Discussion:
[vnfpool] new VNFPool draft charter
Zongning
2014-06-03 09:35:51 UTC
Permalink
Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a "VNF Pool". Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

* How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

* What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

* What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

* When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

* Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

* The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

* The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

* Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

* Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
Diego R. Lopez
2014-06-03 22:23:09 UTC
Permalink
Hi,

It looks really good to me, Ning!

I am just wondering whether we could give state synchronization a special category (of being "out of scope in this phase" or similar) because I think it is a extremely relevant issue.

Be goode,

On 3 Jun 2014, at 11:35 , Zongning <***@huawei.com<mailto:***@huawei.com>> wrote:

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:
1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.
2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.
3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.
4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:
• How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?
• What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?
• What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?
• When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:
• Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.
• The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.
• The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.
• Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.
• Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
_______________________________________________
vnfpool mailing list
***@ietf.org<mailto:***@ietf.org>
https://www.ietf.org/mailman/listinfo/vnfpool


--
"Esta vez no fallaremos, Doctor Infierno"

Dr Diego R. Lopez
Telefonica I+D
http://people.tid.es/diego.lopez/

e-mail: ***@tid.es
Tel: +34 913 129 041
Mobile: +34 682 051 091
-----------------------------------------


________________________________

Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at:
http://www.tid.es/ES/PAGINAS/disclaimer.aspx
Melinda Shore
2014-06-03 22:35:54 UTC
Permalink
Post by Diego R. Lopez
Hi,
It looks really good to me, Ning!
I am just wondering whether we could give state synchronization a
special category (of being "out of scope in this phase" or similar)
because I think it is a extremely relevant issue.
This is an extremely contested question. It seems appropriate to me
to view vnfpool as a layered mechanism, where a layer handling the
reliability/redundancy mechanism may carry other traffic, but that
traffic is opaque to the pooling mechanism. That traffic might
include service state. But, this is my second go-'round on state
transfer (third, if you include sami, the first one being rserpool
itself) and I really think it's best to be left out of scope with
no specific future plans for the time being, but to allow a generic
transport mechanism in the pooling traffic.

Melinda
--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."
Susan Hares
2014-06-05 20:44:25 UTC
Permalink
Melinda:

If we leave it scope for this phase, but open for some future phase I think
it will help the discussion. This will allows to build toward this phase.
Of course, I'm coming from the router/SDN platform view point.

Sue

-----Original Message-----
From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Melinda Shore
Sent: Tuesday, June 03, 2014 6:36 PM
To: ***@ietf.org
Subject: Re: [vnfpool] new VNFPool draft charter
Post by Diego R. Lopez
Hi,
It looks really good to me, Ning!
I am just wondering whether we could give state synchronization a
special category (of being "out of scope in this phase" or similar)
because I think it is a extremely relevant issue.
This is an extremely contested question. It seems appropriate to me to view
vnfpool as a layered mechanism, where a layer handling the
reliability/redundancy mechanism may carry other traffic, but that traffic
is opaque to the pooling mechanism. That traffic might include service
state. But, this is my second go-'round on state transfer (third, if you
include sami, the first one being rserpool
itself) and I really think it's best to be left out of scope with no
specific future plans for the time being, but to allow a generic transport
mechanism in the pooling traffic.

Melinda


--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."
Melinda Shore
2014-06-05 21:02:58 UTC
Permalink
Post by Susan Hares
If we leave it scope for this phase, but open for some future phase I think
it will help the discussion. This will allows to build toward this phase.
Of course, I'm coming from the router/SDN platform view point.
I think it will help among people who'd like to use it but
hurt us politically, to be frank - this is an issue that always
raises a lot of ire in the quarter of the IETF in which we're
now operating. We'd also be faced with the difficult of answering
the question of why a vendor would want a customer to be able to fail,
say, their firewall over to another vendor's firewall. The obvious
answer to that is that customers want it, but that doesn't mean
that vendors will implement that kind of mechanism, and when we've
done polls about this kind of thing we haven't been able to get
any vendor to say they'd do it. So, if people really want to
include text saying that this is future work, by all means do so
but do so with knowledge of how discussions have gone in the very
recent past.

And to be clear, to say that it's out of scope means that it's out of
scope *now* - as we know all decisions may be revisited in the
future.

Melinda
--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."
Susan Hares
2014-06-05 21:18:59 UTC
Permalink
Melinda:

If it is a layer 9 issue (politics) to get the work started, then it out of
scope. I really just want to get going on this work.

Sue

-----Original Message-----
From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Melinda Shore
Sent: Thursday, June 05, 2014 5:03 PM
To: ***@ietf.org
Subject: Re: [vnfpool] new VNFPool draft charter
Post by Susan Hares
If we leave it scope for this phase, but open for some future phase I
think it will help the discussion. This will allows to build toward this
phase.
Post by Susan Hares
Of course, I'm coming from the router/SDN platform view point.
I think it will help among people who'd like to use it but hurt us
politically, to be frank - this is an issue that always raises a lot of ire
in the quarter of the IETF in which we're now operating. We'd also be faced
with the difficult of answering the question of why a vendor would want a
customer to be able to fail, say, their firewall over to another vendor's
firewall. The obvious answer to that is that customers want it, but that
doesn't mean that vendors will implement that kind of mechanism, and when
we've done polls about this kind of thing we haven't been able to get any
vendor to say they'd do it. So, if people really want to include text
saying that this is future work, by all means do so but do so with knowledge
of how discussions have gone in the very recent past.

And to be clear, to say that it's out of scope means that it's out of scope
*now* - as we know all decisions may be revisited in the future.

Melinda

--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."
Zongning
2014-06-06 02:04:19 UTC
Permalink
Hi, Sue, Melinda, and Diego,

I don't see any controversy in your views. As said by Melinda, it's out of scope means that it's out of scope *now*, or *in this phase*.

So I'd propose:
service state synchronization is out of scope - > service state synchronization is out of scope in this phase.

Make sense?

Thanks,

-Ning

-----邮件原件-----
发件人: vnfpool [mailto:vnfpool-***@ietf.org] 代表 Susan Hares
发送时间: 2014年6月6日 5:19
收件人: 'Melinda Shore'; ***@ietf.org
主题: Re: [vnfpool] new VNFPool draft charter

Melinda:

If it is a layer 9 issue (politics) to get the work started, then it out of scope. I really just want to get going on this work.

Sue

-----Original Message-----
From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Melinda Shore
Sent: Thursday, June 05, 2014 5:03 PM
To: ***@ietf.org
Subject: Re: [vnfpool] new VNFPool draft charter
Post by Susan Hares
If we leave it scope for this phase, but open for some future phase I
think it will help the discussion. This will allows to build toward this
phase.
Post by Susan Hares
Of course, I'm coming from the router/SDN platform view point.
I think it will help among people who'd like to use it but hurt us politically, to be frank - this is an issue that always raises a lot of ire in the quarter of the IETF in which we're now operating. We'd also be faced with the difficult of answering the question of why a vendor would want a customer to be able to fail, say, their firewall over to another vendor's firewall. The obvious answer to that is that customers want it, but that doesn't mean that vendors will implement that kind of mechanism, and when we've done polls about this kind of thing we haven't been able to get any vendor to say they'd do it. So, if people really want to include text saying that this is future work, by all means do so but do so with knowledge of how discussions have gone in the very recent past.

And to be clear, to say that it's out of scope means that it's out of scope
*now* - as we know all decisions may be revisited in the future.

Melinda
--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."

_______________________________________________
vnfpool mailing list
***@ietf.org
https://www.ietf.org/mailman/listinfo/vnfpool

_______________________________________________
vnfpool mailing list
***@ietf.org
https://www.i
Ersue, Mehmet (NSN - DE/Munich)
2014-06-06 18:38:16 UTC
Permalink
Hi Ning,

the charter looks good to me and is valuable. I am very much in favor of this work.

Concerning "a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”."

I think a VNF exists as a template plus a SW package, which can be instantiated to create VNF instances. It is probably not the best wording if we say: "A VNF uses a number of VNF instances providing the same function."

Would the text below work for you?
"a VNF may adopt a pooling mechanism, where a "VNF Pool" is used comprising a number of VNF instances with the same function."

s/normal/regular/

Cheers,
Mehmet
Post by Susan Hares
-----Original Message-----
Zongning
Sent: Friday, June 06, 2014 4:04 AM
Subject: [vnfpool] 答复: new VNFPool draft charter
Hi, Sue, Melinda, and Diego,
I don't see any controversy in your views. As said by Melinda, it's out
of scope means that it's out of scope *now*, or *in this phase*.
service state synchronization is out of scope - > service state
synchronization is out of scope in this phase.
Make sense?
Thanks,
-Ning
-----邮件原件-----
发送时间: 2014年6月6日 5:19
主题: Re: [vnfpool] new VNFPool draft charter
If it is a layer 9 issue (politics) to get the work started, then it
out of scope. I really just want to get going on this work.
Sue
-----Original Message-----
Sent: Thursday, June 05, 2014 5:03 PM
Subject: Re: [vnfpool] new VNFPool draft charter
Post by Susan Hares
If we leave it scope for this phase, but open for some future phase
I
Post by Susan Hares
think it will help the discussion. This will allows to build toward this
phase.
Post by Susan Hares
Of course, I'm coming from the router/SDN platform view point.
I think it will help among people who'd like to use it but hurt us
politically, to be frank - this is an issue that always raises a lot of
ire in the quarter of the IETF in which we're now operating. We'd also
be faced with the difficult of answering the question of why a vendor
would want a customer to be able to fail, say, their firewall over to
another vendor's firewall. The obvious answer to that is that
customers want it, but that doesn't mean that vendors will implement
that kind of mechanism, and when we've done polls about this kind of
thing we haven't been able to get any vendor to say they'd do it. So,
if people really want to include text saying that this is future work,
by all means do so but do so with knowledge of how discussions have
gone in the very recent past.
And to be clear, to say that it's out of scope means that it's out of scope
*now* - as we know all decisions may be revisited in the future.
Melinda
--
Melinda Shore
No Mountain Software
"Software longa, hardware brevis."
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
https://www
Zongning
2014-06-09 03:02:17 UTC
Permalink
Hi, Mehmet,

Thanks for your support and suggestions!

Yes, I agree that we can do better wording (combing what you and Linda have suggested) along the lines of -
"a VNF may adopt a pooling mechanism, where a number of VNF instances with the same function can be grouped as a pool to provide the function. We call such a pool a VNF Pool.".

