Optimal scheduling policies for mixed scalar-vector multiprocessor supercomputers

Authors:
Y. Liu;O. Johnson
Affiliations:
Ph.D. Candidate, Department of Computer Science, University of Houston;Professor of Computer Science, University of Houston and Director, CSARC, Houston Area Research Center
Venue:
Proceedings of the 1988 ACM/IEEE conference on Supercomputing
Year:
1988

Citing 3
Cited 0

Effects of job loading policies for multiprogramming systems in processing a job stream

ACM Transactions on Computer Systems (TOCS)
A Finite-Source Queue with Different Customers

Journal of the ACM (JACM)
Optimality of scheduling policy for processing a job stream

ACM Transactions on Computer Systems (TOCS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

A queueing network model is developed for multiprocessing supercomputer systems implemented on two heterogeneous processors which are thought of as a scalar unit and a vector unit which work concurrently on a single process. The model is analysed theoretically for a representative case of model job mixes. It is found that the utilization of one of the two central service stations in the model can only be maximized at the expense of minimizing the utilization of the other central service station. The method of analysis adopted is the technique of variable aggregation based on nearly complete decomposability.