[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] tap:aio Performance

On Fri, Feb 16, 2007 at 02:18:49PM +1300, Peter wrote:
> Hi.
> We have hit bugs with 3.0.3 with blktap, so can't use it there.  (I saw 
> some fixes on the devel list that I think address these problems in 3.0.4).
> 3.0.4 has crashed within a hour to a day after enabling any loop back 
> mounted file systems.  So we can't use loop back mounted devices on 
> 3.0.4.  3.0.3 is fine.  We'd reported this via bug tracker but we don't 
> have a reproducible test case for it.
> The performance on 3.0.4 (with sparse blktap files) is a lot slower 
> (unusably so) compared to 3.0.3 (with sparse loopback mounted files).

As I said in my previous mail - it is meaningless to compare performance
of  tap:aio with file: One writes data to disk when you ask it to (blktap)
the other postpones the writes indefinitely (file).  Of course the one which 
doesn't actually write your data to disk is faster - but that's not useful 
because upon crash you can say goodbye to your data.

It is doubly meaningless to compare the performance when using sparse
files. Every write to a sparse file which requires extra blocks to be
allocated will hit the journal - this destroys performance. If you want
to look at performance, use non-sparse files with blktap, and compare 
against phy:    Any comparison against file: is completely bogus.

|=- Red Hat, Engineering, Emerging Technologies, Boston.  +1 978 392 2496 -=|
|=-           Perl modules: http://search.cpan.org/~danberr/              -=|
|=-               Projects: http://freshmeat.net/~danielpb/               -=|
|=-  GnuPG: 7D3B9505   F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505  -=| 

Xen-users mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.