A workload model representative of static and dynamic characteristics

SummaryA technique to implement a workload model that must be representative of both static and dynamic characteristics of a workload is presented. The main goal of this work is the construction of a representative and compact artificial workload model. The approach taken is first to assign the set of workload components to classes having homogeneous static (i.e., load-independent) characteristics using clustering and then to model the dynamic sequence of components execution with a suitable stochastic process. The representativeness of such a workload model may be verified applying the physically or the function-oriented criteria for the static aspects and the performance-oriented criterion for the dynamic aspects considered. The results of an experimental application of this technique to model the workload of a university environment are presented.

[1]  Günter Haring,et al.  On Stochastic Models of Interactive Workloads , 1983, Performance.

[2]  Kishor S. Trivedi.  Probability and Statistics with Reliability, Queuing, and Computer Science Applications , 1984 .

[3]  Giuseppe Serazzi,et al.  Measurement and Tuning of Computer Systems , 1984, Int. CMG Conference.

[4]  Terry Williams,et al.  Probability and Statistics with Reliability, Queueing and Computer Science Applications , 1983 .

[5]  Domenico Ferrari,et al.  On the foundations of artificial workload design , 1983, SIGMETRICS '84.

[6]  H. Pat Artis.  Capacity planning for MVS computer systems , 1979, PERV.

[7]  Gianfranco Balbo,et al.  Computational algorithms for closed queueing networks , 1980 .

[8]  Erhan Çinlar,et al.  Introduction to stochastic processes , 1974 .

[9]  Paul O'Neill,et al.  Performance statistics of a time sharing network at a small university , 1980, CACM.

[10]  Domenico Ferrari,et al.  A sensitivity study of the clustering approach to workload modeling (extended abstract) , 1985, SIGMETRICS 1985.

[11]  A. O. Pittenger.  Introduction to Stochastic Processes (Erhan Çinlar) , 1977 .

[12]  Giuseppe Serazzi,et al.  A Characterization of the Variation in Time of Workload Arrival Patterns , 1985, IEEE Transactions on Computers.

[13]  Domenico Ferrari,et al.  A Sensitivity Study of the Clustering Approach to Workload Modeling , 1986, Perform. Evaluation.

[14]  Antoine-S Bailly,et al.  Science régionale - Walter Isard, Introduction to régional science. Englewood Cliffs (NJ), Prentice-Hall, 1975 , 1975 .

[15]  Robert F. Ling,et al.  Cluster analysis algorithms for data reduction and classification of objects , 1981 .

[16]  Peter A. W. Lewis,et al.  Statistical Analysis of Non-Stationary Series of Events in a Data Base System , 1976, IBM J. Res. Dev..

[17]  John A. Hartigan,et al.  Clustering Algorithms , 1975 .

[18]  K. Sreenivasan,et al.  On the construction of a representative synthetic workload , 1974, CACM.

[19]  Ashok K. Agrawala,et al.  An Approach to the Workload Characterization Problem , 1976, Computer.

[20]  Stephen S. Lavenberg,et al.  Exploratory Analysis of Access Path Length Data for a Data Base Management System , 1976, IBM J. Res. Dev..

[21]  Reinhard Posch,et al.  The Use of a Synthetic Jobstream in Performance Evaluation , 1979, Comput. J..

[22]  Erol Gelenbe.  Stationary deterministic flows in discrete systems: i , 1982, SIGMETRICS '82.

[23]  Erol Gelenbe,et al.  The Deterministic Flows Model: Parameter Identification , 1985, MMB.

[24]  Günter Haring.  On State-Dependent Workload Characterization by Software Resources , 1982, SIGMETRICS Perform. Evaluation Rev..

[25]  M. L. Bolzoni,et al.  A package for the implementation of static workload models , 1982, SIGMETRICS '82.