PERFORMANCE MODELING AND EVALUATION OF A 2-DIMENSIONAL DISK ARRAY SYSTEM

Authors
Citation
Cs. Lee et Tm. Parng, PERFORMANCE MODELING AND EVALUATION OF A 2-DIMENSIONAL DISK ARRAY SYSTEM, Journal of parallel and distributed computing, 38(1), 1996, pp. 16-27
Citations number
25
Categorie Soggetti
Computer Sciences","Computer Science Theory & Methods
ISSN journal
07437315
Volume
38
Issue
1
Year of publication
1996
Pages
16 - 27
Database
ISI
SICI code
0743-7315(1996)38:1<16:PMAEOA>2.0.ZU;2-O
Abstract
This paper presents a performance model of a two-dimensional disk arra y (TIDA) system, which is composed of several major subsystems includi ng disk cache, intelligent disk array controller, SCSI-like I/O bus, a nd two-dimensional array of disk devices. Accessing conflict in these subsystems and fork/join synchronization of physical disk requests are considered in the model. The representation for the complex behavior, including the interactions among subsystems, of a whole disk array sy stem distinguishes the model from others that model only individual su bsystems. To assist evaluating the architectural alternatives of TIDA, we employ a subsystem access time modeling methodology, in which we m odel for each subsystem the mean subsystem access time per request (SA TPR). Fed with a given set of representative workload parameters, the performance model is used to conduct performance evaluation and the SA TPRs of the subsystems are utilized to identify the bottleneck subsyst em for performance improvement. The results show that (1) the values o f some key design parameters, such as data block size and I/O bus band width that yield the best system throughput are dependent not only on the subsystem performance but also on the interaction among subsystems ; (2) an I/O bus bandwidth of 5 Mbytes/s per disk device is large enou gh for data transfers from/to disk devices equipped with a cache of 1 Mbytes; and (3) the activity of fork/join synchronization of physical disk requests may cause performance degradation, which can be improved by using large I/O bus bandwidth and/or placing a cache in each disk device. (C) 1996 Academic Press, Inc.