Font Size: a A A

The Research On Key Technology For Flood Catastrophe Simulation

Posted on:2009-03-02Degree:MasterType:Thesis
Country:ChinaCandidate:Q L LiFull Text:PDF
GTID:2178360272491968Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the development of computer and Internet , the research on the engineering dynamics catastrophe simulation system in grid computing environment is becoming more and more important on the engineering community and academic field. In our country, there has been some initiating research on software theories and technologies of the catastrophe simulation. Attending to the importance on the analysis, simulation and prediction of the flood catastrophe, this paper combines middleware software basing on the grid with the need of the catasphe simulation, and then study on the flood catastrophe heterogeneous data sources treatment, the special parallel file system treatment, the parallel distributed job cooperative scheduling algorithm and so on.First, after in-depth analysis the existing parall filesystem and netCDF data format,this paper design a efficient, reliable parallel I/O filesystem base on netCDF for the framework,the filesystem uses a new data scatter algorithm,Comparing to the existing parall filesystem,the filesystem improve greatly in fault-tolerant and data backup/recovery, The experiment demonstrates can thoroughly support Simple data processing patterns and get a high parall speedup.Second, we present an integrated software framework for parallel data processing on pc cluster. A modern and feature-rich graphical user interface (GUI) was developed for the system allowing edit and submission of parallel processing flows and interaction with running jobs. Multiple jobs can be executed on the pc cluster and controlled from the same GUI. The framework provides a module integrated development environment (MIDE) to pc cluster module developer. System monitor, database, memory management and the MPI functionalities are provided to developer as API, which can significantly simplify the code development for the system. A multi-user development and processing environment is available. The system can greatly reduce time for transplanting serial processing modules to pc cluster, and get a high speedup.Last, based on the researches in trust relationship in Grid, the paper redesigns the classical Grid scheduling algorithm: MCT and Min-min, and the Trust MCT algorithm and the Trust Min-min algorithm are proposed. According to the different characters of the MCT and the Min-min, the trust value zone is divided into several regions by different rules and the tasks are mapped to the resources in the same trust value region. Simulation results show that the miss ratio of tasks execution is reduced, therefore the completion time is decreased.
Keywords/Search Tags:parallel data processing, Parallel I/O, Filesystem, task scheduling
PDF Full Text Request
Related items