On Thursday 20 September 2007 12:41, Ralf Gross wrote:
> Chris Howells schrieb:
> > > Unless you have a critical problem of speed, I don't particularly
> > > recommend starting with block sizes other than the default.  It would
> > > require a lot
> >
> > Unfortunately I do have a critical speed problem with bacula. I need to
> > back up several terabytes of date in ~5 million files. With the default
> > block size Bacula is *far* slower than our current Legato Networker at
> > backing up this data to LTO 2 tape. Bacula *will not* write any faster
> > to LTO 4 any faster than ~22MB/sec under any circumstances. Legato will
> > write to tape at ~25MB/sec, or even a lot faster if it's easily
> > compressible.

For Chris: well if you consider 22MB/sec vs 25MB/sec *far* slower you have a 
different concept of the world than I do.  Yes, one might expect Bacula to be 
able to write to an LTO-4 faster than an LTO-2, but it many cases the tape 
write time is not the big bottleneck in a Bacula backup.

I believe that when people begin doing real tests with Bacula (not dd or tar), 
they will very likely find that speedups in writing to the tape don't involve 
writing enormous block sizes, but many other factors in their system -- the 
data shown below tends to confirm this ...

You might get much more satisfactory results if you start with a different 
premise from "Bacula *will not* write any faster to ... than ...".  An 
appropriate one might be "what are the things I can do to speed up the 
backups?" or "what really are the bottlenecks", or "why can someone else 
write at 72.28 M bytes/second, but my system does not?".

>
> After spooling bacula is writing to LTO with 70Mb/s here.
>
> 02-Sep 06:47 VU0EM005-sd: Despooling elapsed time = 01:43:59, Transfer
> rate = 72.28 M bytes/second
>
> This is with the default block size and only one single job.

Nice throughput Ralf :-)

>
> Ralf
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Bacula-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bacula-devel

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bacula-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-devel

Reply via email to