Bacula-users

Re: [Bacula-users] Bacula spool on SSD -- solid state drive performance testing?

2009-07-28 10:52:50
Subject: Re: [Bacula-users] Bacula spool on SSD -- solid state drive performance testing?
From: Alan Brown <ajb2 AT mssl.ucl.ac DOT uk>
To: Marc Cousin <mcousin AT sigma DOT fr>
Date: Tue, 28 Jul 2009 15:46:43 +0100 (BST)
On Tue, 28 Jul 2009, Marc Cousin wrote:

> At peak during the night, we have around 40-50 write streams at the same time,
> and we are despooling to 3 LTO3 and 3 LTO1.

That's larger than my installation and I haven't got budget for a
dedicated array controller. We're using software striping on our disks.

> I think the most important is : don't purchase anything without doing
> benchmarks first for this special type of activity. That's what we did : we
> wrote a small program to simulate the expected throughput on the array with
> several streams, and only bought the array when we were satisfied with its
> performance, which required a bit of tuning (choose the right filesystem, the
> right scheduler, tune read ahead...). And I think this time wasn't wasted at
> all...

You're right, it isn't.

The main criterion is keeping up with the tape drive(s). Right now we are,
but that will change with a new robot purchase and that's why I'm looking
ahead.

AB




------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users

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