Analytic response time model for distributed systems

ABSTRACT

Network designers are faced with a combinatorial explosion of choices not only among various vendors' workstations, hosts and servers, but also among application distribution strategies, communication media, communication protocols, and network topologies. To study performance trade-offs among various designs, the authors have developed a generic system thruput and response time model for distributed systems. We have applied the model to actual customer networks. The modeller describes the system in a language of model statements describing operations, resources, submodels, parameters and transactions flows. Model statements are “compiled” into arrays that form the input to the analytical queueing network model. The modeling package is written in APL2 and is based on an earlier VSAPL package developed for manufacturing planning. Since the package is designed with an open set of submodel types, and since APL2 is portable across many architectures and operating systems, we can easily integrate non-APL2 code.

REFERENCES

• 1: Bux, Werner 'Local Area Subnetworks: A Performance Comparison". IEEE Transactions on Communications, Vol. Corn-29, No. 10, October 1981, p. 1465.
• 2: Bux, Werner 'Performance issues in local area networks' IBM Systems Journal, Vol. 23 No. 4, 1984, p. 351.
• 3: Leo H. Groner , Janice H. Cook, Modelling the arithmetic of statistical distributions, Proceedings of the international conference on APL, p.78-85, July 07-11, 1986, Mancheester, United Kingdom
• 4: Leo H. Groner, QMOD: A system for automatically generating and solving analytical queueing network models, Proceedings of the international conference on APL, p.125-128, October 21-23, 1981, San Francisco, California, United States
• 5: Kleinrock,L. Queueina Svstems:Vols 1 and 2. Wiley New York 1975,1976.
• 6: Edward D. Lazowska , John Zahorjan , David R. Cheriton , Willy Zwaenepoel, File access performance of diskless workstations, ACM Transactions on Computer Systems (TOCS), v.4 n.3, p.238-268, Aug. 1986 [doi>10.1145/6420.6423]
• 7: Edward D. Lazowska , John Zahorjan , G. Scott Graham , Kenneth C. Sevcik, Quantitative system performance: computer system analysis using queueing network models, Prentice-Hall, Inc., Upper Saddle River, NJ, 1984

INDEX TERMS

Primary Classification:
  D. Software
  D.3 PROGRAMMING LANGUAGES
      D.3.2 Language Classifications

          Nouns: APL

Additional Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS

  D. Software
  D.4 OPERATING SYSTEMS
      D.4.7 Organization and Design
          Subjects: Distributed systems
      D.4.8 Performance
          Subjects: Queueing theory

  J. Computer Applications
  J.1 ADMINISTRATIVE DATA PROCESSING
      Subjects: Manufacturing

General Terms:
Algorithms, Design, Languages, Performance, Theory