xfs
[Top] [All Lists]

Re: unexpected high fragmentation, any ideas?

To: Russell Cattelan <cattelan@xxxxxxx>
Subject: Re: unexpected high fragmentation, any ideas?
From: Chris Wedgwood <cw@xxxxxxxx>
Date: Sun, 3 Apr 2005 12:03:27 -0700
Cc: Marc Lehmann <schmorp@xxxxxxxxxx>, linux-xfs@xxxxxxxxxxx
In-reply-to: <42500F94.1070603@xxxxxxxxxxx>
References: <20050403004653.GA981@xxxxxxxxxx> <20050403050542.GB5727@xxxxxxxxxxxxxxxxxxxxx> <20050403135805.GC24559@xxxxxxxxxx> <42500F94.1070603@xxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
On Sun, Apr 03, 2005 at 10:45:24AM -0500, Russell Cattelan wrote:

> I think the problem you are running into is that with a slow writing
> app pdflush is pushing pages out to disk to quickly.  A way to test
> that is to increase the pdflush interval, don't remember which proc
> value you need to change for that dirty_writback_centisecs I think.

for really slow writes i found a large biosize helped.  i've had this
in my quilt series for a long time now and use it (with appropriate
mount option):

# This is a BitKeeper generated diff -Nru style patch.
#
# ChangeSet
#   2004/09/30 18:58:10-07:00 cw@xxxxxxxx
#   xfs_mount.h:
#     [PATCH] XFS allow large biosize on mount
# 
# fs/xfs/xfs_mount.h
#   2004/09/30 18:57:59-07:00 cw@xxxxxxxx +1 -1
#   [PATCH] XFS allow large biosize on mount
# 
diff -Nru a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
--- a/fs/xfs/xfs_mount.h        2004-10-22 13:43:53 -07:00
+++ b/fs/xfs/xfs_mount.h        2004-10-22 13:43:53 -07:00
@@ -429,7 +429,7 @@
  * Max and min values for UIO and mount-option defined I/O sizes;
  * min value can't be less than a page.  Currently unused.
  */
-#define XFS_MAX_IO_LOG         16      /* 64K */
+#define XFS_MAX_IO_LOG         24      /* 16MB */
 #define XFS_MIN_IO_LOG         PAGE_SHIFT
 
 /*


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