Applying Preference-Based Fuzzy Clustering Technology to Improve Grid Resources Selection Performance

How to select proper resources for applications is a critical and complex issue which results from applications´ multiple requirements and grid resources´ features of being distributed, heterogeneous, autonomous, and dynamically available. This paper proposes a preference-based fuzzy clustering method to improve grid resources selection performance. It applies fuzzy clustering technology according to application preference to create resource clusters, then the clusters are evaluated and proper one(s) can be selected. This method can meet applications´ requirements and reduce the cost on choosing resources. Furthermore, since resources are classified by different application preferences, this method can also avoid heavy loads´ centralizing on only a few resources so as to improve load balance in grid environments.