Re: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute

["Les Ginsberg (ginsberg)" <[e-mail]>]

Robert –

I agree there is still a concern here – which has to do with how the new metric is calculated.

Imagine  that I have a million users in a metro area and they are all WFH (gee – when would that ever happen?? 😊 ) So they aren’t really mobile.
Are we going to have all users initially directed to Server #1 – and then after some period move all users to Server #2 just because Server #1 is busier than Server #2? (I hope not)
It seems to me that how the metric is calculated has cross-site implications.
This is out of scope for the IGP – we simply use what we are given.
But whatever entity is calculating the metric to be used has to be able to do so in a way that doesn’t cause spurious rerouting.

Seems to me you are asking Linda this question in one of your other posts.

My agreement on the use of the IGP assumes that the entity calculating the metric to be provided to the IGP has the correct intelligence.

   Les


From: Robert Raszuk <[e-mail]>
Sent: Thursday, January 13, 2022 8:52 AM
To: Les Ginsberg (ginsberg) <[e-mail]>
Cc: Linda Dunbar <[e-mail]>; Gyan Mishra <[e-mail]>; [e-mail]
Subject: Re: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute


> and I agree that using the IGP/Flex Algo as you are proposing is a viable solution.

Except that IGP usually does not run between application server and load balancer ...


On Thu, Jan 13, 2022 at 5:37 PM Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>> wrote:
Linda –

Are you saying that the scenario you are trying to address is to have a given “transaction” go to the currently closest/most lightly loaded Application Server?
And there is no intent to support for long lived connections?

If so, then this isn’t really a load balancing issue and I agree that using the IGP/Flex Algo as you are proposing is a viable solution.
The concern then becomes the rate of updates to the new Prefix Metric. If this changes too rapidly this will heavily consume network resources by triggering flooding, SPF, forwarding plane updates at a high rate.
Can you put some language in the draft that indicates the expected rate of updates to the metric and some guidelines on limiting the frequency?

Thanx.

   Les


From: Linda Dunbar <[e-mail]<mailto:[e-mail]>>
Sent: Thursday, January 13, 2022 7:58 AM
To: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Cc: Gyan Mishra <[e-mail]<mailto:[e-mail]>>; Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>
Subject: RE: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute

Robert,

For your scenario of TCP flows lasting more than 8~18 hours,  multiple server instances SHOULDN’T be assigned with the SAME IP address (ANYCAST address).  Each of those instances should have one distinct IP address.

The draft is for different scenario where application are instantiated at multiple locations behind multiple App Layer Load Balancers. They have relative short flows that can go to any App Layer Load Balancers. Multiple Load Balancers  for those applications are assigned with the same IP address. In Kubernetes, multiple load balancers for one type of microservices are assigned with one single Virtual IP address, so that the network can forward as one single destination.

Linda
From: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Sent: Thursday, January 13, 2022 9:37 AM
To: Linda Dunbar <[e-mail]<mailto:[e-mail]>>
Cc: Gyan Mishra <[e-mail]<mailto:[e-mail]>>; Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>
Subject: Re: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute


> Flows among micro-services are very short.

While that can be true - there is nowhere in the document a restriction or even a warning that this solution is aiming for short lived flows only and that users should be well aware about drastic nature of proposed mechanism to their established flows.

In one of the companies I worked for average  TCP flow duration was anywhere from 8-18 hours and it was a very drastic event for the user to loose it in the middle of the day.  Various means where taken and applied to protect such sessions from any form of disconnects.

I think we are shooting here with the wrong weapon to the target.

Thx,
R.

On Thu, Jan 13, 2022 at 4:23 PM Linda Dunbar <[e-mail]<mailto:[e-mail]>> wrote:
Robert,

Your link to Traefik  adds more reasons why “Site index and preference” should be distributed by IGP:

  *   Site index and preferences for a cluster of micro-service instances are rarely dynamically changed. Most of those values are configured. Therefore, the oscillation is minimal.
  *   Flows among micro-services are very short. Put less requirements to flow affinity.
  *

Linda


From: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Sent: Thursday, January 13, 2022 5:00 AM
To: Gyan Mishra <[e-mail]<mailto:[e-mail]>>
Cc: Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>>; Linda Dunbar <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>
Subject: Re: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute


And just to provide a sound alternative to the objective of this work.

