An efficient parallel storage scheme in data grid environment

摘要

It is important to improve data reliability and data accessing efficiency for data-intensive applications in data grid environment. We propose an IDA-based parallel storage scheme in Data Grid environment. The scheme partitions a data file into unrecognizable blocks and distributes them across storage nodes according to user profile and system conditions. A subset of blocks, which could be downloaded in parallel to the remote client, is required to reconstruct the data file. The scheme can be deployed on top of current grid middleware. Experimental results and analysis show that the IDA-based parallel storage scheme has better data reliability and data accessing performance than the existing data replication methods.