Re: nfsd write throughput

[Greg Banks <gnb@sgi.com>]

On Tue, Aug 03, 2004 at 12:32:14PM +0200, Olaf Kirch wrote:
> Hi folks,
> 
> I've been looking at the problem from a different angle...
> 
> Theory:
> 	The main bottleneck is that we spend a long time in commit(),
> 	blocking other WRITE calls from making any progress (thereby
> 	stalling all NFS clients). The reason is what we take inode->i_sem
> 	in nfsd_sync, but the writev() code wants to grab the same
> 	semaphore.
> 
> Circumstantial Evidence:
> 
> I've been doing some tests with the latencies of WRITE and COMMIT, using a
> single stream write. The average time we spend in nfsd_write is miniscule,
> usually it's less than 2 milliseconds. However when a commit comes in,
> we take a hit there as well - something around 500 ms for reiser, and
> 400 ms for ext3. Syncing to reiser frequently takes up to 1.2 seconds,
> while the 400 ms for ext3 is pretty constant.

With IRIX clients, which do far fewer COMMITs than (at least 2.4) Linux
clients, I have seen COMMIT latencies in the order of multiple seconds
as over a gigabyte of data is written to disk at 180 MB/s.

> Right now, nfsd_sync calls
> 
> 	filemap_fdatawrite
> 	filp->f_op->fsync
> 	filemap_fdatawait
> 
> all under the i_sem. However, it seems we don't need the i_sem for
> the filemap_* functions (is that valid - at least sync_page_range
> doesn't?). So I changed the code to make it grab i_sem only for the fsync
> call, but unfortunately, that doesn't seem to make much of a difference,
> as I found out. Most of the time taken by a commit is spent in fsync
> (the delta between the fsync latency and the overall commit latency is
> usually less than 5 ms, i.e. ~1%).
> 
> I also changed nfsd_sync to call filemap_fdatawrite_range instead of
> filemap_fdatawrite, but that doesn't make a noticeable difference either.

I tried this many months ago on 2.4, including tweaking the VFS layer to
pass a ranged flush down to XFS, also without any noticeable effect.
The performance limitation was nfsds being locked out because they were
unable to get the BKL which was held in sync_old_buffers().  In the
single streaming writer case reducing the flush range made no difference
to the number of pages queued to disk and hence no difference to
performance.

BTW this BKLage is the main reason why I gave up hope trying to get the
2.4 kernel's write performance up.

> I then re-enabled my flushfast hack, and the commit latencies went
> down to 30 ms on ext3, with the occasional spike of 300 ms. On reiser,
> the commit latency went down to something like 50 ms on average.
> 
> (The reiserfs rewrite case was fairly bad, however. Rewrite over NFS
> on top of reiser is fairly slow to begin with, much slower than write;
> and the gain from the flushfast patch is minimal - but that's a different
> story)
> 
> 
> Conclusion:
> 
> So this at least supports my theory that the commits are throttling the
> writes quite a bit.

Indeed.

> For the sake of completeness, I did some more iozone
> measurements, and on write/rewrite the performance gain is about 50%
> on both reiser and ext3, for a single client. I would think for several
> clients writing concurrently, the gain should be even more pronounced,
> but I haven't run these tests yet.
> 
> I'm wondering what could happen if we change nfsd_sync to not take the
> i_sem at all... I'll talk to a few VFS folks around here and try to find out.

I imagine they will not be thrilled by the idea.

I still think the best approach is to get the page cache to start
pushing unstable NFS pages to disk more aggresively, after the WRITE
but before the COMMIT.  This should avoid long waits for disk IO
with i_sem held; IIRC the page cache will only hold i_sem long enough
to traverse page lists, allowing another WRITE call to get in soon.

> PS:
> 
> Another thing I noticed was that the commit calls sent by the Linux
> client (2.6.5) are not evenly distributed over time. Much of the time,
> the client will call COMMIT 4-6 times a second, and then all of a sudden
> I see 30-80 calls a second several times in a row.

That's not good.

Greg.
-- 
Greg Banks, R&D Software Engineer, SGI Australian Software Group.
I don't speak for SGI.


-------------------------------------------------------
This SF.Net email is sponsored by OSTG. Have you noticed the changes on
Linux.com, ITManagersJournal and NewsForge in the past few weeks? Now,
one more big change to announce. We are now OSTG- Open Source Technology
Group. Come see the changes on the new OSTG site. www.ostg.com
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs