Performance evaluation of a distributed application

N. Ahituv, Y. Elovici

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Analysing the performance of an application performed on a distributed system is discussed in this paper. An analogy between a distributed system and a production process is portrayed, particularly for an application running on several computers. Consequently, theories of management of production processes are employed to help analyse and manage distributed systems, specifically, the Theory of Constraints (TOC). Using TOC combined with the cost/utilization model, which was initially developed to evaluate the utilization of a single processor and is extended here to handle a distributed system, it is demonstrated how the performance of a distributed system can be examined. The methodology presented here is based on a simple graphic display aimed to allow managers of information systems to locate constrained resources, to optimize the distribution of the computer application, and to examine and pinpoint improper imbalances and fluctuations in the system workload. The model develops into a management decision support tool that may be applied in areas such as buffer policy, assessment of protective capacity, investment in computer resources, and identification of areas for improvement.

Original languageEnglish
Pages (from-to)916-927
Number of pages12
JournalJournal of the Operational Research Society
Volume52
Issue number8
DOIs
StatePublished - 1 Jan 2001

Keywords

  • Computer performance evaluation
  • Cost/utilization approach
  • Distributed applications
  • Distributed systems
  • Parallel systems
  • Theory of constraints

ASJC Scopus subject areas

  • Management Information Systems
  • Strategy and Management
  • Management Science and Operations Research
  • Marketing

Fingerprint

Dive into the research topics of 'Performance evaluation of a distributed application'. Together they form a unique fingerprint.

Cite this