skip to main content
article

Disk scheduling in a multimedia I/O system

Published:01 February 2005Publication History
Skip Abstract Section

Abstract

This article provides a retrospective of our original paper by the same title in the Proceedings of the First ACM Conference on Multimedia, published in 1993. This article examines the problem of disk scheduling in a multimedia I/O system. In a multimedia server, the disk requests may have constant data rate requirements and need guaranteed service. We propose a new scheduling algorithm, SCAN-EDF, that combines the features of SCAN type of seek optimizing algorithm with an Earliest Deadline First (EDF) type of real-time scheduling algorithm. We compare SCAN-EDF with other scheduling strategies and show that SCAN-EDF combines the best features of both SCAN and EDF. We also investigate the impact of buffer space on the maximum number of video streams that can be supported.We show that by making the deadlines larger than the request periods, a larger number of streams can be supported.We also describe how we extended the SCAN-EDF algorithm in the PRISM multimedia architecture. PRISM is an integrated multimedia server, designed to satisfy the QOS requirements of multiple classes of requests. Our experience in implementing the extended SCAN-EDF algorithm in a generic operating system is discussed and performance metrics and results are presented to illustrate how the SCAN-EDF extensions and implementation strategies have succeeded in meeting the QOS requirements of different classes of requests.

References

  1. Abbott, R. and Garcia-Molina, H. 1990. Scheduling I/O requests with deadlines: A performance evaluation. In Proceedings of the IEEE Real-Time Systems Symposium. IEEE Computer Society Press, Los Alamitos, Calif. 113--124.Google ScholarGoogle Scholar
  2. Anderson, D. P., Osawa, Y., and Govindan, R. 1992. A file system for continuous media. ACM Trans. Comput. Syst. 10, 4 (Nov.), 311--337. Google ScholarGoogle ScholarDigital LibraryDigital Library
  3. Anderson, D. P., Osawa, Y., and Govindan, R. 1991. Real-time disk storage and retrieval of digital audio/video data. Tech. rep. UCB/CSD 91/646, University of California, Berkeley, Calif. Google ScholarGoogle Scholar
  4. Blanquer, J., Bruno, J., Gabber, E., Mcshea, M., Ozden, B., and Silberschatz, A. 1999. Resource management for QOS in Eclipse/BSD. FreeBSD Conference. Available HTTP:http://freebsdcon.org/1999/exhibitors/; accessed April 2000.Google ScholarGoogle Scholar
  5. Bolosky, W. J., Fitzgerald, R. P., and Doucer, J. R. 1997. Distributed schedule management in the tiger video file server. In Proceedings of the ACM Symposium on Operating Systems Principles. ACM, New York, 212--223. Google ScholarGoogle Scholar
  6. Bosch, P., Mullender, S. J., and Jansen, P. G. 1999. Clockwise: A mixed-media file system. In Proceedings of the IEEE International Conference on Multimedia Computing and Systems 2. IEEE Computer Society Press, Los Alamitos, Calif., 277--281. Google ScholarGoogle Scholar
  7. Bruno, J., Brustolini, J., Gabber, E., Ozden, B., and Silbershatz, A. 1999. Disk scheduling with quality of service guarantees. In Proceedings of the IEEE International Conference on Multimedia Computing and Systems 2. IEEE Computer Society Press, Los Alamitos, Calif., 400--405. Google ScholarGoogle Scholar
  8. Card, R., Dumas, E., and Mevel, F. 1998. The Linux Kernel Book. Wiley, New York, Chap. 3--16. Google ScholarGoogle Scholar
  9. Chang, E. and Zakhor, A. 1994. Scalable video data placement on parallel disk arrays. In Proceedings of the SPIE Symposium on Electronic Imaging Science and Technology. 208--221.Google ScholarGoogle Scholar
  10. Chang, R. I., Shih, W. K., and Chang, R. C. 2000. Deadline modification scan with maximum scannable groups for multimedia real-time disk scheduling. In Proceedings of the Real-Time Systems Symposium 19, 149--168. Google ScholarGoogle ScholarDigital LibraryDigital Library
  11. Chen, H. J., Krishnamurthy, A., Little, T. D., and Venkatesh, D. 1995. A scalable video-on-demand service for the provision of VCR-like functions. In Proceedings of the of IEEE Conference on Multimedia Computing and Systems. IEEE Computer Society Press, Los Alamitos, Calif., 65--72. Google ScholarGoogle Scholar
  12. Chen, M. S., Kandlur, D., and Yu, P. S. 1994. Support for fully interactive playout in a disk-array-based video server. In Proceedings of the ACM Multimedia Conference. ACM, New York, 391--398. Google ScholarGoogle Scholar
  13. Chen, S., Stankovic, J. A., Kurose, J. F., and Towsley, D. 1991. Performance evaluation of two new disk scheduling algorithms for real-time systems. J. Real-Time Syst. 3, 307--306.Google ScholarGoogle ScholarCross RefCross Ref
  14. Chiueh, T., Venkatramani, C., and Vernick, M. 1996. Design of the stony brook video server. In Proceedings of the SPIE First International Symposium on Technologies and Systems for Voice, Video and Data Communications 2604, 133--145.Google ScholarGoogle Scholar
  15. Freedman, C. and DeWitt, D. 1995. The SPIFFI scalable video-on-demand system. In Proceedings of the ACM SIGMOD Conference. ACM, New York, 352--363. Google ScholarGoogle Scholar
  16. Haskin, R. L. 1998. Tiger Shark---A scalable file system for multimedia. IBM J. Res. Develop. 42, 2 (Mar), 185--197. Google ScholarGoogle ScholarDigital LibraryDigital Library
  17. Jayanta, K. D., Salehi, J. D., Kurose, J. F., and Towsley, D. 1994. Providing vcr capabilities in large-scale video server. In Proceedings of the ACM Multimedia Conference. ACM, New York, 25--32. Google ScholarGoogle Scholar
  18. Jeffay, K., Stanat, D. F., and Martel, C. U. 1991. On non-preemptive scheduling of periodic and sporadic tasks. In Proceedings of the of Real-time Systems Symposium. 129--139.Google ScholarGoogle Scholar
  19. Kim, M. Y. 1986. Synchronized disk interleaving. IEEE Trans. Comput. C-35, 11, 978--988. Google ScholarGoogle ScholarDigital LibraryDigital Library
  20. Lehoczky, J. 1990. Fixed priority scheduling of periodic task sets with arbitrary deadlines. In Proceedings of Real-time Systems Symposium. 201--212.Google ScholarGoogle Scholar
  21. Lin, T. H. and Tarng, W. 1991. Scheduling periodic and aperiodic tasks in hard real-time computing systems. In Proceedings of ACM SIGMETRICS. ACM, New York, 31--38. Google ScholarGoogle Scholar
  22. Liu, C. L. and Layland, J. W. 1973. Scheduling algorithms for multiprogramming in a hard real-time environment. J. ACM, 46--61. Google ScholarGoogle Scholar
  23. Lougher, P. and Shepherd, D. 1993. The design of a storage server for continuous media. The Comput. J. 36, 1 (Feb.), 32--42.Google ScholarGoogle ScholarCross RefCross Ref
  24. Lund, K. and Goebel, V. 2003. Adaptive disk scheduling in a multimedia DBMS. In Proceedings of the ACM Multimedia Conference. ACM, New York, 65--74. Google ScholarGoogle Scholar
  25. Makaroff, D., Neufeld, G., and Hutchinson, N. 1997. An evaluation of vbr disk admission algorithms for continuous media file servers. In Proceedings of the ACM Multimedia Conference. ACM, New York, 143--154. Google ScholarGoogle Scholar
  26. Martin, C., Narayan, P. S., Ozden, B., Rastogi, R., and Silberschatz, A. 1996. The Fellini Multimedia Storage System. Kluwer Academic Publications, Chap. 5.Google ScholarGoogle Scholar
  27. Mokbel, M. F., Aref, W. G., Elbassioni, K., and Kamel, I. 2004. Scalable multimedia disk scheduling. In Proceedings of the International Conference of Data Engineering. 498--509. Google ScholarGoogle Scholar
  28. Molano, A., Juvva, K., and Rajkumar, R. 1997. Real-time file systems: Guaranteeing timing constraints for disk accesses in rt_mach. In Proceedings of the IEEE Real-Time Systems Symposium. IEEE Computer Society Press, Los Alamitos, Calif., 155--165. Google ScholarGoogle Scholar
  29. Patterson, D. A., Gibson, G., and Katz, R. H. 1988. A case for redundant arrays of inexpensive disks (RAID). In Proceedings of the ACM SIGMOD Conference. ACM, New York, 109--116. Google ScholarGoogle Scholar
