xfs
[Top] [All Lists]

Re: Benchmark

To: Florin Andrei <florin@xxxxxxx>
Subject: Re: Benchmark
From: Steve Lord <lord@xxxxxxx>
Date: Fri, 11 May 2001 13:38:30 -0500
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: Message from Florin Andrei <florin@xxxxxxx> of "11 May 2001 11:22:57 PDT." <989605377.13414.3.camel@xxxxxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
> On 11 May 2001 18:21:31 +0800, Federico Sevilla III wrote:
> > 
> > I'm not an expert on XFS, but I do recall someone on the list posting a
> > message recently saying that it had something to do with the fact that XFS
> > does synchronous file deletions, hence the speed issue.
> 
> Is it possible to mount an XFS partition so that deletes are NOT
> synchronous?

No, but you can make all the other transactions synchronous too so that the
deletes do not look as bad when compared to other operations ;-) This is
the wsync option.

The only way to experiment with this is to start commenting out
xfs_trans_set_sync calls in xfs. Can people either have a little
patience on this, or try stuff themselves, this is not going to
get changed from within SGI anytime soon, as I said the other day,
the todo list from this end is LONG.

If you want to experiment with XFS with a low latency log then try this
trick:

        o create a loop device on a file
        o make an xfs filesystem with an external log on the loop device

The writes to the log will still happen, but the disk will not be in the
picture in a synchronous manner since the loop device just copies into
pages in the page cache and flushing happens asynchronously.

> 
> -- 
> Florin Andrei
> 


Steve



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