The Realization Of The Content Addressable Storage System

A study shows that the capacity of the global data storage has increased by one time on the basis of the original every 18 months . At the same time, IDC points that 75% of the growth comes from fixed content data. The so-called fixed content data is not changed after the creation of data such as office documents, E-mail, television news, medical images . Fixed content is unstructured data,which accounts for 85% of the enterprise's total data volume. The number of the fixed content of the enterprise increases to tens of thousands with time going by.It's difficult to find the content needed during the traditional file system storage,and some of them even can not be understood . Therefore, enterprises are faced with the tremendous challenges of data storage and managementIn recent years, data storage technologies, especially networked storage technology,have developed fast. The storage technologies which use the data block as interface, such as DAS, SAN,have a direct access to the storage devices,and have a high storage efficiency . The technologies which use document storage as interface such as NAS, provide an effective mechanism for the use of file-sharing. However, neither the Direct Connect storage (DAS), Network Attached Storage (NAS) and Storage Area Network (SAN), nor the new IP-based SAN, SAN and NAS combination of unified storage , is suitable to unstructured data storage. Object Storage ,comes up a new storage interface - Object Interface. Object Storage Architecture has the advantages of block interfaces as well as the document storage system interface .For data storage, it can allow data to have a direct access to the storage medium hardware in the way of the data block 's, and it has a high storage efficiency. In the meantime, four data management, through other attributes of the packaged object,it can let data be under sharing and management between different platforms likes other documounts . In Object storage, Content Addressed Storage is the model which has been successfully used in a fixed content storage. CAS provides a brand new effective information management way which can greatly reduce the difficult of the application software for managing storage media manipulation and storage of information on the physical address . this has irreplaceable advantages for the storage of the fixed content.At present, CAS has now become one focus of the content management and storage rearchThis article makes a detailed analysis of CAS technical and the ideological content addressable storage, aiming at achieving a prototype of fixed content storage system (FCSS). FCSS offers object-based storage as high-speed indexing, disk space allocation program and quick access interface: Basing on the content of RED-BLACK TREE (Hash) and data (files) physical location mapping function, module allocation of disk storage space object, the API function provides storage interface ; and it gives description of the resources corresponding to the storage resources, documents, users to facilitate the realization of the efficient exchange between data storage and application; For the upper storage management, a number of storage management interfaces are also reached to help companies achieve rapid fixed content storage and management strategies . FCSS makes the organic integrations of content-based storage and management become true. It has greatly enhanced the efficiency of fixed content retrieval and access . It also can provide the enterprise fixed content storage with an efficient solution.