Resource Load Balancing Based on Multi-agent in ServiceBSP Model
ICCS '07 Proceedings of the 7th international conference on Computational Science, Part III: ICCS 2007
Hi-index | 0.00 |
The Bulk Synchronous Parallelism(BSP) model is useful in developing scalable software for parallel computation. To adapt the BSP model to the Grid environment, the BSP-G, a BSP development library based on the services of Globus Toolkit has been proposed. In this paper, we further the work of BSP-G and introduce an agent assisted ServiceBSP model. The agent in this model could make a QoS-aware selection of computational services and monitor the status of those services. During the selection phase, the agent can handle the diversity of QoS information of services by referring to an ontology repository of Quality of Service(QoS). An experiment shows that the agent could provide users with satisfied supersteps by selecting Grid services on QoS demand.