Advertisement

Managing the Performance Impact of Administrative Utilities

  • Sujay Parekh
  • Kevin Rose
  • Joseph Hellerstein
  • Sam Lightstone
  • Matthew Huras
  • Victor Chang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2867)

Abstract

Administrative utilities (e.g., filesystem and database backups, garbage collection in the Java Virtual Machines) are an essential part of the operation of production systems. Since production work can be severely degraded by the execution of such utilities, it is desirable to have policies of the form “There should be no more than an x% degradation of production work due to utility execution.” Two challenges arise in providing such policies: (1) providing an effective mechanism for throttling the resource consumption of utilities and (2) continuously translating from policy expressions of “degradation units” into the appropriate settings for the throttling mechanism. We address (1) by using self-imposed sleep, a technique that forces utilities to slow down their processing by a configurable amount. We address (2) by employing an online estimation scheme in combination with a feedback loop. This throttling system is autonomous and adaptive and allows the system to self-manage its utilities to limit their performance impact, with only high-level policy input from the administrator. We demonstrate the effectiveness of these approaches in a prototype system that incorporates these capabilities into IBM’s DB2 Universal Database server.

Keywords

Feedback Control Production Work Garbage Collection Performance Impact Java Virtual Machine 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Bruno, J., Gabber, E., Özden, B., Silberschatz, A.: The Eclipse operating system: Providing quality of service via reservation domains. In: Proceedings of the USENIX 1998 Annual Technical Conference, New Orleans, LA, pp. 235–246 (1998)Google Scholar
  2. 2.
    Banga, G., Druschel, P., Mogul, J.C.: Resource containers: A new facility for resource management in server systems. In: Proceedings of the Third Symposium on Operating Systems Design and Implementation (OSDI), New Orleans, LA, pp. 45–58 (1999)Google Scholar
  3. 3.
    Aman, J., Eilert, C.K., Emmes, D., Yocom, P., Dillenberger, D.: Adaptive algorithms for managing a distributed data processing workload. IBM Systems Journal 36 (1997)Google Scholar
  4. 4.
    Ryu, K.D., Hollingsworth, J.K.: Exploiting fine-grained idle periods in networks of workstations. IEEE Transactions on Parallel and Distributed Systems 11, 683–698 (2000)CrossRefGoogle Scholar
  5. 5.
    Wilson, P.R.: Uniprocessor garbage collection techniques. In: Bekkers, Y., Cohen, J. (eds.) IWMM-GIAE 1992. LNCS, vol. 637, pp. 1–42. Springer, Heidelberg (1992)CrossRefGoogle Scholar
  6. 6.
    Bacon, D.F., Cheng, P., Rajan, V.T.: A real-time garbage collector with low overhead and consistent utilization. In: Proceedings of the 30th ACM SIGPLANSIGACT Symposium on Principles of Programming Languages, pp. 285–298. ACM Press, New York (2003)CrossRefGoogle Scholar
  7. 7.
    Lu, C., Abdelzaher, T.F., Stankovic, J.A., Son, S.H.: A feedback control architecture and design methodology for service delay guarantees in web servers. Technical Report CS2001-06, University of Virginia, Department of Computer Science (2001) Google Scholar
  8. 8.
    Diao, Y., Gandhi, N., Hellerstein, J.L., Parekh, S., Tilbury, D.M.: Using MIMO feedback control to enforce policies for interrelated metrics with application to the Apache web server. In: Proceedings of Network Operations and Management (2002)Google Scholar
  9. 9.
    Astrom, K.J., Wittenmark, B.: Adaptive Control, 2nd edn. Addison-Wesley Publishing Company, Reading (1994)Google Scholar
  10. 10.
    Ogata, K.: Modern Control Engineering, 3rd edn. Prentice Hall, Englewood Cliffs (1997)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Sujay Parekh
    • 1
  • Kevin Rose
    • 2
  • Joseph Hellerstein
    • 1
  • Sam Lightstone
    • 2
  • Matthew Huras
    • 2
  • Victor Chang
    • 2
  1. 1.IBM T.J. Watson Research CenterHawthorneUSA
  2. 2.IBM Toronto LabTorontoCanada

Personalised recommendations