The Application Of Stock Trading Model In Grid Computing

Grid computing is a fast developing wide area network computing technique. It enables the coordinated use of geographically distributed resources in the absence of central control, omniscience and strong trust relationships. However, the management of resources in the Grid environment is more complex than in general ones, as resources are autonomous, owned by different individuals or organizations, and more (geographically) distributed, heterogeneous in nature. We must develop a suitable resource management model to get a good performance from Grid.In the Grid environment, the most important models of grid resource management architecture are Hierarchical model and Computational Market/Economy Model. The work of this dissertation is based on Computational Market/Economy Model:(1)Surveying the major Grid projects in the world and the resource management schemes adopted in these projectsAfter an introduction of Grid architecture, the dissertation surveys, analyses and compares the major Grid projects in the world. Then, we summarize the resource management architectures, machine and resource organization models, and scheduling models that can be adopted in Grid environments.(2) Designing a simulating stock market model of resource allocation in Grid computing environmentsThe dissertation presents a distributed simulating stock market model of resource allocation in Grid environments, and designs the major components of the trading market. Using parameterized resources, we can define the resource demands of tasks more precisely by many parametersincluding price. Finally we design the corresponding market trading mechanism based on the concept of parameterized resources.