Re: [nfsv4] Draft agenda

Hi group,

Any interest in discussing the implications of nconnect / pNFS sessions
traversing paths of significantly different properties (bandwidth,
latency) during the session if time permits?

Basically my observation, that most clients appear to be doing RR-load
balancing currently - while e.g. iSCSI MPIO would offer more advanced
scheduling mechanisms to optimize overall IOPS.

Best regards,
   Richard

Am 26.03.2023 um 12:28 schrieb Chris Inacio:
> All,
>
> IETF week has started - so here is the final agenda (until people bash
> it.). This is mostly just taking David Black’s asks into account.
>
> Also looking for slides from speakers if you haven’t sent them in,
> please send them to nfsv4-chairs@ietf.org
> <mailto:nfsv4-chairs@ietf.org> as soon as reasonable.  Ideally, all
> slides are in by Thursday midday Japan time.  If you have already sent
> your slides in - thank you.  I definitely want to acknowledge that we
> already have some slides.  If you don’t intend to have slides, please
> let the chairs know that as well.
>
>
> ====
>
> IETF-116 NFSv4 WG Agenda
> Welcome (back) to Yokohama
>
> * Chairs Welcome / Agenda Bash (chairs)
> * WG status and interim meetings (5 min - chairs/all)
>          - interim meetings and draft progress
> * pNFS layout using NVMe (15 min - Hellwig)
> * RDMA Commit (10 min - Tom Talpey)
> * 5661-bis (20 min - Noveck/Chairs)
> * Delstid - Extending opening of files / Reporting of Errors via
> LAYOUTRETURN in NFSv4.2 (15 min - Haynes)
> * Implementation Status (10 min - Lever)
> * Sequence ID Calibration (10 min - Yang Jing)
>
> * end of IETF-116, safe travels
>
>
> Thank you,
> Nfsv4 chairs
>
> ----
> Chris Inacio
> inacio@cert.org
>
>
>
>> On Mar 11, 2023, at 2:01 AM, Chris Inacio <inacio@cert.org> wrote:
>>
>> Hi Dave,
>>
>>
>> Thanks for the feedback.
>>
>> As to those within the meeting changes, I think that’s all doable.
>>  BUT, before I do those changes, Christoph is on site but leaving
>> before the Friday meeting time.  We would like to change the actual WG
>> time to accommodate that if possible.  Stay tuned - and then keep me
>> honest!  You might need another round of suggestions IF we can get the
>> meeting time changed; so I would look to hold off till then.
>>
>>
>> Regards,
>> ----
>> Chris Inacio
>> inacio@cert.org
>>
>>
>>
>>> On Mar 10, 2023, at 11:56 AM, Black, David <David.Black@dell.com> wrote:
>>>
>>> Chris and Dave,
>>> > The timing of this meeting really sucks for those in EDT.   This
>>> was not supposed to happen.
>>> timeanddate.com <http://timeanddate.com/>says that Friday, March 31
>>> 12:00pm (noon) in Yokohama (JST) is 11:00pm on Thursday, March 30 in
>>> Boston (EDT).
>>> I believe that I can make that time work.  It would help if the
>>> agenda were bashed so that Christoph’s pNFS layout using NVMe draft
>>> comes earlier with the aim of covering it (and Tom’s RDMA commit
>>> draft) before midnight EDT.  I’d suggest moving the pNFS layout using
>>> NVMe draft to follow the RDMA commit draft, if that doesn’t cause
>>> problems for other presenters.  Is that possible?
>>> Thanks, --David
>>> *From:*nfsv4 <nfsv4-bounces@ietf.org
>>> <mailto:nfsv4-bounces@ietf.org>>*On Behalf Of*Chris Inacio
>>> *Sent:*Friday, March 10, 2023 9:31 AM
>>> *To:*David Noveck
>>> *Cc:*NFSv4
>>> *Subject:*Re: [nfsv4] Draft agenda
>>>
>>> [EXTERNAL EMAIL]
>>>
>>>
>>>         Also, I would like to hear about any issues with the time
>>>         slot as scheduled.
>>>
>>>     You say you would like to hear about these issues but I don't
>>>     believe you. But at least, you are willing to hear about these
>>>     issues.
>>>     The timing of this meeting really sucks for those in EDT.   This
>>>     was not supposed to happen.
>>>
>>> Dave - as you know, I put comments in the schedule request asking for
>>> a time slot that was first thing in the morning to be as
>>> accommodative as possible to the participants I was aware of.  We
>>> didn’t get that slot.  I can have empathy without power to change the
>>> circumstances.  They powers that be, trying to solve their impossible
>>> scheduling problems, don’t give me preference (or you, or any of the
>>> other participants or chairs.) I would like this to be the best
>>> possible time slot for the working group.  I will ask again when I
>>> understand what people would like - but it doesn’t mean I can get it.
>>> With all that said - everyone, I’m still open to hearing feedback.
>>> Thanks
>>> Chris (as a chair)
>>
>
>
> _______________________________________________
> nfsv4 mailing list
> nfsv4@ietf.org
> https://www.ietf.org/mailman/listinfo/nfsv4