Interfacing MATLAB with a parallel virtual processor for matrix algorithms

This paper describes the results of a project to interface MATLAB with a pa rallel virtual processor (PVP) that allows execution of matrix operations i n MATLAB on a set of computers connected by a network. The software, a conn ection-oriented BSD socket-based client-server model, automatically partiti ons a MATLAB problem and delegates work to server processes running on sepa rate remote machines. Experimental data on the matrix multiply operation sh ows that the speed improvement of the parallel implementation over the sing le-processor MATLAB algorithm depends on the size of the matrices, the numb er of processes, the speed of the processors, and the speed of the network connection. In particular, the advantage of doing matrix multiply in parall el increases as the size of the matrices increase. A speedup of 2.95 times was achieved in multiplying 2048 by 2048 square matrices using 15 workstati ons. The algorithm was also implemented on a network of four PC's, which wa s up to 2.5 times as fast as four workstations. The study also showed that there is an optimal number of processes for a particular problem, and so us ing more processes is not always faster. (C) 2001 Elsevier Science Inc. All rights reserved.