RE: ORION results

From: Luca Canali <>
Date: Fri, 29 Jan 2010 09:30:16 +0100
Message-ID: <>


See below the link for a 'introduction to ORION' wiki page that I made so time ago for sharing with colleagues in the High Energy Phisycs DB community. It is by no means exhaustive, but it may be of help for you to start with.


From: [] On Behalf Of Michael Schmitt Sent: 28 January 2010 23:22
To: ''
Subject: ORION results

Hi All,

I am messing around with ORION on a couple of different machines and I am seeing results from a couple of servers that make me think something is goofy. The files are sitting on a SAN with 15K disks in raid 5, and the results are at the bottom of the email. I was wondering if I was most likely seeing results of caching on the server/SAN somehow (RHEL5 blades)? When I run the same test on the similar hardware, but a different set of disks on the SAN, I get numbers like IOPS 281 , MBPS 378, latency 4.44. Can anyone help explain the differences?

ORION VERSION Commandline:
-run simple -testname linuxprd2 -num_disks 8 (I just used this number because it was the same in the doc)

Maximum Large MBPS=2257.83 _at_ Small=0 and Large=3
Maximum Small IOPS=172705 _at_ Small=36 and Large=0
Minimum Small Latency=0.01 _at_ Small=1 and Large=0


-- Received on Fri Jan 29 2010 - 02:30:16 CST

Original text of this message