All of lore.kernel.org
 help / color / mirror / Atom feed
* Move RBD Cache on clients from RAM to SSD.
@ 2015-06-16  7:59 Mike
  2015-06-16  9:54 ` Sebastien Han
  0 siblings, 1 reply; 5+ messages in thread
From: Mike @ 2015-06-16  7:59 UTC (permalink / raw)
  To: ceph-devel

Hello.

By now rbd_cache uses RAM, can we store cache on disk (SSD, SSD NVE)?
In this way I look some benefits:
* Don't waste a RAM for cache
* Can make a big cache on SSD: 2-8Gb per cache
* If you use Cache Tier - don't waste space on them
* No need addition Ceph servers - use Compute with SSD.

Yes it's for write-through and read-only cache strategies mostly.

Any ideas? My be I am blind and this idea totally stupid?


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Move RBD Cache on clients from RAM to SSD.
  2015-06-16  7:59 Move RBD Cache on clients from RAM to SSD Mike
@ 2015-06-16  9:54 ` Sebastien Han
  2015-06-16 11:43   ` Mike
  0 siblings, 1 reply; 5+ messages in thread
From: Sebastien Han @ 2015-06-16  9:54 UTC (permalink / raw)
  To: Mike; +Cc: ceph-devel@vger.kernel.org

This is not possible at the moment but long ago a BP was register to allow a multiple backend functionality. (can’t find it anymore)
Unfortunately I don’t think this BP got much attention :/

> On 16 Jun 2015, at 09:59, Mike <mike.almateia@gmail.com> wrote:
> 
> Hello.
> 
> By now rbd_cache uses RAM, can we store cache on disk (SSD, SSD NVE)?
> In this way I look some benefits:
> * Don't waste a RAM for cache
> * Can make a big cache on SSD: 2-8Gb per cache
> * If you use Cache Tier - don't waste space on them
> * No need addition Ceph servers - use Compute with SSD.
> 
> Yes it's for write-through and read-only cache strategies mostly.
> 
> Any ideas? My be I am blind and this idea totally stupid?
> 
> --
> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html


Cheers.
–––– 
Sébastien Han 
Senior Cloud Architect 

"Always give 100%. Unless you're giving blood."

Mail: seb@redhat.com 
Address: 11 bis, rue Roquépine - 75008 Paris

--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Move RBD Cache on clients from RAM to SSD.
  2015-06-16  9:54 ` Sebastien Han
@ 2015-06-16 11:43   ` Mike
  2015-06-16 11:47     ` Mark Nelson
  0 siblings, 1 reply; 5+ messages in thread
From: Mike @ 2015-06-16 11:43 UTC (permalink / raw)
  To: Sebastien Han; +Cc: ceph-devel@vger.kernel.org

16.06.2015 12:54, Sebastien Han пишет:
> This is not possible at the moment but long ago a BP was register to allow a multiple backend functionality. (can’t find it anymore)
> Unfortunately I don’t think this BP got much attention :/

Thanks for answer. It's sad, that this feature not interesting anyone.
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Move RBD Cache on clients from RAM to SSD.
  2015-06-16 11:43   ` Mike
@ 2015-06-16 11:47     ` Mark Nelson
  2015-06-16 11:55       ` Mike
  0 siblings, 1 reply; 5+ messages in thread
From: Mark Nelson @ 2015-06-16 11:47 UTC (permalink / raw)
  To: Mike, Sebastien Han; +Cc: ceph-devel@vger.kernel.org

As an alternative, would dm-cache or bcache on the client VM be an option?

Mark

On 06/16/2015 06:43 AM, Mike wrote:
> 16.06.2015 12:54, Sebastien Han пишет:
>> This is not possible at the moment but long ago a BP was register to allow a multiple backend functionality. (can’t find it anymore)
>> Unfortunately I don’t think this BP got much attention :/
>
> Thanks for answer. It's sad, that this feature not interesting anyone.
> --
> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Move RBD Cache on clients from RAM to SSD.
  2015-06-16 11:47     ` Mark Nelson
@ 2015-06-16 11:55       ` Mike
  0 siblings, 0 replies; 5+ messages in thread
From: Mike @ 2015-06-16 11:55 UTC (permalink / raw)
  To: Mark Nelson, Sebastien Han; +Cc: ceph-devel@vger.kernel.org

16.06.2015 14:47, Mark Nelson пишет:
> As an alternative, would dm-cache or bcache on the client VM be an option?
> 
> Mark

I not think it's working witch librbd in usecase with libvirt. This
solution working with kernel rbd module when you can associate rbd block
device with bcache device.
I think so.
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2015-06-16 11:55 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-06-16  7:59 Move RBD Cache on clients from RAM to SSD Mike
2015-06-16  9:54 ` Sebastien Han
2015-06-16 11:43   ` Mike
2015-06-16 11:47     ` Mark Nelson
2015-06-16 11:55       ` Mike

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.