A historical application profiler for use by parallel schedulers

grantor: University of TorontoScheduling Algorithms that use application and system knowledge have been shown to be more effective at scheduling parallel jobs on a multiprocessor than algorithms that do not. This thesis focuses on obtaining such information for use by a scheduler in a network of workstations environment. The log files from three parallel systems are examined to determine the best way of categorizing parallel jobs for storage in a job database and the job information that would be useful to a scheduler. A Historical Profiler is proposed that stores information about programs and users, and manipulates this information to provide schedulers with execution time estimates. Several preemptive and non-preemptive versions of the FCFS, EASY and Least Work First scheduling algorithms are compared to evaluate the utility of the profiler. It is found that both preemption and the use of application execution time estimates obtained from the Historical Profiler lead to improved performance.M.Sc