-Ning

-----邮件原件-----
发件人: Ersue, Mehmet (NSN - DE/Munich) [mailto:***@nsn.com]
发送时间: 2014年6月7日 2:38
收件人: Zongning; ***@ietf.org
主题: RE: [vnfpool] 答复: new VNFPool draft charter

Hi Ning,

the charter looks good to me and is valuable. I am very much in favor of this work.

Concerning "a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”."

I think a VNF exists as a template plus a SW package, which can be instantiated to create VNF instances. It is probably not the best wording if we say: "A VNF uses a number of VNF instances providing the same function."

Would the text below work for you?
"a VNF may adopt a pooling mechanism, where a "VNF Pool" is used comprising a number of VNF instances with the same function."

s/normal/regular/

Cheers,
Mehmet
Post by Susan Hares
-----Original Message-----
Zongning
Sent: Friday, June 06, 2014 4:04 AM
Subject: [vnfpool] 答复: new VNFPool draft charter
Hi, Sue, Melinda, and Diego,
I don't see any controversy in your views. As said by Melinda, it's out
of scope means that it's out of scope *now*, or *in this phase*.
service state synchronization is out of scope - > service state
synchronization is out of scope in this phase.
Make sense?
Thanks,
-Ning
-----邮件原件-----
发送时间: 2014年6月6日 5:19
主题: Re: [vnfpool] new VNFPool draft charter
If it is a layer 9 issue (politics) to get the work started, then it
out of scope. I really just want to get going on this work.
Sue
-----Original Message-----
Sent: Thursday, June 05, 2014 5:03 PM
Subject: Re: [vnfpool] new VNFPool draft charter
Post by Susan Hares
If we leave it scope for this phase, but open for some future phase
I
Post by Susan Hares
think it will help the discussion. This will allows to build toward this
phase.
Post by Susan Hares
Of course, I'm coming from the router/SDN platform view point.
I think it will help among people who'd like to use it but hurt us
politically, to be frank - this is an issue that always raises a lot of
ire in the quarter of the IETF in which we're now operating. We'd also
be faced with the difficult of answering the question of why a vendor
would want a customer to be able to fail, say, their firewall over to
another vendor's firewall. The obvious answer to that is that
customers want it, but that doesn't mean that vendors will implement
that kind of mechanism, and when we've done polls about this kind of
thing we haven't been able to get any vendor to say they'd do it. So,
if people really want to include text saying that this is future work,
by all means do so but do so with knowledge of how discussions have
gone in the very recent past.
And to be clear, to say that it's out of scope means that it's out of scope
*now* - as we know all decisions may be revisited in the future.
Melinda
--
Melinda Shore
No Mountain Software
"Software longa, hardware brevis."
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
https://www
Ersue, Mehmet (NSN - DE/Munich)
2014-06-09 09:49:00 UTC
Permalink
WFM

Cheers,
Mehmet
Post by Susan Hares
-----Original Message-----
Sent: Monday, June 09, 2014 5:02 AM
Subject: 答复: [vnfpool] 答复: new VNFPool draft charter
Hi, Mehmet,
Thanks for your support and suggestions!
Yes, I agree that we can do better wording (combing what you and Linda
have suggested) along the lines of -
"a VNF may adopt a pooling mechanism, where a number of VNF instances
with the same function can be grouped as a pool to provide the
function. We call such a pool a VNF Pool.".
-Ning
-----邮件原件-----
发送时间: 2014年6月7日 2:38
主题: RE: [vnfpool] 答复: new VNFPool draft charter
Hi Ning,
the charter looks good to me and is valuable. I am very much in favor of this work.
Concerning "a VNF may adopt a pooling mechanism by using a number of
VNF instances providing the same function, which we call a “VNF Pool”."
I think a VNF exists as a template plus a SW package, which can be
instantiated to create VNF instances. It is probably not the best
wording if we say: "A VNF uses a number of VNF instances providing the
same function."
Would the text below work for you?
"a VNF may adopt a pooling mechanism, where a "VNF Pool" is used
comprising a number of VNF instances with the same function."
s/normal/regular/
Cheers,
Mehmet
Post by Susan Hares
-----Original Message-----
Zongning
Sent: Friday, June 06, 2014 4:04 AM
Subject: [vnfpool] 答复: new VNFPool draft charter
Hi, Sue, Melinda, and Diego,
I don't see any controversy in your views. As said by Melinda, it's
out
Post by Susan Hares
of scope means that it's out of scope *now*, or *in this phase*.
service state synchronization is out of scope - > service state
synchronization is out of scope in this phase.
Make sense?
Thanks,
-Ning
-----邮件原件-----
发送时间: 2014年6月6日 5:19
主题: Re: [vnfpool] new VNFPool draft charter
If it is a layer 9 issue (politics) to get the work started, then it
out of scope. I really just want to get going on this work.
Sue
-----Original Message-----
Sent: Thursday, June 05, 2014 5:03 PM
Subject: Re: [vnfpool] new VNFPool draft charter
Post by Susan Hares
If we leave it scope for this phase, but open for some future
phase
Post by Susan Hares
I
Post by Susan Hares
think it will help the discussion. This will allows to build
toward
Post by Susan Hares
Post by Susan Hares
this
phase.
Post by Susan Hares
Of course, I'm coming from the router/SDN platform view point.
I think it will help among people who'd like to use it but hurt us
politically, to be frank - this is an issue that always raises a lot
of
Post by Susan Hares
ire in the quarter of the IETF in which we're now operating. We'd
also
Post by Susan Hares
be faced with the difficult of answering the question of why a vendor
would want a customer to be able to fail, say, their firewall over to
another vendor's firewall. The obvious answer to that is that
customers want it, but that doesn't mean that vendors will implement
that kind of mechanism, and when we've done polls about this kind of
thing we haven't been able to get any vendor to say they'd do it.
So,
Post by Susan Hares
if people really want to include text saying that this is future
work,
Post by Susan Hares
by all means do so but do so with knowledge of how discussions have
gone in the very recent past.
And to be clear, to say that it's out of scope means that it's out of scope
*now* - as we know all decisions may be revisited in the future.
Melinda
--
Melinda Shore
No Mountain Software
"Software longa, hardware brevis."
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfp
Susan Hares
2014-06-05 20:41:20 UTC
Permalink
Ning:



This charter is really good. The focus to a Pool for single VNF of a single
type is a great place to start. Once we get a working situation we can go
out to important issues like state synchronization. Like Diego, I agree we
should place the state synchronization “out of scope for this phase”.



Sue Hares



From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Diego R. Lopez
Sent: Tuesday, June 03, 2014 6:23 PM
To: Zongning
Cc: ***@ietf.org
Subject: Re: [vnfpool] new VNFPool draft charter



Hi,



It looks really good to me, Ning!



I am just wondering whether we could give state synchronization a special
category (of being "out of scope in this phase" or similar) because I think
it is a extremely relevant issue.



Be goode,



On 3 Jun 2014, at 11:35 , Zongning <***@huawei.com> wrote:





Hi, folks,



Please see the below new draft charter of VNFPool. We believe that we have
updated the charter to address most of the comments from London BoF. Main
changes are:

1) We narrow down our scope to only focus on the redundancy model and
the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is
out-of-scope. We may consider how to maintain the pool states in a VNF Pool
only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same
functional type. Different types of VNFs are envisoned to be held in
separate VNF Pools.

4) We do not address reliability related control or routing between
adjacent VNFs in the service graph. This is also applied to update the
relation between VNF Pool and SFC WG.



==========================================================================

Network functions such as firewalls, load balancers, and WAN optimizers are
conventionally deployed as specialized hardware servers in both network
operators' networks and data center networks as the building blocks of the
network services. There is a trend to implement such network functions as
software instances running on general purpose servers, via a virtualization
layer (i.e., hypervisors). These virtualized functions are called
Virtualized Network Functions (VNFs), which can be used to build network
services.



The use of VNFs introduces additional challenges to the reliability of the
provided network services. A single VNF instance would typically not have
built-in reliability mechanisms on its host (i.e., a general purpose
server). Instead, there are more factors of risk such as software failure at
various levels including hypervisors and virtual machines, hardware failure,
and instance migration that can make a VNF instance unreliable.



In order to provide a reliable function, a VNF may adopt a pooling mechanism
by using a number of VNF instances providing the same function, which we
call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF
Pool, e.g., selects active/standby VNF instances, and potentially interacts
with a service control entity. Such a service control entity is an entity
that orchestrates a set of network functions to build network services. The
major benefit of using VNF Pool is that reliability mechanisms such as
redundancy model are achieved by the VNF Pool inside the VNF and thus not
visible to the service control entity. A VNF Pool-enabled VNF still appears
as a normal VNF when orchestrated by a service control entity.



Questions that are raised by the addition of a pooling mechanism to VNF
include:

· How to manage the redundancy model, e.g., select active/standby
VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling
mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service
control entity? For example, how can a VNF Pool be addressed by the service
control entity?

· When a live VNF instance goes out of service, how does the
service control entity learn which instance in a VNF Pool will replace it,
and learn the characteristics of the new instance?

The WG initially focuses on several reliability mechanisms that are mainly
associated with a redundancy model based on a VNF Pool in a VNF. Additional
mechanisms that may be needed include state maintenance in a VNF Pool only
for pooling purpose (service state synchronization is out of scope). The WG
assumes that a VNF Pool contains redundant VNF instances of same functional
type. Different types of VNFs are envisoned to be held in separate VNF
Pools. The WG will address the reliability of an individual VNF, but not the
reliability related to the control or the routing between adjacent VNFs that
can form a network service.

Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling
between the Pool Manager and the instances in the pool for instance
registration, backup instances selection, and switching between active and
standby instances.

· The protocol between the Pool Manager and the underlying network
to collect the network information required for appropriate
placement/selection of backup instances.

· The protocol between a VNF and the service control entity to
exchange operational information between a VNF Pool and the service control
entity.

· Identify and analyze reliable transport protocols, such as MPTCP
and SCTP for reliable delivery of the messages associated with the
redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to
identify and mitigate threats against the pooling mechanism. Identification
of an appropriate trust model between pool members, and between pool members
and Pool Manager.

