I/O Processor Allocation for Mesh Cluster Computers

As cluster systems become increasingly popular, more and more parallel applications require need not only com-puting power but also significant I/O performance. How-ever, the I/O subsystem has become the bottleneck of the overall system performance for years due to slower im-provement of the second storage devices. In recent years parallel I/O has drawn an increasing attention as a promis-ing approach to eliminate this bottleneck. To improve I/O efficiency of a cluster system computation tasks must be carefully assigned to processors, so that the communica-tion overheads within the group the processors of the task, and those I/O traffics that connect processors of the task to I/O system are both optimized. Earlier processor allo-cation strategies considered the optimization of communi-cation traffic or I/O traffic only. Since both the communi-cation and I/O traffic can cause network contention, we de-velop a set of binary tree based algorithms to address the is-sues of both communication and I/O traffics simultaneously. The experimental results indicate that for tasks that have different mixture of communication and I/O traffics, our al-gorithms have very good performance in terms of overall parallel I/O efficiency. We also developed two mathemati-cal evaluating criteria – “compactness ” and “spatial com-pactness”, to determine the fitness of allocation algorithms in terms of geometrical adjacency of processors. 1