linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Howells <dhowells@redhat.com>
Cc: Trond.Myklebust@netapp.com, nfsv4@linux-nfs.org,
	linux-kernel@vger.kernel.org, dhowells@redhat.com,
	linux-fsdevel@vger.kernel.org,
	linux-security-module@vger.kernel.org, selinux@tycho.nsa.gov,
	casey@schaufler-ca.com, Daniel Phillips <phillips@phunq.net>
Subject: Re: [PATCH 00/37] Permit filesystem local caching
Date: Fri, 22 Feb 2008 16:14:35 +0000	[thread overview]
Message-ID: <19020.1203696875@redhat.com> (raw)
In-Reply-To: <17916.1203636833@redhat.com>

David Howells <dhowells@redhat.com> wrote:

> > > Have you got before/after benchmark results?
> > 
> > See attached.
> 
> Attached here are results using BTRFS (patched so that it'll work at all)
> rather than Ext3 on the client on the partition backing the cache.

And here are XFS results.

Tuning XFS makes a *really* big difference for the lots of small/medium files
being tarred case.  However, in general BTRFS is much better.

David
---


=========================
FEW BIG FILES TEST ON XFS
=========================

Completely cold caches:

	[root@andromeda ~]# time cat /warthog/bigfile >/dev/null
	real    0m2.286s
	user    0m0.000s
	sys     0m1.828s
	[root@andromeda ~]# time cat /warthog/biggerfile >/dev/null
	real    0m4.228s
	user    0m0.000s
	sys     0m1.360s

Warm NFS pagecache:

	[root@andromeda ~]# time cat /warthog/bigfile >/dev/null
	real    0m0.058s
	user    0m0.000s
	sys     0m0.060s
	[root@andromeda ~]# time cat /warthog/biggerfile >/dev/null
	real    0m0.122s
	user    0m0.000s
	sys     0m0.120s

Warm XFS pagecache, cold NFS pagecache:

	[root@andromeda ~]# time cat /warthog/bigfile >/dev/null
	real    0m0.181s
	user    0m0.000s
	sys     0m0.180s
	[root@andromeda ~]# time cat /warthog/biggerfile >/dev/null
	real    0m1.034s
	user    0m0.000s
	sys     0m0.404s

Warm on-disk cache, cold pagecaches:

	[root@andromeda ~]# time cat /warthog/bigfile >/dev/null
	real    0m1.540s
	user    0m0.000s
	sys     0m0.256s
	[root@andromeda ~]# time cat /warthog/biggerfile >/dev/null
	real    0m3.003s
	user    0m0.000s
	sys     0m0.532s


==========================================
MANY SMALL/MEDIUM FILE READING TEST ON XFS
==========================================

Completely cold caches:

	[root@andromeda ~]# time tar cf - /warthog/aaa >/dev/zero
	real    4m56.827s
	user    0m0.180s
	sys     0m6.668s

Warm NFS pagecache:

	[root@andromeda ~]# time tar cf - /warthog/aaa >/dev/zero
	real    0m15.084s
	user    0m0.212s
	sys     0m5.008s

Warm XFS pagecache, cold NFS pagecache:

	[root@andromeda ~]# time tar cf - /warthog/aaa >/dev/zero
	real    0m13.547s
	user    0m0.220s
	sys     0m5.652s

Warm on-disk cache, cold pagecaches:

	[root@andromeda ~]# time tar cf - /warthog/aaa >/dev/zero
	real    4m36.316s
	user    0m0.148s
	sys     0m4.440s


=======================================================
MANY SMALL/MEDIUM FILE READING TEST ON AN OPTIMISED XFS
=======================================================

mkfs.xfs -d agcount=4 -l size=128m,version=2 /dev/sda6


Completely cold caches:

	[root@andromeda ~]# time tar cf - /warthog/aaa >/dev/zero
	real    3m44.033s
	user    0m0.248s
	sys     0m6.632s

Warm on-disk cache, cold pagecaches:

	[root@andromeda ~]# time tar cf - /warthog/aaa >/dev/zero
	real    3m8.582s
	user    0m0.108s
	sys     0m3.420s

