The Development of the Resource Broker of Desktop Grid Federation for Tree Search Applications

Abstract

As the desktop grid federation has facilitated many large scale e-Science projects for the resource restricted organizations, people start thinking about a wide variety of applications. This paper studies the desktop grids of artificial intelligence computation applications that require dynamic and adaptive task control. This paper discusses the development of the resource broker that assigns tasks based on user credits as well as resource utilization for the users in and across the organizations. The result demonstrates that, in a desktop grid federation using the well-designed broker, a number of organizations can perform large scale applications efficiently via resource sharing.