AMBER Archive (2002)

Subject: Re: Scaling of Sander on Linux Clusters (fwd)

From: Sanjeev B.S. (sanjeev_at_mbu.iisc.ernet.in)
Date: Tue Aug 27 2002 - 12:59:51 CDT

Available on the amber site ... I agree, but i have never been able to
reproduce their figures with my own systems !
Nevertheless, i'm thinking in terms of dozens of CPU (16, 32, 64, 128), so
even if it scales not too badly for 2-4 processors, i need speedup until
there. Without these kind of speedup, clusters are just an utopy.
Of course, you can benefit from the powerfull of one dual-node ...

Stef

One important thing. One can try not just with mpich but also LAM. It
seems LAM was better for clusters, and mpich for shared memory types. Not
sure about the latest versions, but that was my experience a year ago.