A Parameterized Parallel Algorithm for Efficient Biological Sequence Comparison

In this paper we extend and improve a previous result for computing an alignment (or string editing) between two strings and , with and , which requires communication rounds and local computing time, on a distributed memory parallel computer of processors each with memory. The algorithm is based on a compromise between the workload of each processor and the number of communication rounds required, expressed by a new parameter called ! . The proposed algorithm is expressed in terms of this parameter. We implemented this algorithm to determine the best values for ! , tuned to obtain the smallest overall parallel time. We show very promising experimental results obtained on a 64-node Beowulf machine.