Hello!
Some time ago I posted a short questionnaire regarding GCG performances for
databases searches. A few of you answered back and I wish to thank them all
for their time. Unfortunately, answers have been too sparsed to be really
informative.
Let's carry on though with stage II of the procedure. I've designed a csh
script which will measure the time used for FastA, Findpatterns and
Profilesearch searches on a defined data set. (sorry for VMS people, I'd
love to provide a similar bench for you as well, but I would need some help
to do so... anyone ready to help?)
In order to make sure we compare searches against exactly the same data
set, sequence data are provided. So we don't need to worry about what
database are available at your site (if there are any!). Furthermore, as I
didn't want to overflow your disks, I kept the data size to a minimum
(32 Kb). The searches are actually performed against redundant FOSN.
This script and required data can be downloaded from our FTP site at
ftp://ftp-igbmc.u-strasbg.fr/pub/GCGBench/gcgbench.tar
Once you've untarred the archive file, all you have to do is edit the path
for gcg startup scripts in the bgbench file, run the gcgbench script,
answer the four questions, and let it go... The bench will be run as a
background job and results will be automatically sent to my e-mail address.
Though the bench might be running for a few hours (about 6h00 on our Sparc
server 1000E, 4 x 85 Mhz), it should not take more than a few minutes of
your time.
Results will be shared with the whole GCG community.
Thanks for your help,
Fred
Frederic PLEWNIAK
Bioinformatics
IGBMC - Strasbourg - France
