Distributed archiving storage system based on CAS

Abstract

With the rapid growth of data, the need for archival storage is the increasing volume of material converted into the digital domain. The archival data is not removed rather closely referenced by its appliance, will only continue to grow; archival storage system with scalability and powerful search ability is desired. In this paper, a novel distributed archival storage based on CAS is brought. First, a network protocol based on standard HTTP defining the communication between CAS server and CAS client is designed and developed; data stored in CAS can be mapped into a secure URL. Each CAS client is constructed with WEB server and database server, this make CAS can be integrated with application easily and tightly. Second, an object oriented data model with good management and strong search ability is established, by integrating the metadata of files and their relationship. Users can browse and search archival data through WEB. With the help of CAS protocol based on HTTP and data object model, a distributed archival storage system based on content and object is achieved with powerful content navigation and search ability.