The WG plans to deliver a problem statement, a set of use cases, VNF Pool
requirements and an architecture covering the aforementioned technical
aspects, and applicability and gap analysis of existing technologies such as
RSerPool. It is our expectation that we will be able to rely heavily on
existing IETF technologies, but that gaps will be found around problems like
redundancy mechanisms, state maintenance solely for pooling purposes,
reliable transport, and trust/security, all of which will need to be
considered and addressed. The WG will include consideration of the
manageability of a VNF Pool. The WG will seek re-chartering before adopting
any work to develop new, or extend existing, protocols.



In particular, we will work closely with the SFC WG, as we believe that SFC
and VNF Pool are independent but complementary. SFC would essentially see a
VNF Pool-enabled VNF as a normal service function and therefore be able to
merge it into an SFC just like any other service functions. Information
exchanged between the VNF Pool and the SFC may include some operational
information from the VNF Pool including the pool address, pool instance
characteristic, and so on, as requested by the SFC WG.



Goals and Milestones:

December 2014 - Submit VNFPool Problem Statement to IESG for publication as
Informational

April 2015 - Submit VNFPool Use Cases to IESG for publication as
Informational

August 2015 - Submit VNFPool Requirements, including the manageability of
VNF Pool to IESG for publication as Informational

August 2015 - Submit VNFPool Architecture to IESG for publication as
Proposed Standard

December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG
for publication as Informational

==========================================================================



Your comments and suggestions to the charter text are highly appreciated!



Thanks.



-Ning

_______________________________________________
vnfpool mailing list
<mailto:***@ietf.org> ***@ietf.org
<https://www.ietf.org/mailman/listinfo/vnfpool>
https://www.ietf.org/mailman/listinfo/vnfpool




--
"Esta vez no fallaremos, Doctor Infierno"

Dr Diego R. Lopez
Telefonica I+D

http://people.tid.es/diego.lopez/

e-mail: ***@tid.es
Tel: +34 913 129 041
Mobile: +34 682 051 091
-----------------------------------------





_____


Este mensaje se dirige exclusivamente a su destinatario. Puede consultar
nuestra política de envío y recepción de correo electrónico en el enlace
situado más abajo.
This message is intended exclusively for its addressee. We only send and
receive email on the basis of the terms set out at:
http://www.tid.es/ES/PAGINAS/disclaimer.aspx
k***@cs.utwente.nl
2014-06-04 18:26:30 UTC
Permalink
Hi Ning,



The charter looks good to me!



In order to solve the state synchronisation issue, you may want to modify the sentence from:



Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope).



INTO:



Service state synchronization is out of scope. However, state maintenance mechanisms required in a VNF pool for pooling purposes are also in scope.



Diego, Melinda, does the above change address your concerns.



Best regards,

Georgios


________________________________
Van: vnfpool [vnfpool-***@ietf.org] namens Zongning [***@huawei.com]
Verzonden: dinsdag 3 juni 2014 11:35
Aan: ***@ietf.org
Onderwerp: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

• How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

• What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

• What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

• When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

• Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

• The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

• The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

• Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

• Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
Andrew G. Malis
2014-06-04 19:32:26 UTC
Permalink
Georgios,

I support the proposed charter in general, and this proposed change in
particular. This is important work and is a very necessary part of a VNF
implementation. Working on it in the IETF allows us to take advantage of
related experience and expertise and have output that's useful to the work
in the SFC WG.

Cheers,
Andy
Post by k***@cs.utwente.nl
Hi Ning,
The charter looks good to me!
In order to solve the state synchronisation issue, you may want to modify
Additional mechanisms that may be needed include state maintenance in a
VNF Pool only for pooling purpose (service state synchronization is out of
scope).
Service state synchronization is out of scope. However, state maintenance
mechanisms required in a VNF pool for pooling purposes are also in scope.
Diego, Melinda, does the above change address your concerns.
Best regards,
Georgios
------------------------------
*Verzonden:* dinsdag 3 juni 2014 11:35
*Onderwerp:* [vnfpool] new VNFPool draft charter
Hi, folks,
Please see the below new draft charter of VNFPool. We believe that we have
updated the charter to address most of the comments from London BoF. Main
1) We narrow down our scope to only focus on the redundancy model
and the related interfaces associated with the VNF Pool in a single VNF.
2) Service state synchronization between VNF instances is
out-of-scope. We may consider how to maintain the pool states in a VNF Pool
only for pooling purpose.
3) We assume that a VNF Pool contains redundant VNF instances of
same functional type. Different types of VNFs are envisoned to be held in
separate VNF Pools.
4) We do not address reliability related control or routing between
adjacent VNFs in the service graph. This is also applied to update the
relation between VNF Pool and SFC WG.
==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers
are conventionally deployed as specialized hardware servers in both network
operators' networks and data center networks as the building blocks of the
network services. There is a trend to implement such network functions as
software instances running on general purpose servers, via a virtualization
layer (i.e., hypervisors). These virtualized functions are called
Virtualized Network Functions (VNFs), which can be used to build network
services.
The use of VNFs introduces additional challenges to the reliability of the
provided network services. A single VNF instance would typically not have
built-in reliability mechanisms on its host (i.e., a general purpose
server). Instead, there are more factors of risk such as software failure
at various levels including hypervisors and virtual machines, hardware
failure, and instance migration that can make a VNF instance unreliable.
In order to provide a reliable function, a VNF may adopt a pooling
mechanism by using a number of VNF instances providing the same function,
which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to
manage the VNF Pool, e.g., selects active/standby VNF instances, and
potentially interacts with a service control entity. Such a service
control entity is an entity that orchestrates a set of network functions to
build network services. The major benefit of using VNF Pool is that
reliability mechanisms such as redundancy model are achieved by the VNF
Pool inside the VNF and thus not visible to the service control entity. A
VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service
control entity.
· How to manage the redundancy model, e.g., select
active/standby VNF instances in a VNF Pool?
· What pool states need to be maintained to support the pooling
mechanism itself, and how are such states maintained?
· What information is exchanged between a VNF and a service
control entity? For example, how can a VNF Pool be addressed by the service
control entity?
· When a live VNF instance goes out of service, how does the
service control entity learn which instance in a VNF Pool will replace it,
and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly
associated with a redundancy model based on a VNF Pool in a VNF. Additional
mechanisms that may be needed include state maintenance in a VNF Pool
only for pooling purpose (service state synchronization is out of scope).
The WG assumes that a VNF Pool contains redundant VNF instances of same
functional type. Different types of VNFs are envisoned to be held in
separate VNF Pools. The WG will address the reliability of an individual
VNF, but not the reliability related to the control or the routing
between adjacent VNFs that can form a network service.
· Redundancy management within a VNF Pool, such as the
signaling between the Pool Manager and the instances in the pool for
instance registration, backup instances selection, and switching between
active and standby instances.
· The protocol between the Pool Manager and the underlying
network to collect the network information required for appropriate
placement/selection of backup instances.
· The protocol between a VNF and the service control entity to
exchange operational information between a VNF Pool and the service
control entity.
· Identify and analyze reliable transport protocols, such as
MPTCP and SCTP for reliable delivery of the messages associated with the
redundancy management within a VNF Pool.
· Analysis of pooling security characteristics and requirements
to identify and mitigate threats against the pooling mechanism.
Identification of an appropriate trust model between pool members, and
between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool
requirements and an architecture covering the aforementioned technical
aspects, and applicability and gap analysis of existing technologies such
as RSerPool. It is our expectation that we will be able to rely heavily on
existing IETF technologies, but that gaps will be found around problems
like redundancy mechanisms, state maintenance solely for pooling purposes,
reliable transport, and trust/security, all of which will need to be
considered and addressed. The WG will include consideration of the
manageability of a VNF Pool. The WG will seek re-chartering before adopting
any work to develop new, or extend existing, protocols.
In particular, we will work closely with the SFC WG, as we believe that
SFC and VNF Pool are independent but complementary. SFC would essentially
see a VNF Pool-enabled VNF as a normal service function and therefore be
able to merge it into an SFC just like any other service functions.
Information exchanged between the VNF Pool and the SFC may include some
operational information from the VNF Pool including the pool address, pool
instance characteristic, and so on, as requested by the SFC WG.
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of
VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG
for publication as Informational
==========================================================================
Your comments and suggestions to the charter text are highly appreciated!
Thanks.
-Ning
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
Diego R. Lopez
2014-06-04 23:08:46 UTC
Permalink
+1

On 4 Jun 2014, at 21:32 , Andrew G. Malis <***@gmail.com<mailto:***@gmail.com>> wrote:

Georgios,

I support the proposed charter in general, and this proposed change in particular. This is important work and is a very necessary part of a VNF implementation. Working on it in the IETF allows us to take advantage of related experience and expertise and have output that's useful to the work in the SFC WG.

Cheers,
Andy


On Wed, Jun 4, 2014 at 2:26 PM, <***@cs.utwente.nl<mailto:***@cs.utwente.nl>> wrote:

Hi Ning,



The charter looks good to me!



In order to solve the state synchronisation issue, you may want to modify the sentence from:



Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope).



INTO:



Service state synchronization is out of scope. However, state maintenance mechanisms required in a VNF pool for pooling purposes are also in scope.



Diego, Melinda, does the above change address your concerns.



Best regards,

Georgios





________________________________
Van: vnfpool [vnfpool-***@ietf.org<mailto:vnfpool-***@ietf.org>] namens Zongning [***@huawei.com<mailto:***@huawei.com>]
Verzonden: dinsdag 3 juni 2014 11:35
Aan: ***@ietf.org<mailto:***@ietf.org>
Onderwerp: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.


==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

• How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

• What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

• What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

• When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

• Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

• The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

• The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

• Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

• Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning

_______________________________________________
vnfpool mailing list
***@ietf.org<mailto:***@ietf.org>
https://www.ietf.org/mailman/listinfo/vnfpool




--
"Esta vez no fallaremos, Doctor Infierno"

Dr Diego R. Lopez
Telefonica I+D
http://people.tid.es/diego.lopez/

e-mail: ***@tid.es
Tel: +34 913 129 041
Mobile: +34 682 051 091
-----------------------------------------


________________________________

Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at:
http://www.tid.es/ES/PAGINAS/disclaimer.aspx
Linda Dunbar
2014-06-04 20:12:58 UTC
Permalink
Ning,

I made some wording changes to the proposed charter. The attached Word document has the changes mark, making it easier to see. In case the mailing list doesn't allow attachment, here is the proposed changes.

Should we also add a paragraph to show why/how VNFpool is different from SFC?

