RE: Bluestore deferred writes for new objects

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> 
> Is the cutoff for the deferred writes > 64kb or >= 64kb by default?
> I.e., is a 64kb write deferred or not by default?
> 

Looks like its <=
https://github.com/ceph/ceph/blob/864594f473da1f04c476eb27073a8c2ea735a982/src/os/bluestore/BlueStore.cc#L10487

And checking the default options, it looks like it's now 32kb, not 64kb from what I remember, so apologies for that.

> 
> Paul
> Am Do., 11. Okt. 2018 um 00:37 Uhr schrieb Sage Weil <sage@xxxxxxxxxxxx>:
> >
> > On Wed, 10 Oct 2018, Nick Fisk wrote:
> > > Following up from a discussion on the performance call last week.
> > >
> > > Is anybody able to confirm the behaviour of Bluestore deferred writes with new objects?
> > >
> > > From my testing it appears that new object are always directly
> > > written to the underlying block device and not buffered into flash, whereas existing objects <64KB are.
> >
> > I just did a quick test (on master) and it looks like the deferred
> > writes are working as expected in that they *do* apply to new objects
> > (as well as existing ones).  Can you be a bit more specific about what
> > you observed?  (which version?  what workload?)
> >
> > Thanks!
> > sage
> >
> 
> 
> --
> Paul Emmerich
> 
> Looking for help with your Ceph cluster? Contact us at https://croit.io
> 
> croit GmbH
> Freseniusstr. 31h
> 81247 München
> www.croit.io
> Tel: +49 89 1896585 90




[Index of Archives]     [CEPH Users]     [Ceph Large]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux