Adaptive Grid Computing for MPI Applications

J. Zhang and W. Meleis (USA)


Grid computing, resource allocation and co-allocation, workflow, task scheduling, message relay.


Our objective is to provide location-, topology-, and administrative-transparentgrid computing for MPI applications, while hiding the physical details of computing platforms and heterogeneous networks from the application developers and users. To achieve this objective, we introduced a new resource allocation model, workflow structures to specify MPI applications involving multiple tasks, and message relay to enable communication across different networks. We developed the SGR framework, which integrates workflow scheduling, task grouping, and message relay services, while hiding resource allocation, heterogeneous networks, and decentralized resource management systems from application developers and users. The SGR system has been implemented on a Globus-enabled computing grid. We created a simulation environment to investigate our model and various schedulers. Using the findings from simulation, we implemented the SGR framework and tested the model’s implementation on a two-cluster grid. We observed that duplication can improve performance by more than 15%, which matches our simulation results. Moreover, we evaluated our new message relay service for cross-site message passing. The test results indicate that although the SGR’s message relay service has some communication overhead, the system is scalable with respect to the number of processes and the message size.

Important Links:

Go Back