Skip to main content
Log in

The effect of fixed-length record implementation on file system response

  • Published:
Acta Informatica Aims and scope Submit manuscript

Summary

Many file and database management systems are implemented using fixed-length physical records on movable head discs. Such implementation not only causes internal dependence among different components of the file access time, but also induces an external dependence of the overall access time on request input rate. Accordingly, conventional performance estimates are inaccurate here, and the error in using them typically ranges from 10% to 100+% for different loading conditions. This paper presents revised formulae for calculating the file response time, throughput and busy period for such a storage structure. The analysis makes use of the Takács process in queueing theory.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Abate, J., Dubner, H., Weinberg, S.B.: Queueing analysis of the IBM 2314 disk storage facility. JACM 15, 577–589 (1968)

    Google Scholar 

  2. CDC: CYBER 70, 7600, 6000 computer systems record manager reference manual (1973)

  3. Choo, Q.H., Conolly, B.W.: New results in the theory of repeated orders queueing systems. J. Appl. Prob. 16, 631–640 (1979)

    Google Scholar 

  4. CINCOM: TOTAL/7 reference manual (1974)

  5. Cox, D.R., Miller, H.D.: The theory of stochastic processes. Chapman and Hall (1965)

  6. Fife, D.W., Smith, J.L.: Transmission capacity of disk storage systems with concurrent arm positioning. IEEE Trans. EC-14, 575–582 (1965)

    Google Scholar 

  7. Frank, H.: Analysis and optimization of disk storage devices for time-sharing systems. JACM 16, 602–620 (1969)

    Google Scholar 

  8. Fuller, S.H.: Analysis of Drum and Disk Storage Units. Berlin Heidelberg New York: Springer (1975)

    Google Scholar 

  9. Gotlieb, C.C., and MacEwen, G.H.: Performance of movable-head disk storage devices. JACM 20, 604–623 (1973)

    Google Scholar 

  10. IBM: Introduction to virtual storage in system/370 (1973)

  11. IBM: Reference manual for IBM 3830 storage control model 1 and IBM 3330 disk storage (1974)

  12. IBM: Reference manual for IBM integrated storage control (1976)

  13. IBM: Introduction to IBM direct-access storage devices and organization methods (1976)

  14. ICL: 1900 series direct access reference manual (1974)

  15. Leung, C.H.C.: A certain correlation between disk seek and rotational delay. Technical Report RCS 115, Computer Science Dept., University of Reading (1979)

  16. Leung, C.H.C.: Applied Probability Models for the Performance Evaluation of Disk File Processing. Ph.D. Thesis, University of London (1980)

  17. Martin, J.: Computer Data-base Organization. Prentice-Hall (1977)

  18. McGee, W.C.: The information management system IMS/VS Part II: Data base facilities. IBM Syst. J. 16, 96–122 (1977)

    Google Scholar 

  19. Weingarten, A.: The analytical design of real-time disk systems. Information Processing 68 (North Holland), 860–866 (1969)

  20. Wilhelm, N.C.: An anomaly in disk scheduling: A comparison of FCFS and SSTF seek scheduling using an empirical model for disk accesses. Comm. ACM 19, 13–17 (1976)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Leung, C.H.C., Choo, Q.H. The effect of fixed-length record implementation on file system response. Acta Informatica 17, 399–409 (1982). https://doi.org/10.1007/BF00264159

Download citation

  • Received:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF00264159

Keywords

Navigation