Linda


Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be instantiated to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services. Conceptionally, multiple instances of a same function are grouped together and managed by its Pool Manager. Different functions have different pool manager. A Pool Manager is responsible for monitoring the status of the instances in the pool, selecting active/standby instances for specific services, and potentially interacts with other entities, e.g. service control entity or service orchestration system. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy & protection are achieved by the VNF Pool and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.[L1]

Questions that are raised by the addition of a pooling mechanism to VNF include:

* How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

* What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

* What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

* When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms might include state maintenance of instances in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains identical VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.





From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 4:36 AM
To: ***@ietf.org
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a "VNF Pool". Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

* How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

* What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

* What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

* When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

* Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

* The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

* The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

* Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

* Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
________________________________

[L1]Not clear of the intent.



Do you mean: "VNF instances in the VNF pool are still visible by the service control entity"?
Zongning
2014-06-05 01:11:33 UTC
Permalink
Hi, Linda,

Thanks for re-wording. I will incorporate some of the suggested changes in the final charter. Specifically, the below sentence means that a VNF appears to outside entity such as service control entity a normal, single VNF (instance), although internally there is a pool for high reliability.
¡°A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.¡±

We do have a dedicated paragraph about relation between VNFPool and SFC šC it is the last paragraph of the main text, just above the milestone part. In that paragraph, we don¡¯t emphasize much on how VNFPool and SFC are different (as they are different in nature), but describe how they relate and work with each other.

-Ning

·¢ŒþÈË: Linda Dunbar
·¢ËÍʱŒä: 2014Äê6ÔÂ5ÈÕ 4:13
ÊÕŒþÈË: Zongning; ***@ietf.org
Ö÷Ìâ: RE: new VNFPool draft charter

Ning,

I made some wording changes to the proposed charter. The attached Word document has the changes mark, making it easier to see. In case the mailing list doesn¡¯t allow attachment, here is the proposed changes.

Should we also add a paragraph to show why/how VNFpool is different from SFC?

Linda


Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be instantiated to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services. Conceptionally, multiple instances of a same function are grouped together and managed by its Pool Manager. Different functions have different pool manager. A Pool Manager is responsible for monitoring the status of the instances in the pool, selecting active/standby instances for specific services, and potentially interacts with other entities, e.g. service control entity or service orchestration system. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy & protection are achieved by the VNF Pool and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.[L1]

Questions that are raised by the addition of a pooling mechanism to VNF include:

¡€ How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

¡€ What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

¡€ What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

¡€ When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms might include state maintenance of instances in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains identical VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.





From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 4:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a ¡°VNF Pool¡±. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

¡€ How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

¡€ What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

¡€ What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

¡€ When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

¡€ Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

¡€ The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

¡€ The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

¡€ Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

¡€ Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
________________________________

Not clear of the intent.



Do you mean: ¡°VNF instances in the VNF pool are still visible by the service control entity¡±?
Susan Hares
2014-06-05 20:58:30 UTC
Permalink
Linda and Ning.



My understanding from this charter is that the pool would be transparent to
whatever entity used it.



If it was a pool of firewalls, then the VNF would be a firewall. It would be
hidden to SFC or SDN controller wanting to use the firewall that it was part
of a pool.



This comes from section 4 in the problem statement that states:

"The main benefit of using VNF Pool is the pooling mechanism such as
redundancy management are achieved by the VNF Pool inside the VNF and
transparent to the Service Control Entity. The Service control Entity
simply interacts with the Pool Manager in each VNF to request and
orchestrate the network functions with desired reliability level."



To me, in my past work this was important to make the overall process work
simply. Is this the source of your astute question?



Sue





From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Linda Dunbar
Sent: Wednesday, June 04, 2014 4:13 PM
To: Zongning; ***@ietf.org
Subject: Re: [vnfpool] new VNFPool draft charter



Ning,



I made some wording changes to the proposed charter. The attached Word
document has the changes mark, making it easier to see. In case the mailing
list doesn't allow attachment, here is the proposed changes.



Should we also add a paragraph to show why/how VNFpool is different from
SFC?



Linda





Network functions such as firewalls, load balancers, and WAN optimizers are
conventionally deployed as specialized hardware servers in both network
operators' networks and data center networks as the building blocks of the
network services. There is a trend to implement such network functions as
software instances running on general purpose servers, via a virtualization
layer (i.e., hypervisors). These virtualized functions are called
Virtualized Network Functions (VNFs), which can be instantiated to build
network services.



The use of VNFs introduces additional challenges to the reliability of the
provided network services. A single VNF instance would typically not have
built-in reliability mechanisms on its host (i.e., a general purpose
server). Instead, there are more factors of risk such as software failure at
various levels including hypervisors and virtual machines, hardware failure,
and instance migration that can make a VNF instance unreliable.



In order to achieve higher reliability, a pool of VNFs that perform the same
function can be grouped together for the network services. Conceptionally,
multiple instances of a same function are grouped together and managed by
its Pool Manager. Different functions have different pool manager. A Pool
Manager is responsible for monitoring the status of the instances in the
pool, selecting active/standby instances for specific services, and
potentially interacts with other entities, e.g. service control entity or
service orchestration system. The major benefit of using VNF Pool is that
reliability mechanisms such as redundancy & protection are achieved by the
VNF Pool and thus not visible to the service control entity. A VNF
Pool-enabled VNF still appears as a normal VNF when orchestrated by a
service control entity.[L1] <>



Questions that are raised by the addition of a pooling mechanism to VNF
include:

. How to manage the redundancy model, e.g., select active/standby VNF
instances in a VNF Pool?

. What pool states need to be maintained to support the pooling
mechanism itself, and how are such states maintained?

. What information is exchanged between a VNF and a service control
entity? For example, how can a VNF Pool be addressed by the service control
entity?

. When a live VNF instance goes out of service, how does the service
control entity learn which instance in a VNF Pool will replace it, and learn
the characteristics of the new instance?

The WG initially focuses on several reliability mechanisms that are mainly
associated with a redundancy model based on a VNF Pool in a VNF. Additional
mechanisms might include state maintenance of instances in a VNF Pool only
for pooling purpose (service state synchronization is out of scope). The WG
assumes that a VNF Pool contains identical VNF instances of same functional
type. Different types of VNFs are envisoned to be held in separate VNF
Pools. The WG will address the reliability of an individual VNF, but not the
reliability related to the control or the routing between adjacent VNFs that
can form a network service.











From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 4:36 AM
To: ***@ietf.org
Subject: [vnfpool] new VNFPool draft charter



Hi, folks,



Please see the below new draft charter of VNFPool. We believe that we have
updated the charter to address most of the comments from London BoF. Main
changes are:

1) We narrow down our scope to only focus on the redundancy model and
the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is
out-of-scope. We may consider how to maintain the pool states in a VNF Pool
only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same
functional type. Different types of VNFs are envisoned to be held in
separate VNF Pools.

4) We do not address reliability related control or routing between
adjacent VNFs in the service graph. This is also applied to update the
relation between VNF Pool and SFC WG.



==========================================================================

Network functions such as firewalls, load balancers, and WAN optimizers are
conventionally deployed as specialized hardware servers in both network
operators' networks and data center networks as the building blocks of the
network services. There is a trend to implement such network functions as
software instances running on general purpose servers, via a virtualization
layer (i.e., hypervisors). These virtualized functions are called
Virtualized Network Functions (VNFs), which can be used to build network
services.



The use of VNFs introduces additional challenges to the reliability of the
provided network services. A single VNF instance would typically not have
built-in reliability mechanisms on its host (i.e., a general purpose
server). Instead, there are more factors of risk such as software failure at
various levels including hypervisors and virtual machines, hardware failure,
and instance migration that can make a VNF instance unreliable.



In order to provide a reliable function, a VNF may adopt a pooling mechanism
by using a number of VNF instances providing the same function, which we
call a "VNF Pool". Conceptionally, a Pool Manager is used to manage the VNF
Pool, e.g., selects active/standby VNF instances, and potentially interacts
with a service control entity. Such a service control entity is an entity
that orchestrates a set of network functions to build network services. The
major benefit of using VNF Pool is that reliability mechanisms such as
redundancy model are achieved by the VNF Pool inside the VNF and thus not
visible to the service control entity. A VNF Pool-enabled VNF still appears
as a normal VNF when orchestrated by a service control entity.



Questions that are raised by the addition of a pooling mechanism to VNF
include:

. How to manage the redundancy model, e.g., select active/standby VNF
instances in a VNF Pool?

. What pool states need to be maintained to support the pooling
mechanism itself, and how are such states maintained?

. What information is exchanged between a VNF and a service control
entity? For example, how can a VNF Pool be addressed by the service control
entity?

. When a live VNF instance goes out of service, how does the service
control entity learn which instance in a VNF Pool will replace it, and learn
the characteristics of the new instance?

The WG initially focuses on several reliability mechanisms that are mainly
associated with a redundancy model based on a VNF Pool in a VNF. Additional
mechanisms that may be needed include state maintenance in a VNF Pool only
for pooling purpose (service state synchronization is out of scope). The WG
assumes that a VNF Pool contains redundant VNF instances of same functional
type. Different types of VNFs are envisoned to be held in separate VNF
Pools. The WG will address the reliability of an individual VNF, but not the
reliability related to the control or the routing between adjacent VNFs that
can form a network service.

Specifically, the WG will work on the following technical aspects:

. Redundancy management within a VNF Pool, such as the signaling
between the Pool Manager and the instances in the pool for instance
registration, backup instances selection, and switching between active and
standby instances.

. The protocol between the Pool Manager and the underlying network
to collect the network information required for appropriate
placement/selection of backup instances.

. The protocol between a VNF and the service control entity to
exchange operational information between a VNF Pool and the service control
entity.

. Identify and analyze reliable transport protocols, such as MPTCP
and SCTP for reliable delivery of the messages associated with the
redundancy management within a VNF Pool.

. Analysis of pooling security characteristics and requirements to
identify and mitigate threats against the pooling mechanism. Identification
of an appropriate trust model between pool members, and between pool members
and Pool Manager.

The WG plans to deliver a problem statement, a set of use cases, VNF Pool
requirements and an architecture covering the aforementioned technical
aspects, and applicability and gap analysis of existing technologies such as
RSerPool. It is our expectation that we will be able to rely heavily on
existing IETF technologies, but that gaps will be found around problems like
redundancy mechanisms, state maintenance solely for pooling purposes,
reliable transport, and trust/security, all of which will need to be
considered and addressed. The WG will include consideration of the
manageability of a VNF Pool. The WG will seek re-chartering before adopting
any work to develop new, or extend existing, protocols.



In particular, we will work closely with the SFC WG, as we believe that SFC
and VNF Pool are independent but complementary. SFC would essentially see a
VNF Pool-enabled VNF as a normal service function and therefore be able to
merge it into an SFC just like any other service functions. Information
exchanged between the VNF Pool and the SFC may include some operational
information from the VNF Pool including the pool address, pool instance
characteristic, and so on, as requested by the SFC WG.



Goals and Milestones:

December 2014 - Submit VNFPool Problem Statement to IESG for publication as
Informational

April 2015 - Submit VNFPool Use Cases to IESG for publication as
Informational

August 2015 - Submit VNFPool Requirements, including the manageability of
VNF Pool to IESG for publication as Informational

August 2015 - Submit VNFPool Architecture to IESG for publication as
Proposed Standard

December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG
for publication as Informational

==========================================================================



Your comments and suggestions to the charter text are highly appreciated!



Thanks.



-Ning

_____

Not clear of the intent.



Do you mean: "VNF instances in the VNF pool are still visible by the service
control entity"?
Zongning
2014-06-06 01:57:14 UTC
Permalink
Hi, Sue,

Yes, this is exactly what we mean in both PS draft and charter text. The pooling is transparent to outside service control entities/controllers using VNF to implement services.

Thanks.

-Ning

·¢ŒþÈË: Susan Hares [mailto:***@ndzh.com]
·¢ËÍʱŒä: 2014Äê6ÔÂ6ÈÕ 4:59
ÊÕŒþÈË: Linda Dunbar; Zongning; ***@ietf.org
Ö÷Ìâ: RE: [vnfpool] new VNFPool draft charter

Linda and Ning.

My understanding from this charter is that the pool would be transparent to whatever entity used it.

If it was a pool of firewalls, then the VNF would be a firewall. It would be hidden to SFC or SDN controller wanting to use the firewall that it was part of a pool.

This comes from section 4 in the problem statement that states:
¡°The main benefit of using VNF Pool is the pooling mechanism such as redundancy management are achieved by the VNF Pool inside the VNF and transparent to the Service Control Entity. The Service control Entity simply interacts with the Pool Manager in each VNF to request and orchestrate the network functions with desired reliability level.¡±

To me, in my past work this was important to make the overall process work simply. Is this the source of your astute question?

Sue


From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Linda Dunbar
Sent: Wednesday, June 04, 2014 4:13 PM
To: Zongning; ***@ietf.org<mailto:***@ietf.org>
Subject: Re: [vnfpool] new VNFPool draft charter

Ning,

I made some wording changes to the proposed charter. The attached Word document has the changes mark, making it easier to see. In case the mailing list doesn¡¯t allow attachment, here is the proposed changes.

Should we also add a paragraph to show why/how VNFpool is different from SFC?

Linda


Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be instantiated to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services. Conceptionally, multiple instances of a same function are grouped together and managed by its Pool Manager. Different functions have different pool manager. A Pool Manager is responsible for monitoring the status of the instances in the pool, selecting active/standby instances for specific services, and potentially interacts with other entities, e.g. service control entity or service orchestration system. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy & protection are achieved by the VNF Pool and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.[L1]

Questions that are raised by the addition of a pooling mechanism to VNF include:

¡€ How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

¡€ What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

¡€ What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

¡€ When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms might include state maintenance of instances in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains identical VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.





From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 4:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a ¡°VNF Pool¡±. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

¡€ How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

¡€ What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

¡€ What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

¡€ When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

¡€ Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

¡€ The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

¡€ The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

¡€ Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

¡€ Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
________________________________

Not clear of the intent.



Do you mean: ¡°VNF instances in the VNF pool are still visible by the service control entity¡±?
King, Daniel
2014-06-04 21:27:36 UTC
Permalink
Hi All,

Thanks Ning and Melinda for the draft charter text, overall the direction and work items look great to me.

Still, being overly opinionated, I do have a few comments:

Charter Text
- Maybe break existing bullet, into two bullets (and minor grammar fix - "the Pool Manager":

o Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism.
o Identification of an appropriate trust model between pool members, and between pool members and the Pool Manager.

- Also, does Trust imply Policy, or vice versa?

- Adding [secure] as requirement in the transport protocol selection. Or maybe I am predetermining the protocol requirements, but then again we already state "reliable" mechanism:

o Investigate and select a [secure] and reliable transport protocol, such as MPTCP and SCTP, for delivery of the messages associated with the redundancy management within a VNF Pool.

Charter Milestones
- Perhaps specify a security should be investigated along with manageability in the following milestone:

"August 2015 - Submit VNFPool Requirements, including the manageability and security of VNF Pools to the IESG for publication as an Informational document."

- Slight clean up (grammar and adding the word "document") on the charter milestones, these would become (note the inclusion of "security"):

December 2014 - Submit VNFPool Problem Statement to IESG for publication as an Informational document.
April 2015 - Submit VNFPool Use Cases to IESG for publication as an Informational document.
August 2015 - Submit VNFPool Requirements, including the manageability and security of VNF Pools to the IESG for publication as an Informational document.
August 2015 - Submit VNFPool Architecture to IESG for publication as a Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational document.

Spelling NITs in Charter Text
- "Conceptionally" to "Conceptually"
- " envisoned " to "envisioned"

Br, Dan.
Melinda Shore
2014-06-04 21:32:54 UTC
Permalink
Post by King, Daniel
- Also, does Trust imply Policy, or vice versa?
Typically in an IETF context it implies mechanism.
Post by King, Daniel
Charter Milestones - Perhaps specify a security should be
"August 2015 - Submit VNFPool Requirements, including the
manageability and security of VNF Pools to the IESG for publication
as an Informational document."
My general sense is that doing security requirements
separately is generally not successful, and that
including them in this sort of context tends to lead
to overly generic requirements that may not be
applicable to the technology that's ultimately adopted.

Melinda
PEDRO ANDRES ARANDA GUTIERREZ
2014-06-05 07:45:53 UTC
Permalink
Hi,

I support the WG Charter. However, I have a small concern regarding the first sentence in the third paragraph:

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services.

There is no mention to ‘the network services’ before.

Best,/PA

Dr. Pedro A. Aranda Gutiérrez

Technology Exploration -
Network Innovation & Virtualisation

mailto:***@tid.es
Telefónica, Investigación y Desarrollo
C/ D. Ramón de la Cruz,84
28006 Madrid, Spain

Fragen sind nicht da, um beantwortet zu werden.
Fragen sind da, um gestellt zu werden.
Georg Kreisler

De: Zongning <***@huawei.com<mailto:***@huawei.com>>
Fecha: jueves, 5 de junio de 2014 03:11
Para: Linda Dunbar <***@huawei.com<mailto:***@huawei.com>>, "***@ietf.org<mailto:***@ietf.org>" <***@ietf.org<mailto:***@ietf.org>>
Asunto: [vnfpool] 答倍: new VNFPool draft charter

Hi, Linda,

Thanks for re-wording. I will incorporate some of the suggested changes in the final charter. Specifically, the below sentence means that a VNF appears to outside entity such as service control entity a normal, single VNF (instance), although internally there is a pool for high reliability.
“A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.”

We do have a dedicated paragraph about relation between VNFPool and SFC – it is the last paragraph of the main text, just above the milestone part. In that paragraph, we don’t emphasize much on how VNFPool and SFC are different (as they are different in nature), but describe how they relate and work with each other.

-Ning

发件人: Linda Dunbar
发送时闎: 2014幎6月5日 4:13
收件人: Zongning; ***@ietf.org<mailto:***@ietf.org>
䞻题: RE: new VNFPool draft charter

Ning,

I made some wording changes to the proposed charter. The attached Word document has the changes mark, making it easier to see. In case the mailing list doesn’t allow attachment, here is the proposed changes.

Should we also add a paragraph to show why/how VNFpool is different from SFC?

Linda


Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be instantiated to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services. Conceptionally, multiple instances of a same function are grouped together and managed by its Pool Manager. Different functions have different pool manager. A Pool Manager is responsible for monitoring the status of the instances in the pool, selecting active/standby instances for specific services, and potentially interacts with other entities, e.g. service control entity or service orchestration system. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy & protection are achieved by the VNF Pool and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.[L1]

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms might include state maintenance of instances in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains identical VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.





From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 4:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

· The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

· The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

· Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
________________________________

Not clear of the intent.



Do you mean: “VNF instances in the VNF pool are still visible by the service control entity”?

________________________________

Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at:
http://www.tid.es/ES/PAGINAS/disclaimer.aspx
Zongning
2014-06-05 08:15:39 UTC
Permalink
Hi, Pedro,

Thanks for your support. I guess “the network service” here could be “the network function”. Or we could delete the words after “grouped together” to avoid duplication.

-Ning

发件人: PEDRO ANDRES ARANDA GUTIERREZ [mailto:***@tid.es]
发送时闎: 2014幎6月5日 15:46
收件人: Zongning; Linda Dunbar; ***@ietf.org
䞻题: Re: [vnfpool] 答倍: new VNFPool draft charter

Hi,

I support the WG Charter. However, I have a small concern regarding the first sentence in the third paragraph:

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services.

There is no mention to ‘the network services’ before.

Best,/PA

Dr. Pedro A. Aranda Gutiérrez

Technology Exploration -
Network Innovation & Virtualisation

mailto:***@tid.es
Telefónica, Investigación y Desarrollo
C/ D. Ramón de la Cruz,84
28006 Madrid, Spain

Fragen sind nicht da, um beantwortet zu werden.
Fragen sind da, um gestellt zu werden.
Georg Kreisler

De: Zongning <***@huawei.com<mailto:***@huawei.com>>
Fecha: jueves, 5 de junio de 2014 03:11
Para: Linda Dunbar <***@huawei.com<mailto:***@huawei.com>>, "***@ietf.org<mailto:***@ietf.org>" <***@ietf.org<mailto:***@ietf.org>>
Asunto: [vnfpool] 答倍: new VNFPool draft charter

Hi, Linda,

Thanks for re-wording. I will incorporate some of the suggested changes in the final charter. Specifically, the below sentence means that a VNF appears to outside entity such as service control entity a normal, single VNF (instance), although internally there is a pool for high reliability.
“A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.”

We do have a dedicated paragraph about relation between VNFPool and SFC – it is the last paragraph of the main text, just above the milestone part. In that paragraph, we don’t emphasize much on how VNFPool and SFC are different (as they are different in nature), but describe how they relate and work with each other.

