Re: [ippm] New Version Notification for draft-li-ippm-otwamp-on-lag-02.txt

Hi Haoyu,

Yes, comparing even and odd LAGs - deviations from ideal load-sharing are very similar.
Obviously - there’s no need for load-balancing across a single link ;-)

Cheers,
Jeff

> On Feb 4, 2022, at 11:02, Haoyu Song <haoyu.song@futurewei.com> wrote:
> 
> 
> Hi Jeff,
>  
> It’s interesting to know. Any idea how the load is balanced on the odd number of links?
> BTW, in your example, a single link is also power of 2 (i.e., 2^0) and there’s no need for load balancing.
>  
> Best,
> Haoyu
>  
> From: Jeff Tantsura <jefftant.ietf@gmail.com> 
> Sent: Friday, February 4, 2022 10:57 AM
> To: Tianran Zhou <zhoutianran=40huawei.com@dmarc.ietf.org>; Haoyu Song <haoyu.song@futurewei.com>; IETF IPPM WG (ippm@ietf.org) <ippm@ietf.org>
> Subject: RE: [ippm] New Version Notification for draft-li-ippm-otwamp-on-lag-02.txt
>  
> Hi,
>  
> Wrt power of two comment from Haoyu.
> This used to be the case in the early days because of poorly implemented load-balancing in the silicon, it was often the case that with odd number of constituents we’d see a rather uneven distribution.
> Most of these issues have been fixed in the last 3 years, these days - it is very common (especially in DC) to use a LAG as an operational construct and  provision a LAG with a single link, so the future upgrade doesn’t require changes to L3 infra.
>  
> I don’t’ think there a need to update the drawing nor to explicitly mentioning  “power of two” assumption anywhere in the text .
>  
> Cheers,
> Jeff
>  
> From: Tianran Zhou
> Sent: Monday, January 24, 2022 5:06 PM
> To: Haoyu Song; Tianran Zhou; IETF IPPM WG (ippm@ietf.org)
> Subject: Re: [ippm] New Version Notification for draft-li-ippm-otwamp-on-lag-02.txt
>  
> Hi Haoyu,
>  
> Thanks very much for your comments. Your perspective is correct.
> I will revise the document in the next version.
>  
> Best,
> Tianran
>  
> -----Original Message-----
> From: ippm [mailto:ippm-bounces@ietf.org] On Behalf Of Haoyu Song
> Sent: Tuesday, January 25, 2022 1:07 AM
> To: Tianran Zhou <zhoutianran=40huawei.com@dmarc.ietf.org>; IETF IPPM WG (ippm@ietf.org) <ippm@ietf.org>
> Subject: Re: [ippm] New Version Notification for draft-li-ippm-otwamp-on-lag-02.txt
>  
> WG,
>  
> I think the scenario for LAG performance measurement is practical and the method is sound, so I support this work.
> For Figure 1, I think usually for load balancing with hashing, the number of links of a LAG would be power of two.
>  
> Best,
> Haoyu Song
>  
> -----Original Message-----
> From: ippm <ippm-bounces@ietf.org> On Behalf Of Tianran Zhou
> Sent: Sunday, January 23, 2022 10:23 PM
> To: IETF IPPM WG (ippm@ietf.org) <ippm@ietf.org>
> Subject: [ippm] FW: New Version Notification for draft-li-ippm-otwamp-on-lag-02.txt
>  
> Hi WG,
>  
> We think this is an useful work on adapting OWAMP/TWAMP in LAG performance measurement.
> We updated this draft based on the comments from the meeting discussion and mailing list.
> Thanks for all the suggestions.
> More reviews and comments are welcome.
>  
> Cheers,
> Tianran
>  
> -----Original Message-----
> From: internet-drafts@ietf.org [mailto:internet-drafts@ietf.org]
> Sent: Monday, January 24, 2022 2:17 PM
> To: Greg Mirsky <gregimirsky@gmail.com>; Guo Jun <guo.jun2@zte.com.cn>; Jun Guo <guo.jun2@zte.com.cn>; Rakesh Gandhi <rgandhi@cisco.com>; Tianran Zhou <zhoutianran@huawei.com>; Zhenqiang Li <li_zhenqiang@hotmail.com>
> Subject: New Version Notification for draft-li-ippm-otwamp-on-lag-02.txt
>  
>  
> A new version of I-D, draft-li-ippm-otwamp-on-lag-02.txt
> has been successfully submitted by Tianran Zhou and posted to the IETF repository.
>  
> Name:               draft-li-ippm-otwamp-on-lag
> Revision:           02
> Title:                  One-way/Two-way Active Measurement Protocol Extensions for Performance Measurement on LAG
> Document date:           2022-01-24
> Group:               Individual Submission
> Pages:               13
> URL:            https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Farchive%2Fid%2Fdraft-li-ippm-otwamp-on-lag-02.txt&amp;data=04%7C01%7Chaoyu.song%40futurewei.com%7Ca16115736bab48d99c2f08d9df020bed%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637786022170908705%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=O1SD54x5%2FM99g7YjxnzaMjaZuJ9S4UxNGcm6NAkTXng%3D&amp;reserved=0
> Status:         https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fdraft-li-ippm-otwamp-on-lag%2F&amp;data=04%7C01%7Chaoyu.song%40futurewei.com%7Ca16115736bab48d99c2f08d9df020bed%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637786022170908705%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=UYUXVf0BvrCuH20ydOjxCI26KjgpAopEIV40%2BqjuxKY%3D&amp;reserved=0
> Htmlized:       https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fhtml%2Fdraft-li-ippm-otwamp-on-lag&amp;data=04%7C01%7Chaoyu.song%40futurewei.com%7Ca16115736bab48d99c2f08d9df020bed%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637786022170908705%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=gN9hSIQaMM%2BE%2BtDLSFxe95J%2FyX4omBdv6rtBTMB4B8o%3D&amp;reserved=0
> Diff:           https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Frfcdiff%3Furl2%3Ddraft-li-ippm-otwamp-on-lag-02&amp;data=04%7C01%7Chaoyu.song%40futurewei.com%7Ca16115736bab48d99c2f08d9df020bed%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637786022170908705%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=NPJcFQoHUAw47%2FJ%2BO4SeOLEaQUNe%2FAI4UDUGMA%2BKK%2Fo%3D&amp;reserved=0
>  
> Abstract:
>    This document defines extensions to One-way Active Measurement
>    Protocol (OWAMP), and Two-way Active Measurement Protocol (TWAMP) to
>    implement performance measurement on every member link of a Link
>    Aggregation Group (LAG).  Knowing the measured metrics of each member
>    link of a LAG enables operators to enforce the performance based
>    traffic steering policy across the member links.
>  
>  
>                                                                                  
>  
>  
> The IETF Secretariat
>  
>  
> _______________________________________________
> ippm mailing list
> ippm@ietf.org
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Fmailman%2Flistinfo%2Fippm&amp;data=04%7C01%7Chaoyu.song%40futurewei.com%7Ca16115736bab48d99c2f08d9df020bed%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637786022170908705%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&amp;sdata=DHVbIOrP6nfAfjGnIl5N%2F%2B5B0O6qDonb%2F%2Beq2uTEJu4%3D&amp;reserved=0
>  
> _______________________________________________
> ippm mailing list
> ippm@ietf.org
> https://www.ietf.org/mailman/listinfo/ippm
>  
> _______________________________________________
> ippm mailing list
> ippm@ietf.org
> https://www.ietf.org/mailman/listinfo/ippm
>  