On an efficient parallelization of exhaustive sequence comparison algorithms on message passing architectures
1Department of Computer Architecture, University of Malaga, CSIC, Campus de la Universidad Autónoma 28049 Madrid
2Centro Nacional de Biotecnología, CSIC, Campus de la Universidad Autóonoma 28049 Madrid
3To whom reprints request should be sent
We present a new parallel computing approach to the case of exhaustive sequential sequence comparison algorithms on message-passing architectures. In this context a modification of guided self-scheduling as well as efficient buffering strategies are presented. We discuss two specific implementations, one on the Paramid parallel computer, and the other on a cluster of workstations running PVM. In both cases the parallel performance is higher than with any other method presented so far. The code is public domain and can be obtained by anonymous ftp at ftp.cnb.uam.es.