[Bioclusters] Bioinformatics Benchmark System version 3 release candidate 1 ready

Joe Landman bioclusters@bioinformatics.org
Mon, 12 Jul 2004 17:47:41 -0400


Hi Doug:

  All of them of course :).  More exactly, could you use the optimized 
HMMer and BLAST code from http://downloads.scalableinformatics.com/, and 
compare to locally compiled versions, and to "vendor" provided versions 
(e.g. from Professor Eddy's and the NCBI site respectively) ?  

  Also, could you let me know how hard/easy it is to tweak tests to your 
liking?  We are working on documentation, and have feedback from at 
least one person who doesn't like XML.  The idea is that good 
constructive criticism (and contribution) will help improve the tool, so 
if you have an idea on how to improve the input formatting, this would 
help everyone.  Even better is if you have some code to implement the 
idea :)

  The HMMer tests need relabeling (my bad, I forgot to do this), and the 
long one takes about 18000 seconds on a 2.6 GHz P4 Xeon.  Somewhat less 
on the Opteron.  Would like to see it on the G5 with and without 
vectorization :)

Joe

Doug Shubert wrote:

> ok the latest  'bbsv3rc1.tar.gz' installed without an error.
> running the 'bbsv1.xml' benchmark now and will post the results.
> which file(s) are of interest to the list?
>
> Thanks
> Doug
>
-- 
Joseph Landman, Ph.D
Scalable Informatics LLC,
email: landman@scalableinformatics.com
web  : http://scalableinformatics.com
phone: +1 734 612 4615