lkml.org 
[lkml]   [2012]   [Dec]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH, 3.7-rc7, RESEND] fs: revert commit bbdd6808 to fallocate UAPI
On Fri, Dec 07, 2012 at 05:17:05PM -0700, Dave Chinner wrote:
> On Fri, Dec 07, 2012 at 02:03:06PM -0500, Chris Mason wrote:

[ dead and beaten fallocate ponies ]

>
> > On a single flash drive doing random 4K writes, xfs does 950MB/s into
> > regular extents but only 400MB/s into preallocated extents.
> >
> > http://masoncoding.com/presentation/perf-linuxcon12/fallocate.png
>
> This is bordering on irrelevancy, but can you provide the workload
> you were running to generate this graph? Random 4k writes could be
> anything, really.

This one was fio aio/dio, I'll dig out the job file and rerun it on
3.7-rc on Monday. Any real random write is going to show this with
enough load.

>
> In my experience, applications that actually do processing between
> random write IOs don't see anywhere near the same degradation as
> such micro-benchmarks tend to indicate can occur with unwritten
> extents. Are you seeing this level of degradation in real-world applications?
> If you give me a reason to fix it (and the hardware to test it on),
> I'm pretty sure I can bring the overhead down to just a few percent
> on fully featured SSDs like FusionIO devices...

We should have a card I can send, drop me the address.

For the workload...that's harder. We can talk all day about what a
normal random write workload is, but if you have a fio job that you
think represents real world, I can run that.

[ much nodding ;) ]


\
 
 \ /
  Last update: 2012-12-08 03:01    [W:0.952 / U:0.216 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site