-Ning

发件人: Linda Dunbar
发送时闎: 2014幎6月5日 4:13
收件人: Zongning; ***@ietf.org<mailto:***@ietf.org>
䞻题: RE: new VNFPool draft charter

Ning,

I made some wording changes to the proposed charter. The attached Word document has the changes mark, making it easier to see. In case the mailing list doesn’t allow attachment, here is the proposed changes.

Should we also add a paragraph to show why/how VNFpool is different from SFC?

Linda


Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be instantiated to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to achieve higher reliability, a pool of VNFs that perform the same function can be grouped together for the network services. Conceptionally, multiple instances of a same function are grouped together and managed by its Pool Manager. Different functions have different pool manager. A Pool Manager is responsible for monitoring the status of the instances in the pool, selecting active/standby instances for specific services, and potentially interacts with other entities, e.g. service control entity or service orchestration system. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy & protection are achieved by the VNF Pool and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.[L1]

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms might include state maintenance of instances in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains identical VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.





From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 4:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

· The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

· The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

· Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
________________________________

Not clear of the intent.



Do you mean: “VNF instances in the VNF pool are still visible by the service control entity”?

________________________________

Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at:
http://www.tid.es/ES/PAGINAS/disclaimer.aspx
Zu Qiang
2014-06-06 18:40:47 UTC
Permalink
Hello, Zongning
Thanks for updating the charter. It looks better than the first version. However, I have some concerns below:

- It is good to narrow down the scope to redundancy module only. I’m assuming that a transparent redundancy module is the preferable solution of this WG. With transparent solutions, no impacts are expected on the pool user. Therefore, I would avoid any linkage with SFC.

- I also don’t understand the statement “When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?” If transparent solutions are expected, why SCE needs to learn the failover?

- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

- Some statements, such as “Identify and analyze reliable transport protocols”, sounds like we do have a specific solution already. I won’t assume there is always an IP connection between the pool manager and the pool instance. The interface between the two entities can be a TCP based protocol if there is an IP connection, or it can be an API if they both locate in the same server. It is real dependence on the solutions itself. Besides, if HW redundancy is needed that there may be always a “remote” instance. The question is really how the “remote” instance is handled, by whom; what homogeneity is assumed or whether this should be part of the investigation. If it is expected that the VNF instances of a pool come from different vendors then even the state synchronization becomes an issue. If the pools are expected to be built from identical instances then one questions why should the pool manager come from a different vendor. I won’t want to see any solution limitations at this stage of the discussion. And I prefer to keep it open until the solution discussion phase.

- My assumption is that the output from this proposed WG is one or more generic HA solutions. As we are in the early stage of the discussion, I don’t think we shall limit the solution to a specific one. I do have a hard time to understand why the Applicability and Gap Analysis shall be done based on a specific protocol?

- Reading the charter, it is not very clear that what the scope of the VNF is. Is it the service/function, or the entities providing this service/function? Some clarification of this use would be good.

Have a nice day
Zu Qiang

From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 5:36 AM
To: ***@ietf.org
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

· The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

· The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

· Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
N***@telekom.de
2014-06-06 18:59:13 UTC
Permalink
Hi,


- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

From my point of view this should be also in scope. Otherwise you are not able to include non virtualized functions (which is necessary due to the reasons already pointed out by Pedro) in the framework itself. It should be possible to have a pool of virtualized and non-virtualized functions which can be used based on the requirements derived from the services.

Regards

Nic


From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 5:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

· The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

· The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

· Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
Zu Qiang
2014-06-06 19:07:46 UTC
Permalink
Hello, Nic
That’s my understanding as well.

Have a nice weekend
Zu Qiang

From: ***@telekom.de [mailto:***@telekom.de]
Sent: Friday, June 06, 2014 2:59 PM
To: Zu Qiang; ***@huawei.com; ***@ietf.org
Subject: AW: new VNFPool draft charter

Hi,


- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

From my point of view this should be also in scope. Otherwise you are not able to include non virtualized functions (which is necessary due to the reasons already pointed out by Pedro) in the framework itself. It should be possible to have a pool of virtualized and non-virtualized functions which can be used based on the requirements derived from the services.

Regards

Nic


From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 5:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

· The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

· The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

· Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
Zongning
2014-06-09 02:42:22 UTC
Permalink
Hi, Nic,

I agree with you, Qiang, Pedro on this point, and has proposed a sentence to address this in my reply to Qiang’s discussion - “VNF Pool composed by both virtualized and non-virtualized functional instances will be included after further use case and requirements study”.

Please let us know your further suggestions & comments.

Thanks.

-Ning

发件人: ***@telekom.de [mailto:***@telekom.de]
发送时闎: 2014幎6月7日 2:59
收件人: ***@ericsson.com; Zongning; ***@ietf.org
䞻题: AW: new VNFPool draft charter

Hi,


- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

From my point of view this should be also in scope. Otherwise you are not able to include non virtualized functions (which is necessary due to the reasons already pointed out by Pedro) in the framework itself. It should be possible to have a pool of virtualized and non-virtualized functions which can be used based on the requirements derived from the services.

Regards

Nic


From: vnfpool [mailto:vnfpool-***@ietf.org] On Behalf Of Zongning
Sent: Tuesday, June 03, 2014 5:36 AM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [vnfpool] new VNFPool draft charter

Hi, folks,

Please see the below new draft charter of VNFPool. We believe that we have updated the charter to address most of the comments from London BoF. Main changes are:

1) We narrow down our scope to only focus on the redundancy model and the related interfaces associated with the VNF Pool in a single VNF.

2) Service state synchronization between VNF instances is out-of-scope. We may consider how to maintain the pool states in a VNF Pool only for pooling purpose.

3) We assume that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools.

4) We do not address reliability related control or routing between adjacent VNFs in the service graph. This is also applied to update the relation between VNF Pool and SFC WG.

==========================================================================
Network functions such as firewalls, load balancers, and WAN optimizers are conventionally deployed as specialized hardware servers in both network operators' networks and data center networks as the building blocks of the network services. There is a trend to implement such network functions as software instances running on general purpose servers, via a virtualization layer (i.e., hypervisors). These virtualized functions are called Virtualized Network Functions (VNFs), which can be used to build network services.

The use of VNFs introduces additional challenges to the reliability of the provided network services. A single VNF instance would typically not have built-in reliability mechanisms on its host (i.e., a general purpose server). Instead, there are more factors of risk such as software failure at various levels including hypervisors and virtual machines, hardware failure, and instance migration that can make a VNF instance unreliable.

In order to provide a reliable function, a VNF may adopt a pooling mechanism by using a number of VNF instances providing the same function, which we call a “VNF Pool”. Conceptionally, a Pool Manager is used to manage the VNF Pool, e.g., selects active/standby VNF instances, and potentially interacts with a service control entity. Such a service control entity is an entity that orchestrates a set of network functions to build network services. The major benefit of using VNF Pool is that reliability mechanisms such as redundancy model are achieved by the VNF Pool inside the VNF and thus not visible to the service control entity. A VNF Pool-enabled VNF still appears as a normal VNF when orchestrated by a service control entity.

Questions that are raised by the addition of a pooling mechanism to VNF include:

· How to manage the redundancy model, e.g., select active/standby VNF instances in a VNF Pool?

· What pool states need to be maintained to support the pooling mechanism itself, and how are such states maintained?

· What information is exchanged between a VNF and a service control entity? For example, how can a VNF Pool be addressed by the service control entity?

· When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?
The WG initially focuses on several reliability mechanisms that are mainly associated with a redundancy model based on a VNF Pool in a VNF. Additional mechanisms that may be needed include state maintenance in a VNF Pool only for pooling purpose (service state synchronization is out of scope). The WG assumes that a VNF Pool contains redundant VNF instances of same functional type. Different types of VNFs are envisoned to be held in separate VNF Pools. The WG will address the reliability of an individual VNF, but not the reliability related to the control or the routing between adjacent VNFs that can form a network service.
Specifically, the WG will work on the following technical aspects:

· Redundancy management within a VNF Pool, such as the signaling between the Pool Manager and the instances in the pool for instance registration, backup instances selection, and switching between active and standby instances.

· The protocol between the Pool Manager and the underlying network to collect the network information required for appropriate placement/selection of backup instances.

· The protocol between a VNF and the service control entity to exchange operational information between a VNF Pool and the service control entity.

· Identify and analyze reliable transport protocols, such as MPTCP and SCTP for reliable delivery of the messages associated with the redundancy management within a VNF Pool.

· Analysis of pooling security characteristics and requirements to identify and mitigate threats against the pooling mechanism. Identification of an appropriate trust model between pool members, and between pool members and Pool Manager.
The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies such as RSerPool. It is our expectation that we will be able to rely heavily on existing IETF technologies, but that gaps will be found around problems like redundancy mechanisms, state maintenance solely for pooling purposes, reliable transport, and trust/security, all of which will need to be considered and addressed. The WG will include consideration of the manageability of a VNF Pool. The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols.

In particular, we will work closely with the SFC WG, as we believe that SFC and VNF Pool are independent but complementary. SFC would essentially see a VNF Pool-enabled VNF as a normal service function and therefore be able to merge it into an SFC just like any other service functions. Information exchanged between the VNF Pool and the SFC may include some operational information from the VNF Pool including the pool address, pool instance characteristic, and so on, as requested by the SFC WG.

Goals and Milestones:
December 2014 - Submit VNFPool Problem Statement to IESG for publication as Informational
April 2015 - Submit VNFPool Use Cases to IESG for publication as Informational
August 2015 - Submit VNFPool Requirements, including the manageability of VNF Pool to IESG for publication as Informational
August 2015 - Submit VNFPool Architecture to IESG for publication as Proposed Standard
December 2015 - Submit Applicability and Gap Analysis of RSerPool to IESG for publication as Informational
==========================================================================

Your comments and suggestions to the charter text are highly appreciated!

Thanks.

