[Bioperl-l] BioSQL: loading large sequence records, and taxon parsing

Hilmar Lapp hlapp at gmx.net
Sat Jun 21 22:43:52 EDT 2003


On Friday, June 20, 2003, at 05:32  AM, Elia Stupka wrote:

> The only way we have found to increase the loading speed is to split 
> the dataset and fire off multiple loading scripts on different 
> machines.... but you need different machines to do that ;)
>

Actually were these single CPU machines? I've made good experience 
firing off 2 and even 3 processes in parallel from one dual-CPU 
machine. You need enough memory for this though (1 loading process will 
go up to 150-200 MB over time due to the caching that bioperl-db does).

	-hilmar
-- 
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------



More information about the Bioperl-l mailing list