Algorithm Research On Grid Resource Scheduling Based On Computing Market Model

Resource management system is the foundation and core of grid. Because Grid is distributed, heterogeneous and dynamic, it is much more difficult to realize resource sharing in Grid than in traditional systems, traditional methods of resource management will not function well in Grid environment. Therefore, the model of the computer economic was used for resolving resource management and scheduling.Economic-based Grid is a hot problem. In recent ten years, much research about the application of Economics method in distributed environments is carried out which lay a fundamental basis for the development of Economic-based Grid. Economics especially Micro-Economics carries out deep research in the field of discreet space resource allocation which integrates with the Economic-based Grid. An important factor that distinguishes the Economic-based Grid from distributed system is that the Economic-based Grid is influenced by the psychological activity of the participant. Reasonable usage of certain psychological model has a great influence on the construction of Economic-based Grid intelligent proxy.In this paper, the basic concepts and grid technologies are introduced;second, the popular grid infrastructures are described;then resource management systems are discussed, focused on Nimrod-G grid resource management system: furthermore, a gird resource trading management algorithm is offered;also, a new time cost optimization algorithm is presented based on the classical price optimization and time optimization algorithms. We also offered an algorithm which balanced the price optimization and time optimization algorithms. Finally, with the help of GridSim toolkit, we emulated our algorithms.The experiment results show that the new time price algorithm in this paper is feasible. It decreases the price and don't change the time algorithm's time. And the balance algorithms have more practical value.