Re: Effects of setting linux block device readahead size

Sorry, I forgot to mention the Linux kernel version I'm using, etc:

2.6.18-92.1.10.el5 #1 SMP x86_64
CentOS 5.2.

The "adaptive" read-ahead, as well as other enhancements in the kernel, are
taking place or coming soon in the most recent stuff. Some distributions
offer the adaptive read-ahead as an add-on (Debian, for example). This is
an area where much can be improved in Linux http://kerneltrap.org/node/6642
http://kernelnewbies.org/Linux_2_6_23#head-102af265937262a7a21766ae58fddc1a29a5d8d7

I obviously did not test how the new read-ahead stuff impacts these sorts of
tests.

On Thu, Sep 11, 2008 at 12:07 PM, Scott Carey <scott(at)richrelevance(dot)com>wrote:

> Hmm, I would expect this tunable to potentially be rather file system
> dependent, and potentially raid controller dependant. The test was using
> ext2, perhaps the others automatically prefetch or read ahead? Does it
> vary by RAID controller?
>
> Well I went and found out, using ext3 and xfs. I have about 120+ data
> points but here are a few interesting ones before I compile the rest and
> answer a few other questions of my own.
>
> 1: readahead does not affect "pure" random I/O -- there seems to be a
> heuristic trigger -- a single process or file probably has to request a
> sequence of linear I/O of some size to trigger it. I set it to over 64MB of
> read-ahead and random iops remained the same to prove this.
> 2: File system matters more than you would expect. XFS sequential
> transfers when readahead was tuned had TWICE the sequential throughput of
> ext3, both for a single reader and 8 concurrent readers on 8 different
> files.
> 3: The RAID controller and its configuration make a pretty significant
> difference as well.
>
> Hardware:
> 12 7200RPM SATA (Seagate) in raid 10 on 3Ware 9650 (only ext3)
> 12 7200RPM SATA ('nearline SAS' : Seagate ES.2) on PERC 6 in raid 10 (ext3,
> xfs)
> I also have some results with PERC raid 10 with 4x 15K SAS, not reporting
> in this message though
>
> . . . {snip}