Please consider using Traefik - https://traefik.io/<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Ftraefik.io%2F&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=P%2BItvofz3%2Bgg0KEMdfxe9MluMPkQ2v1jL8a1Q50rddo%3D&reserved=0>

Thx,
R.


On Thu, Jan 13, 2022 at 11:56 AM Robert Raszuk <[e-mail]<mailto:[e-mail]>> wrote:
Gyan,

I see what the draft is trying to do now. /* I did not even consider this for the reason described below. */

But what you wrote requires little correction:

"So now the server you are on gets overloaded and a site cost gets advertised in the IGP at which point the connection receives a TCP reset"

if you s/connection/all connections/ then you quickly realize that what is proposed here is a disaster.

Breaking all existing flows going to one LB to suddenly timeout and all go to the other LB(s) is never a technique any one would seriously deploy in a production network.

Leave alone that doing that has potential to immediately overload the other LB(s)/server(s) too.

For me the conclusion is that IGP transport level is not the proper layer to address the requirement.

Cheers,
Robert.


On Thu, Jan 13, 2022 at 7:05 AM Gyan Mishra <[e-mail]<mailto:[e-mail]>> wrote:

Hi Les

Agreed.

My thoughts are that the context of the draft is based on an Anycast VIP address of a server which is proximity based load balancing and not necessarily ECMP/UCMP and only if the proximity is the same for multiple paths to the Anycast VIP would there be a ECMP/UCMP possibility.

Let’s say if it’s proximity based and one path is preferred, the flow will take that path.  So now the server you are on gets overloaded and a site cost gets advertised in the IGP at which point the connection receives a TCP reset and flow re-establishes on the alternate path based on the site cost and remains there until the server goes down or  gets overloaded or a better path comes along.

For ECMP case, ECMP has flow affinity so the flow will stay on the same path long lived until the connection terminates.

So now in ECMP case the flow hashed to a path and maintains its affinity to that path.  Now all of sudden the server gets overloaded and we get a better site cost advertised.  So now the session terminates on current path and establishes again on the Anycast VIP new path based on the site cost advertised.

The failover I believe results in the user refreshing their browser which is better than hanging.

As the VIP prefix is the only one that experiences reconvergence on new path based on site cost if there is any instability with the servers that will be reflected to the IGP Anycast prefix as well.

Is that a good or bad thing.  I think if you had to pick your poison as here the issue Linda is trying to solve is a server issue but leveraging the IGP to force re-convergence when the server is in a half baked state meaning it’s busy and connections are being dropped or very slow QOE for end user.  If you did nothing and let it ride the the user would be stuck on a bad connection.

So this solution dynamically fixed the issue.

As far as oscillation that is not a big deal as you are in a much worse off state connected to a dying server on its last leg as far as memory and CPU.

This solution I can see can apply to any client / server connection and not just 5G and can be used by enterprises as well as SP for their customers to have an drastically improved QOE.

I saw some feedback on the TLV and I think once that is all worked out I am in favor of advancing this draft.

Kind Regards

Gyan


On Wed, Jan 12, 2022 at 10:16 PM Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>> wrote:
Gyan –

The difference between ECMP and UCMP is not significant in this discussion.
I don’t want to speak for Robert, but for me his point was that IGPs can do “multipath” well – but this does not translate into managing flows.
Please see my other responses on this thread.

Thanx.

    Les


From: Gyan Mishra <[e-mail]<mailto:[e-mail]>>
Sent: Wednesday, January 12, 2022 5:26 PM
To: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Cc: Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>>; Linda Dunbar <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>
Subject: Re: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute


Robert

Here are a few examples of UCMP drafts below used in core and data center use cases.

