Characterizing Process Execution Behavior Using Machine Learning Techniques

A deep knowledge of process execution behavior is very useful in formulating better scheduling techniques. In this paper we study the approaches to the process characterization problem. Also we propose a novel method to characterize and categorize process execution behaviors by using machine learning techniques that learn from previous execution instances of programs. The “Waikato Environment for Knowledge Analysis (Weka)”, an open source machine-learning tool is used for training and performance evaluation. We hope this study shall ultimately help in the design of an advanced knowledge-based process scheduler. Finally we propose an architecture that may improve the PBS scheduler in grid systems