Do disk arms move?

Authors:
W. C. Lynch
Affiliations:
Case Western Reserve University, Cleveland, Ohio
Venue:
ACM SIGMETRICS Performance Evaluation Review
Year:
1972

Citing 2
Cited 11

The working set model for program behavior

Communications of the ACM
A comparative analysis of disk scheduling policies

SOSP '71 Proceedings of the third ACM symposium on Operating systems principles

Disk cache—miss ratio analysis and design considerations

ACM Transactions on Computer Systems (TOCS)
Disk file allocation based on the buddy system

ACM Transactions on Computer Systems (TOCS)
Disk arm movement in anticipation of future requests

ACM Transactions on Computer Systems (TOCS)
Improving Disk Performance Via Latency Reduction

IEEE Transactions on Computers
A Performance Evaluation of RAID Architectures

IEEE Transactions on Computers
An anomaly in disk scheduling: a comparison of FCFS and SSTF seek scheduling using an empirical model for disk accesses

Communications of the ACM
Advances in Disk Technology: Performance Issues

Computer
Design and Performance Analysis of a Disk Array System

IEEE Transactions on Computers
On the effectiveness of buffered and multiple arm disks

ISCA '78 Proceedings of the 5th annual symposium on Computer architecture
The optimal choice of window sizes for working set dispatching

SIGMETRICS '74 Proceedings of the 1974 ACM SIGMETRICS conference on Measurement and evaluation
Workloads (creation and use)

Communications of the ACM

Quantified Score

Hi-index	0.04

Visualization

Abstract

Measurement of the lengths of disk arm movements in a 2314 disk storage facility of an IBM 360/67 operating under the Michigan Terminal System yielded the unexpected data that the arms need not move in 63% of the accesses and need move for an average of only 30ms. in the remaining 37% of the cases. A description and analysis of a possible mechanism of action is presented. The predictions of this model do not disagree with the measured data.