  30. Plagemann, T., Goebel, V., Halvorsen, P., and Anshus, O. 2000. Operating system support for multimedia systems. The Comput. Commun. J. 23, 3 (Feb.), 267--289. Google ScholarGoogle ScholarDigital LibraryDigital Library
  31. Reddy, A. L. N. 1992. A study of I/O system organizations. In Proceedings of the International Symposium on Computer Architecture. 308--317. Google ScholarGoogle Scholar
  32. Reddy, A. L. N. and Banerjee, P. 1989. An evaluation of multiple-disk I/O systems. IEEE Trans. Comput. C-38, 12 (Dec.), 1680--1690. Google ScholarGoogle ScholarDigital LibraryDigital Library
  33. Reddy, A. L. N. and Haskin, R. 1996. The Communications Handbook. CRC Press, Chapt. 106.Google ScholarGoogle Scholar
  34. Reddy, A. L. N. and Wyllie, J. 1993. Disk scheduling in a multimedia I/O system. In Proceedings of the ACM Multimedia Conference. ACM, New York, 225--233. Google ScholarGoogle Scholar
  35. Rompogiannakis, T., Nerjes, G., Muth, P., Paterakis, M., Triantafillou, P., and Weikum, G. 1998. Disk scheduling for mixed-media workloads in a multimedia server. In Proceedings of the ACM Multimedia Conference. ACM, New York, 297--302. Google ScholarGoogle Scholar
  36. Rose, O. 1995. Statistical Properties of MPEG video traffic and their impact on traffic modeling in ATM systems. Tech. Rep. 101, Institute of Computer Science, University of Wurzburg. Google ScholarGoogle Scholar
  37. Salem, K. and Garcia-Molina, H. 1986. Disk striping. In Proceedings of the International Conference on Data Engineering. 336--342. Google ScholarGoogle Scholar
  38. Shenoy, P., Goyal, P., and Vin, H. M. 1999. Architectural considerations for next generation file systems. In Proceedings of the ACM Multimedia Conference. ACM, New York, 457--467. Google ScholarGoogle Scholar
  39. Shenoy, P. J., Goyal, P., Rao, S., and Vin, H. M. 1998. Symphony: An integrated multimedia file system. In Proceedings of the ACM/SPIE Multimedia Computing and Networking. ACM, New York, 124--138.Google ScholarGoogle Scholar
  40. Shenoy, P. J. and Vin, H. M. 1998. Cello: A disk scheduling framework for next generation operating systems. Proc. ACM SIGMETRICS 26, 1 (June), 44--55. Google ScholarGoogle ScholarDigital LibraryDigital Library
  41. Shih, W. K., Liu, J. W., and Liu, C. L. 1992. Modified rate monotone algorithm for scheduling periodic jobs with deferred deadlines. Tech. Rep. University of Illinois, Urbana-Champaign.Google ScholarGoogle Scholar
  42. Wijayaratne, K. B. R. 2001. PRISM: A file server architecture for providing integrated services. Ph.D. dissertation, Department of Computer Science, Texas A&M University, Texas. Google ScholarGoogle Scholar
  43. Wijayaratne, R. and Reddy, A. L. N. 1999a. Integrated QOS management for disk I/O. In Proceedings of the IEEE International Conference on Multimedia Computing and Systems 1. IEEE Computer Society Press, Los Alamitos, Calif., 487--492. Google ScholarGoogle Scholar
  44. Wijayaratne, R. and Reddy, A. L. N. 1999b. Techniques for improving the throughput of VBR streams. ACM/SPIE Multimed. Comput. Netw. 3654, 216--227.Google ScholarGoogle Scholar
  45. Wijayaratne, R. and Reddy, A. L. N. 2000. Providing QOS guarantees for disk I/O. Multimed. Syst 8, 1 (January), 57--68. Google ScholarGoogle ScholarDigital LibraryDigital Library
  46. Wijayaratne, R. and Reddy, A. L. N. 2001. System support for providing integrated services from networked multimedia storage servers. In Proceedings of the ACM Multimedia Conference. ACM, New York, 270--279. Google ScholarGoogle Scholar
  47. Yee, J. and Varaiya, P. 1992. Disk scheduling policies for real-time multimedia applications. Tech. rep. University of California, Berkeley, Berkeley, Calif.Google ScholarGoogle Scholar
  48. Yu, P. S., Chen, M. S., and Kandlur, D. D. 1993. Grouped sweeping scheduling for dasd-based multimedia storage management. IEEE Multimed. Syst. 1, 99--109.Google ScholarGoogle ScholarDigital LibraryDigital Library

