Implementing a performance forecasting system for metacomputing: the Network Weather Service

ABSTRACT

In this paper we describe the design and implementation of a system called the Network Weather Service (NWS) that takes periodic measurements of deliverable resource performance from distributed networked resources, and uses numerical models to dynamically generate forecasts of future performance levels. These performance forecasts, along with measures of performance fluctuation (e.g. the mean square prediction error) and forecast lifetime that the NWS generates, are made available to schedulers and other resource management mechanisms at runtime so that they may determine the quality-of-service that will be available from each resource.We describe the architecture of the NWS and implementations that we have developed and are currently deploying for the Legion [13] and Globus/Nexus [7] metacomputing infrastructures. We also detail NWS forecasts of resource performance using both the Legion and Globus/Nexus implementations. Our results show that simple forecasting techniques substantially outperform measurements of current conditions (commonly used to gauge resource availability and load) in terms of prediction accuracy. In addition, the techniques we have employed are almost as accurate as substantially more complex modeling methods. We compare our techniques to a sophisticated time-series analysis system in terms of forecasting accuracy and computational complexity.

REFERENCES

• 1: AppLeS. http://www-cse.ucsd.edu/groups/hpcl/apples/apples.html.
• 2: Francine Berman , Richard Wolski, Scheduling From the Perspective of the Application, Proceedings of the High Performance Distributed Computing (HPDC '96), p.100, August 06-09, 1996
• 3: Francine D. Berman , Rich Wolski , Silvia Figueira , Jennifer Schopf , Gary Shao, Application-level scheduling on distributed heterogeneous networks, Proceedings of the 1996 ACM/IEEE conference on Supercomputing (CDROM), p.39-es, January 01-01, 1996, Pittsburgh, Pennsylvania, United States [doi>10.1145/369028.369109]
• 4: M. V. Devarakonda , R. K. Iyer, Predictability of Process Resource Usage: A Measurement-Based Study on UNIX, IEEE Transactions on Software Engineering, v.15 n.12, p.1579-1586, December 1989 [doi>10.1109/32.58769]
• 5: Silvia M. Figueira , Francine Berman, Modeling the Effects of Contention on the Performance of Heterogeneous Applications, Proceedings of the High Performance Distributed Computing (HPDC '96), p.392, August 06-09, 1996
• 6: Ian Foster , Jonathan Geisler , Carl Kesselman , Steven Tuecke, Managing multiple communication methods in high-performance networked computing systems, Journal of Parallel and Distributed Computing, v.40 n.1, p.35-48, Jan. 10, 1997 [doi>10.1006/jpdc.1996.1266]
• 7: I. Foster and C. Kesselman. Globus: A metacomputing infrastructure toolkit. International Journal of Supercomputer Applications, 1997. to appear.
• 8: Ian Foster , Carl Kesselman , Steven Tuecke, The Nexus approach to integrating multithreading and communication, Journal of Parallel and Distributed Computing, v.37 n.1, p.70-82, Aug. 25, 1996 [doi>10.1006/jpdc.1996.0108]
• 9: R. Gallant and G. Tauchen. Snp: A program for nonparametric time series analysis. In http://www.econ.duke.edu/Papers/Abstracts/abstract.95.26.html.
• 10: R. Gallant and G. Tauchen. Seminonparametric estimation of conditionally constrained heterogeneous processes: Asset pricing applications. Econometrica 57, pages 1091-1120, 1989.
• 11: C. Granger and P. Newbold. Forecasting Economic Time Series. Academic Press, 1986.
• 12: Andrew S. Grimshaw, Easy-to-Use Object-Oriented Parallel Processing with Mentat, Computer, v.26 n.5, p.39-51, May 1993 [doi>10.1109/2.211896]
• 13: Andrew S. Grimshaw , William A. Wulf , James C. French , Alfred C. Weaver , Paul F. Reynolds, Jr., Legion: The Next Logical Step Toward a Nationwide Virtual Computer, University of Virginia, Charlottesville, VA, 1994
• 14: Will E. Leland , Murad S. Taqqu , Walter Willinger , Daniel V. Wilson, On the self-similar nature of Ethernet traffic (extended version), IEEE/ACM Transactions on Networking (TON), v.2 n.1, p.1-15, Feb. 1994 [doi>10.1109/90.282603]
• 15: Netperf. http://www.cup.hp.com/netperf/netperfpage.html.
• 16: R. Wolski. Dynamically forecasting network performance using the network weather service. Technical Report TR-CS96-494, U.C. San Diego, October 1996. available from http://www.cs.ucsd.edu/users/rich/publications.html.
• 17: R. Wolski, Forecasting network performance to support dynamic scheduling using the network weather service, Proceedings of the 6th International Symposium on High Performance Distributed Computing (HPDC '97), p.316, August 05-08, 1997