Predicting the CPU availability of time-shared Unix systems on the computational grid

TitlePredicting the CPU availability of time-shared Unix systems on the computational grid
Publication TypeConference Papers
Year of Publication1999
AuthorsWolski R, Spring N, Hayes J
Conference NameThe Eighth International Symposium on High Performance Distributed Computing, 1999. Proceedings
Date Published1999///
ISBN Number0-7803-5681-0
Keywordsaccuracy, Application software, Autocorrelation, Availability, Central Processing Unit, computational grid, correlation methods, CPU availability prediction, CPU resources predictability, CPU sensor, Dynamic scheduling, grid computing, Load forecasting, long-range autocorrelation dependence, medium-term forecasts, network operating systems, Network Weather Service, NWS, performance evaluation, self-similarity degree, short-term forecasts, successive CPU measurements, Time measurement, Time sharing computer systems, time-shared Unix systems, time-sharing systems, Unix, Unix load average, vmstat utility, Weather forecasting

Focuses on the problem of making short- and medium-term forecasts of CPU availability on time-shared Unix systems. We evaluate the accuracy with which availability can be measured using the Unix load average, the Unix utility “vmstat” and the Network Weather Service (NWS) CPU sensor that uses both. We also examine the autocorrelation between successive CPU measurements to determine their degree of self-similarity. While our observations show a long-range autocorrelation dependence, we demonstrate how this dependence manifests itself in the short- and medium-term predictability of the CPU resources in our study