Abstract: | The paper describes experiences gained during the process of Implementing a standard serial benchmark (SLALOM) on a distributed computing system (Pleiades running ESP/C++). Based on a real problem In computer graphics, SLALOM Is a timed benchmark that Includes problem set-up and I/O costs. The purpose of our experiments has been to maximize the speedup of a distributed implementation of SLALOM when compared to Its serial implementation. As a result of these experiments, we have developed several guidelines for adapting algorithms for distributed systems. |