A General Data Grid: Framework and Implementation

Today, data grids have become an important emerging platform for managing and processing a very large amount of data distributed across multiple grid nodes and stored in relational databases. However, there are still obstacles for potential grid users to be involved into the trend and the data grid application development is far from the data grids. While the traditional established data grid architectures are not particularly suitable for the some grid service, such as data replica services, user authentication services, and the Optimal Path selection services. Our project aims to exploit a novel architecture named General Data Grid, which integrates the metadata services, data replica services, java message services, and the Optimal Path selection on Data Grid environment. At the end of this paper, we describe the key implement on the GDGrid and present a simple example application concerning finding the optimal route. Our experiment of GDGrid shows the algorithm of the route selection (Heart-Beat algorithm) is effectively, improve the performance greatly, and afford fault-tolerance management to the great extent.