|
 ABSTRACT
Small computers often use floppy disks for storage. Since such disks are significantly slower than hard disks, the response time of a given application program is due predominantly to the time required to access data in files. Access time is dependent on three factors: hardware (disk drive and interface), the operating system in use, and the patterns of file access of application programs. A simple-to-use method of predicting access times with good accuracy is presented. The method makes use of simple calibration experiments that provide information related to the hardware and operating system characteristics of the system in use. This is combined with patterns of file access information for a given application program to predict access time for that program. Actual examples for several small computer systems are 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
|
Special issue on queueing networks. Computing Surveys, 10, 3, (September 1978). A survey of queueing network analysis techniques for analysis of performance of computer systems. Discussion of the assumptions and limitations of this approach to analysis.
|
| |
2
|
|
| |
3
|
|
| |
4
|
|
 |
5
|
|
| |
6
|
Finkel, L. and Brown, J. R., Data File Programming in BASIC. John Wiley and Sons, New York, 1981, Chapters 4-8. Different file organizations are explained including their access through the Basic language.
|
| |
7
|
Hellerman, H. and Conroy, T. F., Computer System Performance. McGraw-Hill, New York, 1975, Chapters 5-6. A good survey of techniques for computer system performance analysis.
|
| |
8
|
Hertrich, F. R., Average motion times of positioners in random access devices. IBM Journal, 9, 2 (March 1965) pp 124-133. A discussion of the design of a hard disk system with emphasis on the rationalization for the shape of the disk seek curve.
|
| |
9
|
Knuth, D. E., The Art of Computer Programming. Vol 3, Addison- Wesley, Reading, MA, 1973, pp 362-363. A simplified model of disk access is presented with sample calculations of access time. Simplifying assumptions are also presented.
|
| |
10
|
Rosenberg, R., Hard disk drives. Mini-micro Systems, 12, 2 (February 1979), pp 47-68. A good survey of hard disk systems with listing of pertinent disk parameters such as storage capacity and seek rates.
|
| |
11
|
Sauer, C. H. and Chandy, K. M., Computer Systems Performance Modeling. Prentice-Hall, Englewood Cliffs, N}, 1981, Chapters 2-3. A discussion of queueing network techniques for performance analysis including numerical methods for solution.
|
| |
12
|
Shugart, SA4OO0 fixed disk drive. OEM Manual #39005-1, Shugart Associates, Sunnyvale, CA, 1978, pp 1-10. Specifications for a Winchester hard disk drive.
|
| |
13
|
Thompson, K., UNIX implementation. Bell Systems Technical Journal, 57, #6 (July-August 1978), pp 1931-1946. The organization of the UNIX file system is explained along with its effect on response time. Some of the hidden disk accesses for directory and mapping information are presented.
|
| |
14
|
Western Digital. The FD179X-01 floppy disk formatter/controller family. Document #A0104, Western Digital Corporation, 1978, pp 1- 20. Specifications of an advanced large-scale integrated circuit controller chip and the many options available for formatting and controlling a floppy disk.
|
| |
15
|
White, R. M., Disk-storage technology. Scientific American (August 1980), pp 138-147. A general discussion of disk technology, design, and performance.
|
|
Adobe Acrobat
QuickTime
Windows Media Player
Real Player
Pdf
D.4