Load Balancing Distributing File System Servers
Alexandra Glagoleva, Archana S. Sathaye · IGI Global eBooks · 2011
In this paper, we present a new methodology for managing read-write filesets across multiple file servers of a Distributed File System, thus balancing the load of file access requests across servers. The proposed methodology is based on a rule-based data mining technique and graph theory algorithms. The rule-based technique generates rules from access request data to identify present file access patterns in the system. We then use the rules, graph analysis and statistical information (usage and size of the filesets) to relocate the filesets between different file servers. The algorithm for fileset relocation is based on the graph coloring problem. We tested our algorithms on data collected for five months on DFS file servers in a production environment. Experiments with the data show that our methodology can make intelligent decisions about file system transfers in order to balance the access request load across DFS servers.