xfs
[Top] [All Lists]

Re: logging of metadata ops

To: Stewart Smith <stewart@xxxxxxxxxxxxxxxx>
Subject: Re: logging of metadata ops
From: Ming Zhang <mingz@xxxxxxxxxxx>
Date: Fri, 10 Mar 2006 12:01:08 -0500
Cc: Linux-xfs@xxxxxxxxxxx
In-reply-to: <1142005498.9240.63.camel@xxxxxxxxxxxxxxxxxxxxx>
References: <1142005498.9240.63.camel@xxxxxxxxxxxxxxxxxxxxx>
Reply-to: mingz@xxxxxxxxxxx
Sender: linux-xfs-bounce@xxxxxxxxxxx
not sure but see if LTT http://www.opersys.com/LTT/ can help u to log.

ming


On Fri, 2006-03-10 at 16:44 +0100, Stewart Smith wrote:
> on creat(), write(), close() of a file in a directory, at what point is
> the creation of the file and its addition to the directory flushed to
> the log on disk?
> 
> I'm trying to work out what limitations may exist in the "create one
> million tables" benchmark that we've been talking about on planetmysql
> (and now planet postgresql as well).
> 
> For MyISAM this would mean creating 3,000,000 files in a directory.
> Currently I'm seeing about 100 table creates/sec (so about 300
> files/sec) although bonnie++ shows 716 file creations per second.
> 
> So, since MySQL is also writing data to these files, could anything
> interesting be happening with the log?
> 
> It's also quite possible that MySQL is the bottleneck here and we should
> be able to do 716/3=238 table creations/sec.
> 
> Although it'd be good to know if that number could be improved upon
> simply by tuning the FS log.
> 
> (this message comes from the random useless benchmark of the week dept)


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