https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-unequal-lb-15<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fhtml%2Fdraft-ietf-bess-evpn-unequal-lb-15&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=G8h11WpirgxNmk2wzr6QN9DsnYBNGQ42ft7Cz9E8pAk%3D&reserved=0>

https://datatracker.ietf.org/doc/html/draft-mohanty-bess-weighted-hrw-02<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fhtml%2Fdraft-mohanty-bess-weighted-hrw-02&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=zcwF%2F%2FKh77N6jyXDOXuftPALvaUb%2B2Kvj2G2tedfvL0%3D&reserved=0>

https://datatracker.ietf.org/doc/html/draft-ietf-idr-link-bandwidth<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fhtml%2Fdraft-ietf-idr-link-bandwidth&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=nJcXT7NffiSt7CJh%2F2bnqaa7ClnxMSCf%2BVproPb34s0%3D&reserved=0>

https://datatracker.ietf.org/doc/html/draft-mohanty-bess-ebgp-dmz<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fhtml%2Fdraft-mohanty-bess-ebgp-dmz&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=AbVKlR%2FrX1vhMzdzHL7J8VgiU2oxaqxSu9oMx9onJRo%3D&reserved=0>



There are many use cases in routing for unequal cost load balancing capabilities.

Kind Regards

Gyan

On Wed, Jan 12, 2022 at 2:23 PM Robert Raszuk <[e-mail]<mailto:[e-mail]>> wrote:
Linda,

> IGP has been used for the Multi-path computation for a long time

IGP can and does ECMP well. Moreover, injecting metric of anycast server destination plays no role in it as all paths would inherit that external to the IGP cost.

Unequal cost load balancing or intelligent traffic spread has always been done at the application layer *for example MPLS*

Thx a lot,
R.

On Wed, Jan 12, 2022 at 8:18 PM Linda Dunbar <[e-mail]<mailto:[e-mail]>> wrote:
Robert,

Please see inline in green:

From: Robert Raszuk <[e-mail]<mailto:[e-mail]>>
Sent: Wednesday, January 12, 2022 1:00 PM
To: Linda Dunbar <[e-mail]<mailto:[e-mail]>>
Cc: Les Ginsberg (ginsberg) <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>
Subject: Re: [Lsr] Seeking feedback to the revised draft-dunbar-lsr-5g-edge-compute

Hi Linda,


[LES:] It is my opinion that what you propose will not achieve your goals – in part because IGPs only influence forwarding on a per packet basis – not a per flow/connection basis.

[Linda] Most vendors do support flow based ECMP, with Shortest Path computed by attributes advertised by IGP.

I am with Les here. ECMP has nothing to do with his point.

[Linda] Les said that “IGP only influence forwarding on a per packet basis”.  I am saying that vendors supporting “forwarding per flow” with equal cost computed by IGP implies  that forwarding on modern routers are no longer purely per packet basis.


Draft says:

When those multiple server instances share one IP address (ANYCAST), the transient network and load conditions can be incorporated in selecting an optimal path among server instances for UEs.

So if we apply any new metric to indicate load of a single anycast address how is this going to help anything ?

[Linda] The “Load” or “Aggregated Site Cost” is to differentiate multiple paths with the same routing distance.


You would need a mechanism where the network is smart and say per src-dst tuple or more spreads the traffic. IGP does not play that game today I am afraid.
[Linda] There is one SRC and multiple paths to one DST. IGP has been used for the Multi-path computation for a long time.

Thank you, Linda

Thx a lot,
R.







_______________________________________________
Lsr mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/lsr<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Fmailman%2Flistinfo%2Flsr&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=4VvMlYIqiEyjlRQHQkRLXocNchpxnDGqBSfKG96GCaY%3D&reserved=0>
--

[Image removed by sender.]<https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.verizon.com%2F&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=RBLTHUJUf6MujlMd%2FIpMJH36fjYXm3diFx6nS9I28E0%3D&reserved=0>

Gyan Mishra

Network Solutions Architect

Email [e-mail]<mailto:[e-mail]>

M 301 502-1347

--

[Image removed by sender.]<https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.verizon.com%2F&data=04%7C01%7Clinda.dunbar%40futurewei.com%7C4361a0b36c614fbf8e7908d9d6aa8b4b%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637776850245075888%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=RBLTHUJUf6MujlMd%2FIpMJH36fjYXm3diFx6nS9I28E0%3D&reserved=0>

Gyan Mishra

Network Solutions Architect

Email [e-mail]<mailto:[e-mail]>

M 301 502-1347