xfs
[Top] [All Lists]

Re: XFS: Abysmal write performance because of excessive seeking (allocat

To: Emmanuel Florac <eflorac@xxxxxxxxxxxxxx>
Subject: Re: XFS: Abysmal write performance because of excessive seeking (allocation groups to blame?)
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Tue, 10 Apr 2012 07:47:36 +1000
Cc: Stefan Ring <stefanrin@xxxxxxxxx>, stan@xxxxxxxxxxxxxxxxx, Linux fs XFS <xfs@xxxxxxxxxxx>
In-reply-to: <20120409133913.2fb9d94b@xxxxxxxxxxxxxx>
References: <CAAxjCEwBMbd0x7WQmFELM8JyFu6Kv_b+KDe3XFqJE6shfSAfyQ@xxxxxxxxxxxxxx> <20350.9643.379841.771496@xxxxxxxxxxxxxxxxxx> <20350.13616.901974.523140@xxxxxxxxxxxxxxxxxx> <CAAxjCEzkemiYin4KYZX62Ei6QLUFbgZESdwS8krBy0dSqOn6aA@xxxxxxxxxxxxxx> <4F7F7C25.8040605@xxxxxxxxxxxxxxxxx> <CAAxjCEyJW1b4dbKctbrgdWjykQt8Hb4Sw1RKdys3oUsehNHCcQ@xxxxxxxxxxxxxx> <20120409001943.GI18323@dastard> <20120409133913.2fb9d94b@xxxxxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Mon, Apr 09, 2012 at 01:39:13PM +0200, Emmanuel Florac wrote:
> Le Mon, 9 Apr 2012 10:19:43 +1000 vous écriviez:
> 
> > A decent RAID controller with a BBWC and a single LUN benefits from
> > parallelism just as much as a large disk arrays do because the BBWC
> > minimises the write IO latency and the controller to do a better job
> > of scheduling its IO.
> 
> BTW recently I've found that for storage servers, noop io scheduler
> often is the best choice, I suppose precisely because it doesn't try to
> outsmart the RAID controller logic...

We've been recommending the use of the no-op (or worst case,
deadline) scheduler for XFS on hardware RAID for quite a few years.
I only test against the no-op scheduler, because I got sick of
having to track down regressions caused by "smart" CFQ heuristics....

Cheers,

Dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

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