Sim-MSC adopts new hierarchical performance model to predicts the performance of applications on multi-core-SMP-clusters.

Hierarchical Parallel Programming Model and Parallelization and Optimization Techniques Based on SMP Cluster;

Research and Application of MPI and OpenMP Hybrid Programming Model Based on SMP Clusters;

Nowadays, the most popular parallel computer architectures in the world are SMP, MPP, COW and SMP Cluster, etc.

In order to give a full play of SMP clusters,this research studied the method to improve the performance of Fox algorithm for parallel matrix multiplication exploiting optimization on single processor,OpenMP and MPI,which involves the levels of instruction,shared memory and distributed memory respectively.

Based on two-level parallel structures of SMP cluster,a two-level parallel technique is developed,which uses a Message Passing Interface MPI between the nodes and Compiler directives OpenMP inside the nodes.

On the basis of the p arallel I/O model framework based on wave-pushing type,the models of each da ta channel are built under the system architecture of SMP cluster based-linux .

Research and experiment show that local blocking technique and way of data distribution have remarkable effect on performance,and the aggregation optimization is effective on quad-core cluster,especially for the case of load-balance data distribution.

A multi-cluster computing model is introduced.

This paper studies the management model of multi-cluster, presumes two methods: user-decided and system-decided, and stabiles a MCJSS structure based on PBS.

Facing this new trend of development for the internet, many big corporations release their products of grid computing with the main concern on the grid of multi-cluster.