A policy-based sensor selection system with goal oriented singular value decomposition technique

Abstract

Modern data center relies on the use of numerous sensors to collect system and environment data to guide the optimal operations of the data centers to meet the desired business goals. The amount of sensors and monitored systems in the data center could be huge, and it includes many types of sensors generating seemingly unrelated data. This presents a challenge to the efficient utilization of these data to determine problems and how well the data center will deliver to the business. In this demo, we present an automatic policy selection management approach with goal guided singular value decomposition technique integrated into IBM Tivoli Monitoring System (ITM) to efficiently utilize the sensor data. © 2009 IEEE.