-Ning
Melinda Shore
2014-06-06 19:09:42 UTC
Permalink
Post by Zu Qiang
- It is good to narrow down the scope to redundancy module
only. I’m assuming that a transparent redundancy module is the
preferable solution of this WG. With transparent solutions, no impacts
are expected on the pool user. Therefore, I would avoid any linkage with
SFC.
We tried that! There's some fairly deep concern about
overlap, and about whether or not some of this might encroach
on sfc turf - that was probably the main issue raised at
the BOF at IETF 89. Because of that it's necessary to highlight
the differences between the work and clarify the relationship
between what sfc is doing and what vnfpool would do.
Post by Zu Qiang
- I also don’t understand the statement “When a live VNF
instance goes out of service, how does the service control entity learn
which instance in a VNF Pool will replace it, and learn the
characteristics of the new instance?” If transparent solutions are
expected, why SCE needs to learn the failover?
That's a good point.
Post by Zu Qiang
- My assumption is that the output from this proposed WG is one
or more generic HA solutions. As we are in the early stage of the
discussion, I don’t think we shall limit the solution to a specific one.
I do have a hard time to understand why the Applicability and Gap
Analysis shall be done based on a specific protocol?
IETF work tends to be bottom-up - specific, discrete, concrete
pieces of work are identified and scoped. We don't solve A Big
Problem and assume it's generally applicable, but rather do work
of the nature of "Here's how to use <x> to do <y>." It may in
fact turn out that there's interest in developing an availability
mechanism based on a different architectural model, and that's
okay.
Post by Zu Qiang
- Reading the charter, it is not very clear that what the scope
of the VNF is. Is it the service/function, or the entities providing
this service/function?
I think this is the question that "service state synchronization is
out of scope" answers.

Melinda
--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."
Zu Qiang
2014-06-06 19:20:41 UTC
Permalink
Have a nice day
Zu Qiang
Post by Zu Qiang
- My assumption is that the output from this proposed WG is one
or more generic HA solutions. As we are in the early stage of the
discussion, I don't think we shall limit the solution to a specific one.
I do have a hard time to understand why the Applicability and Gap
Analysis shall be done based on a specific protocol?
IETF work tends to be bottom-up - specific, discrete, concrete pieces of work are
identified and scoped. We don't solve A Big Problem and assume it's generally
applicable, but rather do work of the nature of "Here's how to use <x> to do
<y>." It may in fact turn out that there's interest in developing an availability
mechanism based on a different architectural model, and that's okay.
[Zu Qiang] Just for my clarification. Does that means all the WG discussion must be based on that specific solution from day one?
Melinda
--
Melinda Shore
No Mountain Software
"Software longa, hardware brevis."
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
Melinda Shore
2014-06-06 19:26:56 UTC
Permalink
Post by Zu Qiang
[Zu Qiang] Just for my clarification. Does that means all the WG discussion must be based on that specific solution from day one?
The working group hasn't been chartered yet, but if a group
is formed based on some version of the current charter, the
documents need to stick to that charter. If there's a proposal
for a reliability mechanism based on a different model the
best thing to do is to prepare a draft and get feedback on it,
and create interest. But, if there's no specific proposal
but broad interest in discussing availability/reliability
mechanisms more generally, it may be possible to create a
non-WG mailing list.

Melinda
--
Melinda Shore
No Mountain Software
***@nomountain.net

"Software longa, hardware brevis."
Zongning
2014-06-09 02:37:40 UTC
Permalink
Hi, Qiang,

I don’t believe the WG discussion will be based on any specific solution at this stage. Reading the charter, it says:
"The WG plans to deliver a problem statement, a set of use cases, VNF Pool requirements and an architecture covering the aforementioned technical aspects, and applicability and gap analysis of existing technologies.", and
"The WG will seek re-chartering before adopting any work to develop new, or extend existing, protocols."

Please let us know any statement in the charter that makes you feel that we are doing solution right now. We can do re-wording. :-)

Thanks.

-Ning

-----邮件原件-----
发件人: vnfpool [mailto:vnfpool-***@ietf.org] 代表 Zu Qiang
发送时间: 2014年6月7日 3:21
收件人: Melinda Shore; ***@ietf.org
主题: Re: [vnfpool] new VNFPool draft charter



Have a nice day
Zu Qiang
Post by Zu Qiang
- My assumption is that the output from this proposed WG is one
or more generic HA solutions. As we are in the early stage of the
discussion, I don't think we shall limit the solution to a specific one.
I do have a hard time to understand why the Applicability and Gap
Analysis shall be done based on a specific protocol?
IETF work tends to be bottom-up - specific, discrete, concrete pieces
of work are identified and scoped. We don't solve A Big Problem and
assume it's generally applicable, but rather do work of the nature of
"Here's how to use <x> to do <y>." It may in fact turn out that
there's interest in developing an availability mechanism based on a different architectural model, and that's okay.
[Zu Qiang] Just for my clarification. Does that means all the WG discussion must be based on that specific solution from day one?
Melinda
--
Melinda Shore
No Mountain Software
"Software longa, hardware brevis."
_______________________________________________
vnfpool mailing list
https://www.ietf.org/mailman/listinfo/vnfpool
_______________________________________________
vnfpool mailing list
***@ietf.org
https://www.ietf.org/mailman/listinfo/vnfpool
Zongning
2014-06-09 02:28:14 UTC
Permalink
Hi, Qiang,

Thanks for the comments. Please see inline.


- It is good to narrow down the scope to redundancy module only. I’m assuming that a transparent redundancy module is the preferable solution of this WG. With transparent solutions, no impacts are expected on the pool user. Therefore, I would avoid any linkage with SFC.

[Ning] Yes, the pooling mechanism will be transparent to the pool user. But a VNF including a VNF Pool (we call a VNFPool-enabled VNF) will be used as a normal VNF in a service chain to build a network service. Therefore, from the service orchestration point of view, some features (mostly operational, NOT pooling mechanism) of a VNFPool-enabled VNF such as address/location, load capability, etc., are still needed by the service control entity.


- I also don’t understand the statement “When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?” If transparent solutions are expected, why SCE needs to learn the failover?

[Ning] Excellent point! You are right that this sentence is somehow misleading. How about a new sentence like - “After a VNF instance failover, how does the Pool Manager notify the service control entity some characteristic changes of the VNF, but without disclosure of the pooling procedure?”.


- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

[Ning] I agree with you, Nic and Pedro that the solution should be applicable to non-virtualized function, from the pooling perspective. But, consider that we currently don’t describe any specific solution in the charter, how about add a new sentence like – “VNF Pool composed by both virtualized and non-virtualized functional instances will be included after further use case and requirements study”. With this statement, I’d solicit you all to bring use case and requirement for the pooling of non-virtualized function, to help the future solution developed by the WG to cover both. ☺


- Some statements, such as “Identify and analyze reliable transport protocols”, sounds like we do have a specific solution already. I won’t assume there is always an IP connection between the pool manager and the pool instance. The interface between the two entities can be a TCP based protocol if there is an IP connection, or it can be an API if they both locate in the same server. It is real dependence on the solutions itself. Besides, if HW redundancy is needed that there may be always a “remote” instance. The question is really how the “remote” instance is handled, by whom; what homogeneity is assumed or whether this should be part of the investigation. If it is expected that the VNF instances of a pool come from different vendors then even the state synchronization becomes an issue. If the pools are expected to be built from identical instances then one questions why should the pool manager come from a different vendor. I won’t want to see any solution limitations at this stage of the discussion. And I prefer to keep it open until the solution discussion phase.

[Ning] I think the statement “Identify and analyze reliable transport protocols” is more like a requirement for the reliable delivery of pooling messages. We give examples of existing IETF solution like MPTCP, but this doesn’t mean we have already determined the solution at this stage. You are right that we are open to solutions and solution discussion will be in next phase.


- My assumption is that the output from this proposed WG is one or more generic HA solutions. As we are in the early stage of the discussion, I don’t think we shall limit the solution to a specific one. I do have a hard time to understand why the Applicability and Gap Analysis shall be done based on a specific protocol?

[Ning] I believe you had misunderstood the Applicability and Gap Analysis I-D (to RSerPool) at this point. ☺ I think anyone (other than RSerPool folks) can bring their I-D about applicability and gap analysis to solution “x”. Currently we have an I-D based on RSerPool. But this doesn’t mean that the WG will start solution based on RSerPool. We will be looking at more reference solutions and find out which parts of them can be reused/extended/referred/etc.


- Reading the charter, it is not very clear that what the scope of the VNF is. Is it the service/function, or the entities providing this service/function? Some clarification of this use would be good.

[Ning] In our charter, a VNF includes a pool of VNF instances implementing a certain function (not a whole service), and a manager to interface with the outside service control entity to provide such function. So I assume a VNF has both roles in your question. Does it make sense? Or you have some suggested text to further clarify this issue? ☺


I hope my reply helps to address your concerns. Otherwise please let me know.

-Ning
Zu Qiang
2014-06-10 01:30:51 UTC
Permalink
Hello, Zongning
Thanks for your reply. See inline below.
Have a nice day
Zu Qiang

From: Zongning [mailto:***@huawei.com]
Sent: Sunday, June 08, 2014 10:28 PM
To: Zu Qiang; ***@ietf.org
Subject: ŽðžŽ: new VNFPool draft charter

Hi, Qiang,

Thanks for the comments. Please see inline.


- It is good to narrow down the scope to redundancy module only. I¡¯m assuming that a transparent redundancy module is the preferable solution of this WG. With transparent solutions, no impacts are expected on the pool user. Therefore, I would avoid any linkage with SFC.

[Ning] Yes, the pooling mechanism will be transparent to the pool user. But a VNF including a VNF Pool (we call a VNFPool-enabled VNF) will be used as a normal VNF in a service chain to build a network service. Therefore, from the service orchestration point of view, some features (mostly operational, NOT pooling mechanism) of a VNFPool-enabled VNF such as address/location, load capability, etc., are still needed by the service control entity.
[Zu Qiang] My understanding is that SFC is just one of the VNF pool users. It shall not make any different with another network applications which will use the VNF pool for redundancy. Therefore I don¡¯t see any need for a linkage with SFC in the charter.


- I also don¡¯t understand the statement ¡°When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?¡± If transparent solutions are expected, why SCE needs to learn the failover?