Index Terms

  1. Disk scheduling in a multimedia I/O system

    Recommendations

    Reviews

    Eno Thereska

    The progress made over a ten-year period on scheduling policies for multimedia systems is summarized in this paper, which also analyzes in detail a scheduling algorithm called SCAN-EDF, which the authors first developed in 1993. The original 1993 paper [1] described the algorithm and basic performance evaluations; this paper provides a retrospective look, based on measurements taken of the algorithm on a real multimedia server (called PRISM). Whereas resources like the central processing unit (CPU) and the network can be partitioned among several request streams using the earliest deadline first (EDF) algorithm, a disk's bandwidth cannot be partitioned efficiently using this scheduling algorithm. The reason is that mechanical delays induced when moving the disk head are significant (measured in milliseconds in 2005), and the time wasted when positioning from one request stream to the next can be significantly larger than the time required to read the blocks that belong to the request stream. Such waste exists for other resources too (such as CPU context switch times), but it is especially large for disks. Traditional bandwidth-maximizing scheduling algorithms, like shortest seek time first (SSTF) and shortest positioning time first (SPTF), do not work for multimedia streams either, since they are not fair, and may induce starvation. Hence, these algorithms are not considered. The authors instead propose extending a commonly used fair-scheduling policy, SCAN, which provides fairness at the expense of throughput, with EDF. SCAN is not as efficient as SSTF, but approximates SSTF fairly well. Combined with EDF, the amount of disk bandwidth is maximized while still meeting request stream deadlines. The authors compare this new algorithm with other ones (such as EDF, SSTF, and SCAN in isolation), and show it to be efficient. This paper summarizes other related work well. The main deficiency with this paper is that the evaluation considers out-of-date disks, which are more representative of disks ten years ago than of disks today. For example, the disk's capacity is 4.3 gigabytes, the average seek time is 11 milliseconds, and the number of sectors per track is just 63. These are nontrivial parameters that should be updated. In modern disks, the seek time is five milliseconds, the capacity is much larger, and the rotational speeds are very fast. The video and audio rate requirements have been constant during the last ten years (for example, 30 frames per second), so any evaluation done in 2005 should consider the new technology. Another possible addition to this work would be to consider disks with preemptible requests. Recent work by Dimitrijevic and others [2] at the 2003 File and Storage Technologies (FAST) conference illustrates storage systems that can use preemption. Such preemption would perhaps make scheduling in a multimedia environment easier. Online Computing Reviews Service

    Access critical reviews of Computing literature here

    Become a reviewer for Computing Reviews.

    Comments

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Sign in

    Full Access

    • Published in

      cover image ACM Transactions on Multimedia Computing, Communications, and Applications
      ACM Transactions on Multimedia Computing, Communications, and Applications  Volume 1, Issue 1
      February 2005
      125 pages
      ISSN:1551-6857
      EISSN:1551-6865
      DOI:10.1145/1047936
      Issue’s Table of Contents

      Copyright © 2005 ACM

      Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

      Publisher

      Association for Computing Machinery

      New York, NY, United States

      Publication History

      • Published: 1 February 2005
      Published in tomm Volume 1, Issue 1

      Permissions

      Request permissions about this article.

      Request Permissions

      Check for updates

      Qualifiers

      • article

    PDF Format

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader