Abstract
The information processing industry has shifted from the conventional mathematical computation to information management. Interactive use of a system with large data bases is now widespread, and the data base has become one of the key factors central to an overall system design. This, in turn, has prompted the development of new peripheral devices.
The paper focuses on developing and analyzing stochastic models for disk storage systems. A general model is developed to describe the sequence of operations required for accessing a record on a disk. The model is then used to analyze how technological and architectural changes affect system performance. As concrete examples, performance comparison of IBM 3330 disk system with IBM 3380 disk system will be conducted. All analytical results are validated via simulation.
Similar content being viewed by others
References
J. Abate, H. Dubner and S. Weinberg, Queueing analysis of the IBM 2314 disk storage facility, J. ACM 15, No. 4 (1968) 477–489.
Y. Bard, Task queueing in auxiliary storage devices with rotational position sensing, Rep. G320-2070, IBM Cambridge Science Center, Cambridge, Massachusetts, 1971.
J.W. Boyse and D.R. Warn, A straightforward model for computer performance predictions, Computing Surveys 7, No. 2 (1975) 73–93.
A. Brandwajn, Models of DASD subsystems with multiple access paths: a throughput-driven approach, IEEE Trans. on Computers 32, No. 5 (1983) 451–463.
W.M. Chow and W.W. Chiu, An analysis of swapping policies in virtual storage systems, IBM Research Report RC5776, 1975.
E.G. Coffman and M. Hofri, A class of FIFO queues arising in computer systems, Operations Research 26, No. 5 (1978) 864–880.
R.W. Conway, W.L. Maxwell and L.W. Miller,Theory of Scheduling (Addison-Wesley, Reading, Mass, 1967).
B.T. Doshi, Stochastic decomposition in a GI/G/1 queue with vacations, 1984, submitted for publication.
H. Frank, Analysis and optimization of disk storage devices for time-sharing systems, J. ACM 16, No. 4 (1969) 602–620.
E. Gelenbe and Iasnogorodski, A queue with serves of walking type, Ann. Inst. Henri Poincare 16, No. 1 (1980) 63–73.
C.C. Gotlieb and G.H. MacEwen, Performance of movable-head disk storage devices, J. ACM 20, No. 4 (1973) 604–623.
M. Hofri, Disk scheduling: FCFS vs. SSTF revisited, Comm. ACM 23 (1980) 645–653.
J. Keilson and L. Servi, Oscillating random walk models for GI/G/1 vacation systems with Bernoulli schedules, 1984, submitted for publication.
H. Kobayashi (1978),Modeling and Analysis: An Introduction to System Performance Evaluation Methodology (Addison-Wesley Publishing Company, 1978).
T. Lang, E. Nahouraii, K. Kasuga and E.B. Fernández, An architectural extension for a large data base incorporating a processor for disk search, IBM Los Angeles Scientific Center, G32D-2688, 1977.
T. Manocha, Analysis of an input/output subsystem with seek-type direct-access devices,IEEE Computer Conference (1972) pp. 155–158.
M. Neuts, Some explicit formulas for the steady-state behavior of a queue with semi-Markovian service times, Adv. Appl. Probl. 9, No. 1 (1977) 141–157.
W.C. Oney, Queueing analysis of the scan policy for moving head disks, J. ACM 22, No. 3 (1975) 397–412.
M. Scholl and L. Kleinrock, On the M/G/1 queue with rest periods and certain service-independent queueing, Operations Research 31, No. 4 (1983) 705–719.
C.F. Skinner, A priority queueing system with server walking time, Operations Research 15, No. 1 (1967) 14–31.
U. Sumita and Y. Masuda, Additive processes defined on a Markov chain with replacements: numerical exploration via the matrix Laguerre transform, 1987, in preparation:
N.C. Wilhelm, A general model for the performance of disk systems, J. ACM 24, No. 1 (1977) 14–31.
Author information
Authors and Affiliations
Additional information
Supported by the National Science Foundation under Grant No. ECS-8404071.
Rights and permissions
About this article
Cite this article
Gavish, B., Sumita, U. Analysis of channel and disk subsystems in computer systems. Queueing Syst 3, 1–23 (1988). https://doi.org/10.1007/BF01159084
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01159084