Active network file system for data mining and multimedia

Storage Area Networks (SAN) exploit advances in networking to make storage remote. SANs enable large scale storage and sharing which in turn facilitate data mining and multimedia applications. These applications are also compute intensive. Active disks make use of the computational power available in the disk to reduce storage traffic. Many of the file system proposals for active disks work at the block level. In this paper we argue for the necessity of filtering at application level. The proposed active network file system (ANFS) is based on the familiar network file system (NFS). We outline the implementation of ANFS in Linux and demonstrate its applicability to data mining and multimedia applications.