GCG db search performance

Endre Barta barta at abc.hu
Fri Apr 18 05:17:33 EST 1997

Dear Guy,

> Testing the performance of GCG on different platforms is surely
> interesting. I decided to try ... and the difficulties began.
> First, fasta did not start. I saw that bgbench contained a line 
> #fastabench   >> LOG
> So, I removed the "outcomment" #
> then I got a
> fasta: Command not found.
> I realized that this is because fastabench starts as a UNIX subshell
> and a subshell does not inherit the aliases from its parent shell
> and by consequence the GCG aliases defined by genetics do not work.
> So, I replaced the line by 
> source fastabench   >> LOG
> then I got a
> fasta: cannot open 5000aa
> This was easily fixed, I renamed 5000aa 5000aa.pep
> then I got a 
> awk: cannot open 1e6.fasta

You certanly run fasta in batch mode by default, and the output is of course 
not in place at that stage.

Try edit bgbench, and use full path for fasta, findpatterns and profilesearch
e.g. $GCGCOREROOT/gcgbin/execute/fasta, or turn off running these programs
in batch mode.

Also, when using Solaris as a normal user, the psrinfo -v command in gcgbench
should be replaced to /usr/sbin/psrinfo -v.


	Endre Barta

*									  *
*	Endre Barta, Ph.D.		Agricultural Biotechnology Center *
*	Hungarian EMBnet Node		Computer Department		  *
*	e-mail: barta at abc.hu		Godollo, P.O.Box: 411	       	  *
*	Tel: (36)-28-430 127		Szent-Gyorgyi Albert ut 4.	  *
*	FAX: (36)-28-420 096		H-2101 Hungary			  *
*               <a href="http://hubi.abc.hu/~barta"></a>                  *
*   								          *

More information about the Info-gcg mailing list

Send comments to us at biosci-help [At] net.bio.net