Font Size: a A A

Research Of Grid Architecture Based On High Available Data Computing

Posted on:2007-02-12Degree:MasterType:Thesis
Country:ChinaCandidate:X R FengFull Text:PDF
GTID:2178360182998062Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
The grid technology is a hotspot in HPC research. The architecture design is the basis for constructing the grid system. First, this article introduces the definition, condition of study and applied circumstance of grid in domestic and international, application of cluster system, the research tread of high end computer in domestic and international. It introduces the two popular grid structures nowadays, and analyzes each structure's virtues and shortcomings.Computational grid is a more variable and unreliable computing environment. It should provide appropriate availability assurance for the applications. How to measure the availability is the key issue for providing high availability service. In this paper, the characteristics of the availability of computational grid are analyzed and it is proposed that the availability of computational grid should be measured from the application's point of view. Then a measurement model for the availability of applications in the computational grid is presented. In the model the overall availability of a grid application is measured by using a serial RBD, for the tasks on a grid node, the availability is measured by using a probabilistic model.With the extension of the grid scale, the cost to manage and to maintain the global resources information will take up a higher percentage of the total cost, causing a bottleneck in cost of the grid. So this article introduces a new model of manage, resources pond-memory stick, it can schedule the task to accommodation computing resources.According to the research contents that the paper have mentioned, this article puts forward the grid architecture based on high available data computing, first, it introduces the detection of the resources and high available resources in the grid, use these resources to build a frame of high available data computing, and, it introduces the process of data computing. Because the environment of grid is complications, it is hard to build the environment of high available grid computing, the article from two standpoints to verify the grid architecture based on high availability of data computing, first, it is build the native system based on high availability of data computing, and, it analyzes the high availability from theories about grid environment. Last, according the experiment, it gets the high availability of this architecture.
Keywords/Search Tags:Cluster, High availability, Task scheduling, Manage resources, Architecture, Grid computing
PDF Full Text Request
Related items