  parent reply	other threads:[~2008-02-22 16:14 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-20 16:05 [PATCH 00/37] Permit filesystem local caching David Howells
2008-02-20 16:06 ` [PATCH 01/37] KEYS: Increase the payload size when instantiating a key David Howells
2008-02-20 16:06 ` [PATCH 02/37] KEYS: Check starting keyring as part of search David Howells
2008-02-20 16:06 ` [PATCH 03/37] KEYS: Allow the callout data to be passed as a blob rather than a string David Howells
2008-02-20 16:06 ` [PATCH 04/37] KEYS: Add keyctl function to get a security label David Howells
2008-02-20 16:06 ` [PATCH 05/37] Security: Change current->fs[ug]id to current_fs[ug]id() David Howells
2008-02-20 16:06 ` [PATCH 06/37] Security: Separate task security context from task_struct David Howells
2008-02-22  4:47   ` Casey Schaufler
2008-02-20 16:06 ` [PATCH 07/37] Security: De-embed task security record from task and use refcounting David Howells
2008-02-22  4:57   ` Casey Schaufler
2008-02-20 16:06 ` [PATCH 08/37] Security: Add a kernel_service object class to SELinux David Howells
2008-02-20 16:06 ` [PATCH 09/37] Security: Allow kernel services to override LSM settings for task actions David Howells
2008-02-22  5:06   ` Casey Schaufler
2008-02-22 13:06     ` David Howells
2008-02-20 16:06 ` [PATCH 10/37] Security: Make NFSD work with detached security David Howells
2008-02-20 16:06 ` [PATCH 11/37] FS-Cache: Release page->private after failed readahead David Howells
2008-02-20 16:07 ` [PATCH 12/37] FS-Cache: Recruit a couple of page flags for cache management David Howells
2008-02-20 16:07 ` [PATCH 13/37] FS-Cache: Provide an add_wait_queue_tail() function David Howells
2008-02-20 16:07 ` [PATCH 14/37] FS-Cache: Generic filesystem caching facility David Howells
2008-02-20 16:07 ` [PATCH 15/37] CacheFiles: Add missing copy_page export for ia64 David Howells
2008-02-20 16:07 ` [PATCH 16/37] CacheFiles: Be consistent about the use of mapping vs file->f_mapping in Ext3 David Howells
2008-02-20 16:07 ` [PATCH 17/37] CacheFiles: Add a hook to write a single page of data to an inode David Howells
2008-02-20 16:07 ` [PATCH 18/37] CacheFiles: Permit the page lock state to be monitored David Howells
2008-02-20 16:07 ` [PATCH 19/37] CacheFiles: Export things for CacheFiles David Howells
2008-02-20 16:07 ` [PATCH 20/37] CacheFiles: A cache that backs onto a mounted filesystem David Howells
2008-02-20 16:07 ` [PATCH 21/37] NFS: Add comment banners to some NFS functions David Howells
2008-02-20 16:07 ` [PATCH 22/37] NFS: Add FS-Cache option bit and debug bit David Howells
2008-02-20 16:08 ` [PATCH 23/37] NFS: Permit local filesystem caching to be enabled for NFS David Howells
2008-02-20 16:08 ` [PATCH 24/37] NFS: Register NFS for caching and retrieve the top-level index David Howells
2008-02-20 16:08 ` [PATCH 25/37] NFS: Define and create server-level objects David Howells
2008-02-20 16:08 ` [PATCH 26/37] NFS: Define and create superblock-level objects David Howells
2008-02-20 16:08 ` [PATCH 27/37] NFS: Define and create inode-level cache objects David Howells
2008-02-20 16:08 ` [PATCH 28/37] NFS: Use local disk inode cache David Howells
2008-02-20 16:08 ` [PATCH 29/37] NFS: Invalidate FsCache page flags when cache removed David Howells
2008-02-20 16:08 ` [PATCH 30/37] NFS: Add some new I/O event counters for FS-Cache events David Howells
2008-02-20 16:08 ` [PATCH 31/37] NFS: FS-Cache page management David Howells
2008-02-20 16:08 ` [PATCH 32/37] NFS: Add read context retention for FS-Cache to call back with David Howells
2008-02-20 16:08 ` [PATCH 33/37] NFS: nfs_readpage_async() needs to be accessible as a fallback for local caching David Howells
2008-02-20 16:09 ` [PATCH 34/37] NFS: Read pages from FS-Cache into an NFS inode David Howells
2008-02-20 16:09 ` [PATCH 35/37] NFS: Store pages from an NFS inode into a local cache David Howells
2008-02-20 16:09 ` [PATCH 36/37] NFS: Display local caching state David Howells
2008-02-20 16:09 ` [PATCH 37/37] NFS: Add mount options to enable local caching on NFS David Howells
2008-02-20 19:58 ` [PATCH 00/37] Permit filesystem local caching Serge E. Hallyn
2008-02-20 20:11 ` David Howells
2008-02-21  3:07 ` Daniel Phillips
2008-02-21 12:31 ` David Howells
2008-02-21 14:55 ` David Howells
2008-02-21 15:17   ` Kevin Coffman
2008-02-21 22:44   ` Daniel Phillips
2008-02-21 22:52     ` Muntz, Daniel
2008-02-22  0:07   ` David Howells
2008-02-22  0:57     ` Daniel Phillips
2008-02-22 12:48     ` David Howells
2008-02-22 22:25       ` Daniel Phillips
2008-02-23  1:22       ` David Howells
2008-02-21 23:33 ` David Howells
2008-02-22 13:52   ` Chris Mason
2008-02-22 16:12   ` David Howells
2008-02-22 16:47   ` David Howells
2008-02-25 23:19   ` David Howells
2008-02-26  0:43     ` Daniel Phillips
2008-02-26  2:00     ` David Howells
2008-02-26 10:26       ` Daniel Phillips
2008-02-26 14:33       ` David Howells
2008-02-26 19:43         ` Daniel Phillips
2008-02-26 21:09         ` David Howells
2008-02-22 16:14 ` David Howells [this message]
  -- strict thread matches above, loose matches on Subject: below --
2008-02-22 16:01 Rick Macklem
2008-02-08 16:51 David Howells

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=19020.1203696875@redhat.com \
    --to=dhowells@redhat.com \
    --cc=Trond.Myklebust@netapp.com \
    --cc=casey@schaufler-ca.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=nfsv4@linux-nfs.org \
    --cc=phillips@phunq.net \
    --cc=selinux@tycho.nsa.gov \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).