All of lore.kernel.org
 help / color / mirror / Atom feed
* dm-clock queue
@ 2015-11-04 22:35 Sage Weil
  2015-11-05 22:19 ` J. Eric Ivancich
  2015-11-06  5:58 ` Christoph Hellwig
  0 siblings, 2 replies; 6+ messages in thread
From: Sage Weil @ 2015-11-04 22:35 UTC (permalink / raw)
  To: g.marripudi, ivancich; +Cc: mbenjami, sjust, ceph-devel

Hi Gunna, Eric-

I wanted to make sure you were connected as we've talked to both of you 
independently about the new request queue in the OSD to support dm-clock 
and I want to make sure our efforts are coordinated.  I thnk the first 
goal is probably to implement something that works and performs well for 
just a few request classes (clients, recovery, scrub, snaptrim).  
Eventually we'll also need to determine if/how to do so for a large client 
count so that we can do client qos.  Just solving the first problem alone 
may be a big win, though: we hear lots of complaints about the effect of 
recovery on client io.

Anyway, just wanted to make sure you two were connected and kick off the 
conversation.

sage

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: dm-clock queue
  2015-11-04 22:35 dm-clock queue Sage Weil
@ 2015-11-05 22:19 ` J. Eric Ivancich
  2015-11-06  5:45   ` Gunna (Gunneswara) Marripudi-SSI
  2015-11-06  5:58 ` Christoph Hellwig
  1 sibling, 1 reply; 6+ messages in thread
From: J. Eric Ivancich @ 2015-11-05 22:19 UTC (permalink / raw)
  To: Sage Weil, g.marripudi; +Cc: mbenjami, sjust, ceph-devel

Thank you, Sage. I will contact you in a separate message, Gunna, to 
better understand your efforts and interests.

Eric


On 11/04/2015 05:35 PM, Sage Weil wrote:
> Hi Gunna, Eric-
>
> I wanted to make sure you were connected as we've talked to both of you
> independently about the new request queue in the OSD to support dm-clock
> and I want to make sure our efforts are coordinated.  I thnk the first
> goal is probably to implement something that works and performs well for
> just a few request classes (clients, recovery, scrub, snaptrim).
> Eventually we'll also need to determine if/how to do so for a large client
> count so that we can do client qos.  Just solving the first problem alone
> may be a big win, though: we hear lots of complaints about the effect of
> recovery on client io.
>
> Anyway, just wanted to make sure you two were connected and kick off the
> conversation.
>
> sage
>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* RE: dm-clock queue
  2015-11-05 22:19 ` J. Eric Ivancich
@ 2015-11-06  5:45   ` Gunna (Gunneswara) Marripudi-SSI
  0 siblings, 0 replies; 6+ messages in thread
From: Gunna (Gunneswara) Marripudi-SSI @ 2015-11-06  5:45 UTC (permalink / raw)
  To: J. Eric Ivancich, Sage Weil
  Cc: mbenjami@redhat.com, sjust@redhat.com, ceph-devel@vger.kernel.org

Sage,

Thanks for the introduction.

Eric,

Glad to meet you over email. I will respond to you on a separate thread.

Regards,

Gunna

-----Original Message-----
From: J. Eric Ivancich [mailto:ivancich@redhat.com] 
Sent: Thursday, November 05, 2015 2:19 PM
To: Sage Weil; Gunna (Gunneswara) Marripudi-SSI
Cc: mbenjami@redhat.com; sjust@redhat.com; ceph-devel@vger.kernel.org
Subject: Re: dm-clock queue

Thank you, Sage. I will contact you in a separate message, Gunna, to better understand your efforts and interests.

Eric


On 11/04/2015 05:35 PM, Sage Weil wrote:
> Hi Gunna, Eric-
>
> I wanted to make sure you were connected as we've talked to both of 
> you independently about the new request queue in the OSD to support 
> dm-clock and I want to make sure our efforts are coordinated.  I thnk 
> the first goal is probably to implement something that works and 
> performs well for just a few request classes (clients, recovery, scrub, snaptrim).
> Eventually we'll also need to determine if/how to do so for a large 
> client count so that we can do client qos.  Just solving the first 
> problem alone may be a big win, though: we hear lots of complaints 
> about the effect of recovery on client io.
>
> Anyway, just wanted to make sure you two were connected and kick off 
> the conversation.
>
> sage
>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: dm-clock queue
  2015-11-04 22:35 dm-clock queue Sage Weil
  2015-11-05 22:19 ` J. Eric Ivancich
@ 2015-11-06  5:58 ` Christoph Hellwig
  2015-11-06  6:17   ` Gunna (Gunneswara) Marripudi-SSI
  1 sibling, 1 reply; 6+ messages in thread
From: Christoph Hellwig @ 2015-11-06  5:58 UTC (permalink / raw)
  To: Sage Weil; +Cc: g.marripudi, ivancich, mbenjami, sjust, ceph-devel

Can someone explain what dm-clock is?

^ permalink raw reply	[flat|nested] 6+ messages in thread

* RE: dm-clock queue
  2015-11-06  5:58 ` Christoph Hellwig
@ 2015-11-06  6:17   ` Gunna (Gunneswara) Marripudi-SSI
  2015-11-06  6:22     ` Christoph Hellwig
  0 siblings, 1 reply; 6+ messages in thread
From: Gunna (Gunneswara) Marripudi-SSI @ 2015-11-06  6:17 UTC (permalink / raw)
  To: Christoph Hellwig, Sage Weil
  Cc: ivancich@redhat.com, mbenjami@redhat.com, sjust@redhat.com,
	ceph-devel@vger.kernel.org

This generally refers to QoS work on Ceph.

This was discussed in one of the perf calls about a month ago; on 9/30/2015. 
https://bluejeans.com/playback/s/ZqUMSWr0fBl8GR1ZrU0OrZDnM3KnglsYxTiUG5X7oUeCixrlsANNPQFuXlPojBp9#

Please refer to the following paper for additional details.

https://www.usenix.org/legacy/event/osdi10/tech/full_papers/Gulati.pdf

Regards,

Gunna

-----Original Message-----
From: Christoph Hellwig [mailto:hch@infradead.org] 
Sent: Thursday, November 05, 2015 9:58 PM
To: Sage Weil
Cc: Gunna (Gunneswara) Marripudi-SSI; ivancich@redhat.com; mbenjami@redhat.com; sjust@redhat.com; ceph-devel@vger.kernel.org
Subject: Re: dm-clock queue

Can someone explain what dm-clock is?

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: dm-clock queue
  2015-11-06  6:17   ` Gunna (Gunneswara) Marripudi-SSI
@ 2015-11-06  6:22     ` Christoph Hellwig
  0 siblings, 0 replies; 6+ messages in thread
From: Christoph Hellwig @ 2015-11-06  6:22 UTC (permalink / raw)
  To: Gunna (Gunneswara) Marripudi-SSI
  Cc: Christoph Hellwig, Sage Weil, ivancich@redhat.com,
	mbenjami@redhat.com, sjust@redhat.com, ceph-devel@vger.kernel.org

Oh, ok - so ti's not a device mapper module.  Thanks a for the
clarification!

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2015-11-06  6:22 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-11-04 22:35 dm-clock queue Sage Weil
2015-11-05 22:19 ` J. Eric Ivancich
2015-11-06  5:45   ` Gunna (Gunneswara) Marripudi-SSI
2015-11-06  5:58 ` Christoph Hellwig
2015-11-06  6:17   ` Gunna (Gunneswara) Marripudi-SSI
2015-11-06  6:22     ` Christoph Hellwig

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.