CPU Usage Pattern Discovery Using Suffix Tree

In the dynamic resource-sharing environment, resource availability often varies from time to time. Resource prediction can be used to enhance scheduler effectiveness on their scheduling strategies and resource allocation. The prediction results can also be used by the applications to adjust themselves to suit the resource availability to get better performance. In this paper, we use a suffix tree to discover CPU usage patterns to find opportunities for exploiting the available CPU resources. We introduced a prediction strategy that uses discovered frequent patterns to predict CPU load. We defined that a CPU usage behavior as a set of CPU usage patterns. Our experiment results showed that CPU usage does exhibit certain behavior and our model is capable of discovering the usage and utilized it to perform prediction. The discovered patterns are interesting because some of the discovered cyclic patterns seem to be related to users' usage behaviour. In order to justify our model prediction capability, we compared our prediction model with the state-of-the-art methods such as Network Weather Services

[1]  Lingyun Yang,et al.  Conservative Scheduling: Using Predicted Variance to Improve Scheduling Decisions in Dynamic Environments , 2003, ACM/IEEE SC 2003 Conference (SC'03).

[2]  Mor Harchol-Balter,et al.  Exploiting process lifetime distributions for dynamic load balancing , 1995, SIGMETRICS.

[3]  Richard Wolski,et al.  The network weather service: a distributed resource performance forecasting service for metacomputing , 1999, Future Gener. Comput. Syst..

[4]  Ian T. Foster,et al.  Homeostatic and tendency-based CPU load predictions , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[5]  Matt W. Mutka.  An examination of strategies for estimating capacity to share among private workstations , 1991, SIGSMALL '91.

[6]  Ramez Elmasri,et al.  Fundamentals of Database Systems, 5th Edition , 2006 .

[7]  Eamonn J. Keogh,et al.  Dimensionality Reduction for Fast Similarity Search in Large Time Series Databases , 2001, Knowledge and Information Systems.

[8]  Richard Wolski,et al.  Dynamically forecasting network performance using the Network Weather Service , 1998, Cluster Computing.

[9]  Ian Foster,et al.  The Grid 2 - Blueprint for a New Computing Infrastructure, Second Edition , 1998, The Grid 2, 2nd Edition.

[10]  Peter A. Dinda,et al.  The statistical properties of host load , 1999, Sci. Program..

[11]  Domenico Ferrari,et al.  An Empirical Investigation of Load Indices for Load Balancing Applications , 1987, Performance.

[12]  Peter A. Dinda,et al.  An evaluation of linear models for host load prediction , 1999, Proceedings. The Eighth International Symposium on High Performance Distributed Computing (Cat. No.99TH8469).

[13]  George Luger,et al.  Artificial Intelligence: Structures and Strategies for Complex Problem Solving (5th Edition) , 2004 .

[14]  Eamonn J. Keogh,et al.  A symbolic representation of time series, with implications for streaming algorithms , 2003, DMKD '03.

[15]  Peter A. Dinda,et al.  Host load prediction using linear models , 2000, Cluster Computing.

[16]  G. Kiczales,et al.  Proceedings the , 1997 .

[17]  Jessica Lin,et al.  Visually mining and monitoring massive time series , 2004, KDD.