[nfsv4] Performance drop in iozone for reading large files

[Bryce Harrington]

On Fri, Aug 25, 2006 at 05:23:37PM -0400, Trond Myklebust wrote:
> On Fri, 2006-08-25 at 14:01 -0700, Bryce Harrington wrote:
> > There is a performance issue in READ operations we've noticed in iozone
> > runs for a while now.  If you look at this plot, of NFSv4 with 1G mem:
> > 
> >    http://crucible.osdl.org/runs/1604/test_output/iozone.sys.log.png
> > 
> > You can see that we're getting fairly steady performance up until the
> > 1048576kb case, where read performance drops to about a third.
> 
> Is the readahead algorithm working correctly? Could you please try
> applying the following patch, and then check your syslog for warnings?

Here it is:

http://crucible.osdl.org/runs/1617/nfs03.console.log

Looks like it switches between 12K, 240K, and 60K.  Does that sound
right?

Thanks,
Bryce

> Date: Fri, 25 Aug 2006 17:23:24 -0400
> From: Trond Myklebust <Trond.Myklebust at netapp.com>
> Subject: No Subject
> 
> 
> ---
> 
>  mm/filemap.c |    2 ++
>  1 files changed, 2 insertions(+), 0 deletions(-)
> 
> diff --git a/mm/filemap.c b/mm/filemap.c
> index b9a60c4..d087fc3 100644
> --- a/mm/filemap.c
> +++ b/mm/filemap.c
> @@ -849,6 +849,8 @@ static void shrink_readahead_size_eio(st
>  		return;
>  
>  	ra->ra_pages /= 4;
> +	printk(KERN_WARNING "Reducing readahead size to %luK\n",
> +			ra->ra_pages << (PAGE_CACHE_SHIFT - 10));
>  }
>  
>  /**