Title :
Multivariate statistical techniques for parallel performance prediction
Author :
Clement, Mark J. ; Quinn, Michael J.
Author_Institution :
Dept. of Comput. Sci., Brigham Young Univ., Provo, UT, USA
Abstract :
Performance prediction can play an important role in improving the efficiency of multicomputers in executing scalable parallel applications. An accurate model of program execution time must include detailed algorithmic and architectural characterizations. The exact values for critical model parameters such as message latency and cache miss penalty can often be difficult to determine. This research uses multivariate data analysis to estimate the values of these coefficients in an analytical model. Representing the coefficients as random variables with a specified mean and variance improves the utility of a performance model. Confidence intervals for predicted execution time can be generated using the standard error values for model parameters. Improvements in the model can also be made by investigating the cause of large variance values for a particular architecture
Keywords :
data analysis; parallel processing; performance evaluation; software performance evaluation; statistical analysis; algorithmic characterization; architectural characterization; cache miss penalty; confidence interval; critical model parameters; large variance values; message latency; model parameters; multicomputer efficiency; multivariate data analysis; multivariate statistical techniques; parallel performance prediction; performance model; predicted execution time; program execution time; random variables; standard error values; variance; Analytical models; Application software; Computer science; Data analysis; Debugging; Delay; Hardware; Predictive models; Random variables; Supercomputers;
Conference_Titel :
System Sciences, 1995. Proceedings of the Twenty-Eighth Hawaii International Conference on
Conference_Location :
Wailea, HI
Print_ISBN :
0-8186-6930-6
DOI :
10.1109/HICSS.1995.375512
