Intelligent Storage for Information Retrieval
Author:
- 15 February 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 214-220
Abstract
The rapid development of Internet, the fast dropping of storage cost and the great improvement of storage capacity have created an environment with enormous amount of data. How to get the data we want and how to manage these huge volumes of data become great challenges. There are a number of existing approaches for solving these challenges including semantic Web, middleware, semantic multimedia information and databases. We are taking a different approach and believe that today's storage systems need to be fundamental changed to meet these challenges. In this paper, we first discuss the possible changes in storage systems. We then present the intelligence to be included in future storage systems. With the intelligence, a new architecture of storage systems can be formed. The essential concepts to be parts of the future storage systems include active storage, object-based storage, autonomic storage, and application-aware storage. We argue the object-based storage device (OSD) concept is the most fundamental. We discuss a number of research issues of intelligent storage.Keywords
This publication has 11 references indexed in Scilit:
- Adaptive replication in peer-to-peer systems2004
- Design of the iSCSI protocol2003
- Integrated QOS management for disk I/O2003
- Disk scheduling with quality of service guarantees2003
- Replication strategies in unstructured peer-to-peer networks2002
- Consistency management in DenoMobile Networks and Applications, 2000
- Cello1998
- Incremental updates of inverted lists for text document retrieval1994
- A linear-time scheme for version reconstructionACM Transactions on Programming Languages and Systems, 1994
- Coda: a highly available file system for a distributed workstation environmentIEEE Transactions on Computers, 1990