A Processor Utilization Model for a Multiprocessor Computer System

Richard E. Nance and U. Narayan Bhat
Additional contact information
Richard E. Nance: Virginia Polytechnic Institute and State University, Blacksburg, Virginia
U. Narayan Bhat: Southern Methodist University, Dallas, Texas

Operations Research, 1978, vol. 26, issue 5, 881-895

Abstract: Using a G / M / s / N queuing system model, we develop a processor utilization model for a simplified multiprocessor computer system. Jobs are assumed to arrive according to a general input process, and each job is assigned randomly to an available processor. A finite capacity input buffer is used if no processor is available. The mathematical model is based on the busy period analysis, and two utilization measures are derived: processor utilization (the fraction of processor occupation time during a busy cycle) and system utilization (the fraction of actual utilization time for all processors). A computational model for these utilization measures is developed through an imbedded Markov chain approach. Experimentation with the computational model reveals the sensitivity of the model to variability in the arrival process. Comparison of 2-processor and 4-processor systems from the operator perspective indicates a qualified preference for the behavior of the 2-processor system. This preference must be carefully interpreted since processor costs, the increase in overhead with an increase in processors, and behavioral variables reflecting the user perspective are excluded.

Date: 1978
References: Add references at CitEc
Citations:

Downloads: (external link)
http://dx.doi.org/10.1287/opre.26.5.881 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:inm:oropre:v:26:y:1978:i:5:p:881-895

Access Statistics for this article

More articles in Operations Research from INFORMS Contact information at EDIRC.
Bibliographic data for series maintained by Chris Asher ().