Using hypergraph-based clustering scheme for traversal prediction in virtual environments

In many virtual environments (VE) applications, the size of the database is not only extremely large, it is also growing rapidly. Even for relatively simple searches, the time required to move the data off storage media is expensive. However, object correlations are common semantic patterns in VE. They can be exploited for improve the effectiveness of storage caching, prefetching, data layout, and disk scheduling. However, little approaches for discovering object correlations in VE to improve the performance of storage systems. In this paper, we develop a class of view-based projection-generation method for mining various frequent sequential traversal patterns in the VE. The frequent sequential traversal patterns are used to predict the user navigation behavior. Furthermore, the hypergraph-based clustering scheme can help reduce disk access time with proper placement patterns into disk blocks. Finally, we have done extensive experiments to demonstrate how these proposed techniques not only significantly cut down disk access time, but also enhance the accuracy of data prefetching.