期刊名称:International Journal of Computer Trends and Technology
电子版ISSN:2231-2803
出版年度:2011
卷号:2
期号:2-2
出版社:Seventh Sense Research Group
摘要:— The Distributed Computing Systems are significant building blocks in design and implementation of the business critical applications. It is always important for such environments to provide high performance for all kinds of workload variations. There are different solutions identified to deal with the performance problems in Distributed environments. There are recent investigations in exploring the feedback control based solutions in various computing environments. The selfmanaging capabilities are inherent qualities of Adaptive Control and such mechanisms are investigated to address performance problems in computing including distributed computing systems. The objective of Adaptive Controllers is to provide the intelligence in tuning the control parameters that would maintain the system in a desired state. In this paper we explore and identify the applicability of Data Mining based approaches to build Adaptive Controllers in selfmanaging the performance. We implemented a simple Adaptive Control using TimeSeries Analysis. It predicts the frequency of occurrence of statements in a database so that the database driver can cache them according to the predicted values which demonstrates the feasibility of building Data Mining based Adaptive Controller.
关键词:Data Mining; Adaptive Control; Distributed Computing Systems