Abstract
A computer system's workload is represented by its multiprogramming level, which is defined as the number of tasks (jobs, customers) which actively compete for resources within the system. In a product-form queuing network model of the system, the workload is modeled by assuming that the multiprogramming level is either fixed (i.e., closed model) or that the multiprogramming level depends upon an outside arrival process (i.e., open model). However, in many actual systems, closed and open models are both inappropriate since the multiprogramming level is neither fixed nor governed by an outside arrival process.
In an actual system., the multiprogramming level varies due to features such as task spawning, killing, blocking, parallel processing, and/or simultaneous resource possession. The multiprogramming level is a random variable with an associated distribution. This paper demonstrates that improved models can result from using this multiprogramming level distribution information. Several examples relative to open versus closed models, subsystem models, actual system models, and blocking models are given which demonstrate the applicability of using multiprogramming level distributions. This applicability, shown via the examples, is the main contribution of the paper. The examples also motivate interesting theoretical results relating to open models, closed models, and subsystem models.
- BBA 84 S.C. Bruell, G. Balbo, and P.V. Afshaxi. Mean value analysis of mixed, multiple class BCMP network with load dependent service stations. Performance Evaluation ~, 4 {Nov. 1984), 241-260.Google Scholar
Cross Ref
- BCMP 75 F. Baskett, K.M. Chandy, R.R. Muntz, and F.G. Palacios. Open, closed, and mixed networks of queues with different classes of customers. J. ol the ACM ~, 2 (Apr. 1975), 248-260. Google Scholar
Digital Library
- BD 78 G. Balbo and P.J. Denning. Approximating load distributions in time sharing systems. Proc. ~nd Int'l Syrup. on Operating Systems, IRIA Laboria, Rocquencourt, France, (Oct. 1978).Google Scholar
- Buze 73 J.P. Buzen. Computational algorithms for closed queueing networks with exponential servers. CACM 16, 9 (Sept. 1073), 527-531. Google Scholar
Digital Library
- CHW 75 K.M. Chandy, U. Herzog, and L. Woo. Parametric analysis of queueing networks. IBM J. Res. Develop. i9, 1 (Jan. 1975), 36-42.Google Scholar
Digital Library
- DG 84 L.W. Dowdy and K.D. Gordon. Algorithms for nonlntegral degrees of muRiprogramming in closed queuing networks. Performance Evaluation 4, 1 (Feb. 1984), 19-29.Google Scholar
Cross Ref
- DG 85 L.W. Dowdy and K.D. Gordon. Performance sensitivity with respect to the distribution of multiprogramming level. (submitted). Tech. Rep. CS-84-07, Computer Science Dept.,theVanderbilt Univ., Nashville, Tennessee.Google Scholar
- JL 83 P.A. Jacobson and E.D. Lazowska. A reduction technique for evaluating queueing networks with serialization delays. Performance '83, (A.K. Agrawala and S.K. Tripathi, eds.), North Holland, (May 1983), 45-59. Google Scholar
Digital Library
- Lam 77 S.S. Larn. Queueing networks with population size constraints. IBM J. Res. Develop. /~1, 4 {July 1977), 370-378.Google Scholar
- LC 77 L. Lipsky and J.D. Church. Applications of a queueing network model for a computer system. Computing Surveys 9, 3 (Sept. 1977), 205-221. Google Scholar
Digital Library
- RK 75 M. Reiser and H. Kobayashi. Queuing networks with multiple closed chains: theory and computational algorithms. IBM J. Rcs. Develop. 19, 3 (May 1975), 283-294.Google Scholar
Digital Library
- RL 80 M. Reiser and S.S. Lavenberg. Mean value analysis of closed multichain queueing networks. J. of the A CM PT, 2 (Apr. 1980), 313-322. Google Scholar
Digital Library
- Zaho 83 J. Zahorjan. The influence of workload representation on response times in queueing models of computer systems. Proc. of the 1988 A CM Sigmctrica Conf. on Mcas. and Modeling of Comp. Sys., Minneapolis, Minnesota, (Aug. 1983), 70-81. Google Scholar
Digital Library
Index Terms
On the applicability of using multiprogramming level distributions
Recommendations
On the applicability of using multiprogramming level distributions
SIGMETRICS '85: Proceedings of the 1985 ACM SIGMETRICS conference on Measurement and modeling of computer systemsA computer system's workload is represented by its multiprogramming level, which is defined as the number of tasks (jobs, customers) which actively compete for resources within the system. In a product-form queuing network model of the system, the ...
Performance of two-stage tandem queues with blocking: the impact of several flows of signals
This paper considers a two-stage tandem G-queue with blocking, service requirements of phase type and arrivals of units and of signals - which cancel one unit waiting in line or in service - both assumed to be Markovian arrival processes. The main ...
Reversibility of Tandem Blocking Queueing Systems
<P>This paper is concerned with queueing systems of several service stations in series in which each station may consist of multi-servers. An infinite number of customers always waits in front of the first station, and each customer passes through all ...






Comments