Bacula-users

Re: [Bacula-users] LTO-4 performance?

2008-05-23 04:49:12
Subject: Re: [Bacula-users] LTO-4 performance?
From: Alan Brown <ajb2 AT mssl.ucl.ac DOT uk>
To: John Drescher <drescherjm AT gmail DOT com>
Date: Fri, 23 May 2008 09:48:49 +0100 (BST)
On Thu, 22 May 2008, John Drescher wrote:

> Yes this is very important. What I have done to solve that on my
> systems is use the hard drive or raid device directly. /dev/sda  or
> /dev/md0 work well and for my LTO2 system I get 35MB/s to 40MB/s. This
> assumes you have real data on the hard drive.

LTO2 tape throughput can easily outperform most drives on sustained
transfer rates.

LTO3 and LTO4 WILL outrun every drive currently in production. The only
way to keep up with them is some form of high speed array.

What that means is that unless you have a specialised setup, most tests
involving direct disk <-> tape throughput will be limited by disk speed.



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users