This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Software Congestion, Mobile Servers, and the Hyperbolic Model
August 1989 (vol. 15 no. 8)
pp. 947-962

The phenomenon of software congestion is examined. The term refers to situations in which the performance bottleneck of a system is an element of software, rather than a hardware device. Software congestion can occur in any system which contains one or more elements of software whose services may be simultaneously desired by multiple clients, but which can service only one client at a time. It is shown that the use of models which ignore software congestion can produce results that are completely irrelevant to actual system behavior. Furthermore, software congestion is frequently invisible to conventional performance measurement tools. A notational scheme, called mobile servers representation, is introduced for describing those systems in which software congestion may be important. An approximate analytical model, called the hyperbolic model, is developed for analyzing systems with software congestion.

[1] B. D. Jensen, private communication.
[2] S. Das, private communication.
[3] R. J. T. Morris and J. S. Kaufman, "Performance comparison of I/O access disciplines for transaction-processing systems,"Comput. J., vol. 25, no. 1, pp. 74-83, 1982.
[4] C. H. Sauer and E. A. MacNair, "Simultaneous resource procession in queueing models of computers,"Perform. Eval. Rev., vol. 7, no. 1&2, pp. 41-52. Spring-Summer 1978.
[5] A. B. Pritsker,Introduction to Simulation and SLAM II. New York: Wiley, 1986.
[6] P. A. Jacobson and E. D. Lazowska, "Analyzing queueing networks with simultaneous resource possession,"Commun. ACM, vol. 25, pp. 142-151, Feb. 1982.
[7] K. M. Chandy, U. Herzog, and L. Woo, "Parametric analysis of queueing networks,"IBM J. Res. Develop., vol. 19, no. 1, Jan. 1975.
[8] E. D. Lazawskaet al., Quantitative System Performance--Computer System Analysis Using Queueing Network Models. Englewood Cliffs, NJ: Prentice-Hall, 1984.
[9] E. Souza e Silva, and R. R. Muntz, "Approximate solutions for a class of non-product form network models,"Performance Eval., vol. 7, pp. 221-242, 1987.
[10] J. S. Kaufman, private communication.
[11] V. L. Wallace and R. S. Rosenberg, "Markovian models and numerical analysis of computer system behavior," inAFIPS Spring Joint Computer Conf. Proc., 1966, pp. 141-148.
[12] B. D. Jensen, private communication.
[13] J. S. Kaufman, private communication.
[14] J. S. Kaufman, "Finite and infinite source interactions," inPerformance '84, E. Gelenbe, Ed. Amsterdam, The Netherlands: Elsevier Science, 1984, pp. 345-359.
[15] K.C. Sevcik and I. Mitrani, "The Distribution of Queueing Network States at Input and Output Instants,"J. ACM, Vol. 28, No. 2, Apr. 1981, pp. 358-371.
[16] L. Kleinrock,Queueing Systems, vol. 1. New York: Wiley, 1975, p. 187.
[17] J. S. Kaufman and Y. T. Wang, "Approximate analysis of a simultaneous resource possession problem," inProc. ICCC 88, Tel Aviv. Oct. 30, 1988.
[18] L. Kleinrock,Queueing Systems, vol. 2. New York: Wiley, 1976, pp. 114-117.

Index Terms:
system performance bottleneck; system description; simultaneously desired software elements; irrelevant results; residual capacity model; software congestion system analysis; mobile servers; software congestion; multiple clients; system behavior; invisible; performance measurement tools; notational scheme; mobile servers representation; analytical model; hyperbolic model; fault tolerant computing; performance evaluation; specification languages
Citation:
M.L. Fontenot, "Software Congestion, Mobile Servers, and the Hyperbolic Model," IEEE Transactions on Software Engineering, vol. 15, no. 8, pp. 947-962, Aug. 1989, doi:10.1109/32.31352
Usage of this product signifies your acceptance of the Terms of Use.