Skip to Main Content
Contemporary operating systems provide system resource management mechanisms that permit the installation manager to define his system performance objectives. The installation manager thinks of these objectives in terms of response time or throughput, but in relaity they have to be translated into system internal terms such as time slice deadlines or service units. This is not an easy translation to make, since it requires experimentation with different system resource management policies. Identifying the appropriate system resouce management policy require methodological approach. Such a methodology consists, first, of performance measurements, workload characterization, and performance evaluation, followed by a series of performance tuning experiments. We found that by using such a methodology, we were able to characterize the workload and the behavior of the IBM Research Division's TSS/360 system, then verify that our proposed scheduling change positively improved system responsiveness.
Note: The Institute of Electrical and Electronics Engineers, Incorporated is distributing this Article with permission of the International Business Machines Corporation (IBM) who is the exclusive owner. The recipient of this Article may not assign, sublicense, lease, rent or otherwise transfer, reproduce, prepare derivative works, publicly display or perform, or distribute the Article.