Agents Based Intelligent Grid Matchmaking System for EDA Tools on a Cluster Grid

Abstract

Agent based technologies have become more and more widespread during the last few years and the number of practical applications has noticeable increased, especially in the new Grid Computing paradigm. Grid Computing system implementations are easily transferable into Cluster Grid environments. So, the main purpose of this research is to validate integration functionalities and performances of an Agent based Intelligent Grid Matchmaking System into a Cluster Grid. For this reason it has been created anAgent System for automatic matchmaking between specific jobs of Electronic Design Automation (EDA) simulation tools with a specific type of hardware architecture node, pertaining to the Cluster Grid. Experimental tests have focused on evaluation of a specific job completion time analyzing its probability density function.