Implementing parallel hmm-pfam on the EARTH multithreaded architecture

Abstract

Hmmpfam is a widely used computation-intensive bioinformatics software for sequence classification. This poster describes a new parallel implementation of hmmpfam on EARTH, which is an event-driven fine-grain multithreaded programming execution model. The comparison results of the original PVM implementation and our implementation shows notable improvements on absolute speedup and scalability. On a cluster of 128 dual-CPU nodes, the execution time of a representative testbench is reduced from 15.9 hours to 4.3 minutes.