Shared Virtual Memory and Generalized Speedup

Generalized speedup is de ned as parallel speed over sequential speed. In this paper the generalized speedup and its relation with other existing performance metrics, such as traditional speedup, e ciency, scalability, etc., are carefully studied. In terms of the introduced asymptotic speed, we show that the di erence between the generalized speedup and the traditional speedup lies in the de nition of the e ciency of uniprocessor processing, which is a very important issue in shared virtual memory machines. A scienti c application has been implemented on a KSR-1 parallel computer. Experimental and theoretical results show that the generalized speedup is distinct from the traditional speedup and provides a more reasonable measurement. In the study of di erent speedups, various causes of superlinear speedup are also presented. This research was supported by the National Aeronautics and Space Administration under NASA contract NAS119480 while the rst author was in residence at the Institute for Computer Applications in Science and Engineering (ICASE), NASA Langley Research Center, Hampton, VA 23681-0001.