A scalability metric for algorithm-machine on NOW and MPP

Along with the rapid development of parallel processing technology and the popularity of NOW, the scalability of parallel algorithm-machine combinations, which measures the capacity of a parallel algorithm to effectively utilize an increasing number of processors, becomes more and more important. We present a new metric, called the time scale metric, to measure and evaluate the scalability of parallel algorithms and machines, and extend it to fit the characteristics of NOW. The experimental results show that the time scale metric is a practical and accurate method to evaluate the scalability of parallel algorithms on MPP and NOW.