A Grid Resource Discovery Scheme Based on Resource Classification

A decentralized grid resource discovery scheme based on resource classification is studied in this paper, in which information nodes with the same type of registered resources are organized into communities, and a DHT based bootstrap network is constructed to navigate between different communities efficiently.Periodical topology maintenance communications in bootstrap network are used to piggyback and disseminate popular data in bootstrap network to achieve better load balance.A greedy search based resource location method is also introduced to identify qualified resource providers with good qualities inside communities, which improves scheduling efficiency.The performance improvement is evaluated by simulation under different cases, and overhead is also studied.