xfs
[Top] [All Lists]

Re: 2.6.11-rc3: 80-95% systime when serving files via NFS

To: Dave Chinner <dgc@xxxxxxx>
Subject: Re: 2.6.11-rc3: 80-95% systime when serving files via NFS
From: Sonny Rao <sonny@xxxxxxxxxxx>
Date: Tue, 8 Feb 2005 11:54:56 -0500
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <20050208153427.G147390@melbourne.sgi.com>
References: <222BE5975A4813449559163F8F8CF5030C22D2@cohsrv1.cohaesio.com> <20050208153427.G147390@melbourne.sgi.com>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4.2.1i
On Tue, Feb 08, 2005 at 03:34:27PM +1100, Dave Chinner wrote:
<snip> 
> Seeing as you are running 2.6.11-rc3, you're in luck. Nathan's
> inode hash tweaks were merged into -rc3, so you can increase the
> size of the inode hash with a mount option. The mount option
> is "ihashsize", and it specifies the width of the hash table.
> i.e. `mount -o ihashsize=xxxxx /dev/sdb1 /mntpt`
> 
> I'm not sure what the default your filesystem will be using, but
> a maximum of 16*PAGE_SIZE (=64k chains on i386/x86-64) applies to the
> default setting.
> 
> If you specify a size this maximum does not apply, but if you
> specify a value too large memory allocation for the hash will
> silently fail and the size will be halved repeatedly until the
> allocation succeeds.
> 


Hmm, I've seen this before too running something which resembles
SPEC-SFS where we have millions of inodes in cache.

I'm guessing that increasing the hash-size will help us here as well?

Sonny

Oprofile and slab Data from kernel 2.6.9
Machine is a 4-way Power5 box.

Sorry about the long lines.

Here's a snippet from Oprofile:

CPU: ppc64 POWER5, speed 1904 MHz (estimated)
Counted CYCLES events (Processor cycles) with a unit mask of 0x00 (No
unit mask) count 100000
samples  %        image name               app name
symbol name
27374923 30.6599  vmlinux-autobench-2.6.9-autokern1 
vmlinux-autobench-2.6.9-autokern1 .dedicated_idle
8371486   9.3761  xfs.ko                   xfs                      .xfs_iget

Snippet from slabinfo:

xfs_acl                0      0    304   13    1 : tunables   54   27    8 : 
slabdata      0      0      0
xfs_chashlist     766053 766640     32  112    1 : tunables  120   60    8 : 
slabdata   6845   6845      3
xfs_ili           779799 4449280    192   20    1 : tunables  120   60    8 : 
slabdata 222464 222464    480
xfs_ifork              0      0     64   59    1 : tunables  120   60    8 : 
slabdata      0      0      0
xfs_efi_item           0      0    352   11    1 : tunables   54   27    8 : 
slabdata      0      0      0
xfs_efd_item           0      0    360   11    1 : tunables   54   27    8 : 
slabdata      0      0      0
xfs_buf_item        3682   7350    184   21    1 : tunables  120   60    8 : 
slabdata    350    350      0
xfs_dabuf              3    576     24  144    1 : tunables  120   60    8 : 
slabdata      2      4      0
xfs_da_state           5    152    488    8    1 : tunables   54   27    8 : 
slabdata      5     19      2
xfs_trans             88    468    872    9    2 : tunables   54   27    8 : 
slabdata     52     52     68
xfs_inode         5435445 6247990    528    7    1 : tunables   54   27    8 : 
slabdata 892570 892570    216
xfs_btree_cur          0    100    192   20    1 : tunables  120   60    8 : 
slabdata      0      5      0
xfs_bmap_free_item      0      0     24  144    1 : tunables  120   60    8 : 
slabdata      0      0      0
xfs_buf_t           5210   6860    512    7    1 : tunables   54   27    8 : 
slabdata    980    980    216
linvfs_icache     5429958 6130187    584    7    1 : tunables   54   27    8 : 
slabdata 875741 875741      0


<Prev in Thread] Current Thread [Next in Thread>