|
ABSTRACT
A model for disk accesses based on published measurements is developed. The model is used to show that under highly probable conditions, FCFS seek scheduling is superior to SSTF scheduling in the sense of having a lower mean queue length. A simple example of an arrival sequence illustrating this anomaly is presented.
REFERENCES
Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.
| |
1
|
Denning, P.J. Effects of scheduling on file memory operations. AFIPS Conf. Proc., Vol. 30, 1967 SJCC, AFIPS Press, Montvale, N.J., 1967, pp. 9--21.
|
 |
2
|
|
 |
3
|
|
| |
4
|
Teorey, T.J. Properties of disk scheduling policies in multiprogrammed computer systems. AFIPS Conf. Proc., Vol. 41, 1972 FJCC AFIPS Press, Montvale, N.J., 1972, pp. 1-11.
|
 |
5
|
|
 |
6
|
|
 |
7
|
|
| |
8
|
Cox, D.R., and Smith, W.L. Queues. Chapman and Hall, London, 1961.
|
| |
9
|
Wilhelm, N.C. A general model for the performance of disk systems (unpublished).
|
| |
10
|
Wilhelm, N.C., Analysis of FCFS and SSTF seek scheduling using an empirical model for disk accesses. Tech. Rep. 64, Digital Systems Lab., Stanford U., Aug. 1973.
|
 |
11
|
|
|