[Ning] Excellent point! You are right that this sentence is somehow misleading. How about a new sentence like - ¡°After a VNF instance failover, how does the Pool Manager notify the service control entity some characteristic changes of the VNF, but without disclosure of the pooling procedure?¡±.
[Zu Qiang] It¡¯s good that you agree with me. The PM-SCE interface is more for the purpose of the pool OAM. The PM shall report to the SCE with the pool status (including both the application and the HW) at any time or periodically, right? Therefore the above statement shall be in the scope of the normal OAM report, isn¡¯t it? Failover is just a normal HA procedure. The SCE doesn¡¯t have to know it. If the failover is due to an application error, the PM may have enough capability to handle it. But if the failover is due to a HW or VM problem where new network resource allocation may be needed, it may be good to inform the SCE. If that is agreeable, I propose some rewording such as ¡°How does the Pool Manager notify the Service Control Entity of some characteristic changes of the VNF, e.g. system failure, but without disclosure of the pooling procedure?¡±


- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

[Ning] I agree with you, Nic and Pedro that the solution should be applicable to non-virtualized function, from the pooling perspective. But, consider that we currently don¡¯t describe any specific solution in the charter, how about add a new sentence like šC ¡°VNF Pool composed by both virtualized and non-virtualized functional instances will be included after further use case and requirements study¡±. With this statement, I¡¯d solicit you all to bring use case and requirement for the pooling of non-virtualized function, to help the future solution developed by the WG to cover both. :)
[Zu Qiang] Great.


- Some statements, such as ¡°Identify and analyze reliable transport protocols¡±, sounds like we do have a specific solution already. I won¡¯t assume there is always an IP connection between the pool manager and the pool instance. The interface between the two entities can be a TCP based protocol if there is an IP connection, or it can be an API if they both locate in the same server. It is real dependence on the solutions itself. Besides, if HW redundancy is needed that there may be always a ¡°remote¡± instance. The question is really how the ¡°remote¡± instance is handled, by whom; what homogeneity is assumed or whether this should be part of the investigation. If it is expected that the VNF instances of a pool come from different vendors then even the state synchronization becomes an issue. If the pools are expected to be built from identical instances then one questions why should the pool manager come from a different vendor. I won¡¯t want to see any solution limitations at this stage of the discussion. And I prefer to keep it open until the solution discussion phase.

[Ning] I think the statement ¡°Identify and analyze reliable transport protocols¡± is more like a requirement for the reliable delivery of pooling messages. We give examples of existing IETF solution like MPTCP, but this doesn¡¯t mean we have already determined the solution at this stage. You are right that we are open to solutions and solution discussion will be in next phase.
[Zu Qiang] If that¡¯s the case, please make it clear in the charter. And I propose to reword with ¡°Identify and analyze reliable interfaces transport protocols¡±


- My assumption is that the output from this proposed WG is one or more generic HA solutions. As we are in the early stage of the discussion, I don¡¯t think we shall limit the solution to a specific one. I do have a hard time to understand why the Applicability and Gap Analysis shall be done based on a specific protocol?

[Ning] I believe you had misunderstood the Applicability and Gap Analysis I-D (to RSerPool) at this point. :) I think anyone (other than RSerPool folks) can bring their I-D about applicability and gap analysis to solution ¡°x¡±. Currently we have an I-D based on RSerPool. But this doesn¡¯t mean that the WG will start solution based on RSerPool. We will be looking at more reference solutions and find out which parts of them can be reused/extended/referred/etc.
[Zu Qiang] If that¡¯s the case, it would be good to make it clear in the charter to avoid any misunderstandings or layer 9 issues.


- Reading the charter, it is not very clear that what the scope of the VNF is. Is it the service/function, or the entities providing this service/function? Some clarification of this use would be good.

[Ning] In our charter, a VNF includes a pool of VNF instances implementing a certain function (not a whole service), and a manager to interface with the outside service control entity to provide such function. So I assume a VNF has both roles in your question. Does it make sense? Or you have some suggested text to further clarify this issue? :)
[Zu Qiang] So the WG will work on the Pool enabled VNF (not the generic VNF), right? :) And a Pool enabled VNF including the VNF function and the network entities (e.g. HW, interface, etc.), right? It is better to add this clarification in the charter.

I hope my reply helps to address your concerns. Otherwise please let me know.

-Ning
Zongning
2014-06-10 02:52:39 UTC
Permalink
Hi, Qiang,

Please see inline.


- It is good to narrow down the scope to redundancy module only. I¡¯m assuming that a transparent redundancy module is the preferable solution of this WG. With transparent solutions, no impacts are expected on the pool user. Therefore, I would avoid any linkage with SFC.

[Ning] Yes, the pooling mechanism will be transparent to the pool user. But a VNF including a VNF Pool (we call a VNFPool-enabled VNF) will be used as a normal VNF in a service chain to build a network service. Therefore, from the service orchestration point of view, some features (mostly operational, NOT pooling mechanism) of a VNFPool-enabled VNF such as address/location, load capability, etc., are still needed by the service control entity.
[Zu Qiang] My understanding is that SFC is just one of the VNF pool users. It shall not make any different with another network applications which will use the VNF pool for redundancy. Therefore I don¡¯t see any need for a linkage with SFC in the charter.

[Ning] The reason we describe the relation between SFC and VNFPool is mainly based on the conclusion of London BoF, where people wanted to see such relation to be clearly addressed in the new charter. I agree with you that SFC is just one type of service control entities/pool users, so we don¡¯t state how SFC differ from other pool users in the charter. If you still feel uncomfortable, I¡¯d replace the last sentence to ¡°Just like the communication between any pool user and VNF Pool, the information exchanged between the VNF Pool and the SFC may include some operational information of the VNF Pool¡±. Make sense?


- I also don¡¯t understand the statement ¡°When a live VNF instance goes out of service, how does the service control entity learn which instance in a VNF Pool will replace it, and learn the characteristics of the new instance?¡± If transparent solutions are expected, why SCE needs to learn the failover?

[Ning] Excellent point! You are right that this sentence is somehow misleading. How about a new sentence like - ¡°After a VNF instance failover, how does the Pool Manager notify the service control entity some characteristic changes of the VNF, but without disclosure of the pooling procedure?¡±.
[Zu Qiang] It¡¯s good that you agree with me. The PM-SCE interface is more for the purpose of the pool OAM. The PM shall report to the SCE with the pool status (including both the application and the HW) at any time or periodically, right? Therefore the above statement shall be in the scope of the normal OAM report, isn¡¯t it? Failover is just a normal HA procedure. The SCE doesn¡¯t have to know it. If the failover is due to an application error, the PM may have enough capability to handle it. But if the failover is due to a HW or VM problem where new network resource allocation may be needed, it may be good to inform the SCE. If that is agreeable, I propose some rewording such as ¡°How does the Pool Manager notify the Service Control Entity of some characteristic changes of the VNF, e.g. system failure, but without disclosure of the pooling procedure?¡±

[Ning] Your understanding is correct. But I don¡¯t think ¡°system failure¡± is clear enough, as ¡°system¡± is a fairly general term. How about ¡°resource re-allocation¡±, or ¡°load/capacity change¡±?


- There was some discussions in the mailing list regarding the virtualized/non-virtualized case where a network application may run on a server with/without a hypervisor. In both cases (with/without a hypervisor), the application may not have any redundancy solutions. Both cases shall be covered by the solutions provided by this WG. However the charter only have virtualized functions covered. Is the non-virtualized use case out of the scope?

[Ning] I agree with you, Nic and Pedro that the solution should be applicable to non-virtualized function, from the pooling perspective. But, consider that we currently don¡¯t describe any specific solution in the charter, how about add a new sentence like šC ¡°VNF Pool composed by both virtualized and non-virtualized functional instances will be included after further use case and requirements study¡±. With this statement, I¡¯d solicit you all to bring use case and requirement for the pooling of non-virtualized function, to help the future solution developed by the WG to cover both. :)
[Zu Qiang] Great.


- Some statements, such as ¡°Identify and analyze reliable transport protocols¡±, sounds like we do have a specific solution already. I won¡¯t assume there is always an IP connection between the pool manager and the pool instance. The interface between the two entities can be a TCP based protocol if there is an IP connection, or it can be an API if they both locate in the same server. It is real dependence on the solutions itself. Besides, if HW redundancy is needed that there may be always a ¡°remote¡± instance. The question is really how the ¡°remote¡± instance is handled, by whom; what homogeneity is assumed or whether this should be part of the investigation. If it is expected that the VNF instances of a pool come from different vendors then even the state synchronization becomes an issue. If the pools are expected to be built from identical instances then one questions why should the pool manager come from a different vendor. I won¡¯t want to see any solution limitations at this stage of the discussion. And I prefer to keep it open until the solution discussion phase.

[Ning] I think the statement ¡°Identify and analyze reliable transport protocols¡± is more like a requirement for the reliable delivery of pooling messages. We give examples of existing IETF solution like MPTCP, but this doesn¡¯t mean we have already determined the solution at this stage. You are right that we are open to solutions and solution discussion will be in next phase.
[Zu Qiang] If that¡¯s the case, please make it clear in the charter. And I propose to reword with ¡°Identify and analyze reliable interfaces transport protocols¡±

[Ning] I am OK with your proposal.


- My assumption is that the output from this proposed WG is one or more generic HA solutions. As we are in the early stage of the discussion, I don¡¯t think we shall limit the solution to a specific one. I do have a hard time to understand why the Applicability and Gap Analysis shall be done based on a specific protocol?

[Ning] I believe you had misunderstood the Applicability and Gap Analysis I-D (to RSerPool) at this point. :) I think anyone (other than RSerPool folks) can bring their I-D about applicability and gap analysis to solution ¡°x¡±. Currently we have an I-D based on RSerPool. But this doesn¡¯t mean that the WG will start solution based on RSerPool. We will be looking at more reference solutions and find out which parts of them can be reused/extended/referred/etc.
[Zu Qiang] If that¡¯s the case, it would be good to make it clear in the charter to avoid any misunderstandings or layer 9 issues.

[Ning] Deal! No issues above layer 7. :)


- Reading the charter, it is not very clear that what the scope of the VNF is. Is it the service/function, or the entities providing this service/function? Some clarification of this use would be good.

[Ning] In our charter, a VNF includes a pool of VNF instances implementing a certain function (not a whole service), and a manager to interface with the outside service control entity to provide such function. So I assume a VNF has both roles in your question. Does it make sense? Or you have some suggested text to further clarify this issue? :)
[Zu Qiang] So the WG will work on the Pool enabled VNF (not the generic VNF), right? :) And a Pool enabled VNF including the VNF function and the network entities (e.g. HW, interface, etc.), right? It is better to add this clarification in the charter.

[Ning]Yes, the WG will focus on Pool enabled VNF. However, to be clear, we are not working on HW/hypervisor layer functionalities. We focus on VNF function, and the associated resource status for pool OAM purpose only (as discussed in the second item above).

I hope my reply helps to address your concerns. Otherwise please let me know.

-Ning

Loading...