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

Fwd: Inconsistency in bonnie++ results for repeated runs



If anyone has any ideas on the possible cause of this then please let me know 
and CC Judith.

----------  Forwarded Message  ----------

Subject: Inconsistency in bonnie++ results for repeated runs
Date: Fri, 6 Feb 2004 04:53
From: Judith Lebzelter <judith@osdl.org>
To: <russell@coker.com.au>

Hello;

I found that my bonnie++ results are varying more than I would have
thought that they should, on this run for example:
    http://khack.osdl.org/stp/286377/logs/run-log.txt

Sequential Input for the third set of runs for both the Per Chr
 and Block measurements dropped off too much:

Version  @version@      ------Sequential Output------ --Sequential Input-
 --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
 Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec
 %CP stp1-000         1G 11676  99 39927  22 16889   8 10566  89 38059  12
 201.2   0 stp1-000         1G 11673  99 39130  23 16927  11 10564  89 38071 
 11 180.2   0 stp1-000         1G 11214  96 31424  17  4086   2  7721  65
 14441   5 137.2   0

Rewrites dropped as well.

Is there something I have not configured right or an option I am
missing?  For most of our runs the three repeated data points do have a
reasonable deviation, but occasionally, one of the runs seems too far off
the other two.

This is the command being run for the above results:
    /usr/bin/time bonnie++ -x 3 -u 0 -n1

On a 1CPU Host and on a single IDE disk/500M RAM, but I have found similar
issues on the 2CPU hosts.

We are using Bonnie++ Version: 1.02.

Is there a mailing list that would be more appropriate than direct mail?
If so please forward it.

Thanks;

Judith Lebzelter
OSDL


_______________________________________________
Testdev mailing list
Testdev@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/testdev

-------------------------------------------------------

-- 
http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/    Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page



Reply to: