| Abstract in Chinese | 第1-9页 |
| Abstract | 第9-12页 |
| 1 Introduction | 第12-15页 |
| ·The Prospect of Network Parallel Computing | 第12页 |
| ·PVM and Its Characteristics | 第12页 |
| ·The Current Research Status | 第12-13页 |
| ·Main Work in This Study | 第13-14页 |
| ·The Significance of This Study | 第14-15页 |
| 2 Building PVM on Wi1132 Network Parallel Environments | 第15-16页 |
| ·Installing and Setting PVM on Wi1132 System | 第15页 |
| ·Starting PVM and Configuring Virtual Machine | 第15-16页 |
| 3 Parallel Programming Under PVM on Win32 Network Parallel Environments | 第16-20页 |
| ·Decomposing Task | 第16页 |
| ·Designing Communication | 第16-17页 |
| ·Choosing Programming Models | 第17-18页 |
| ·Writing, Compiling and Executing PVM Parallel Programs | 第18-19页 |
| ·Debugging PVM Parallel Programs | 第19页 |
| ·Optimizing PVM Parallel Numerical Programs | 第19-20页 |
| 4 Factors Affecting the Performance of PVM Parallel Programs | 第20-26页 |
| ·Load-Balancing | 第20-21页 |
| ·The Importance of Load-Balancing | 第20页 |
| ·Methods for Load-Balancing | 第20-21页 |
| ·Communication Costs | 第21-23页 |
| ·Communication Costs Are the Key Factor Affecting the Performance of PVM Parallel Programs | 第21-23页 |
| ·Techniques of Decreasing Communication Costs | 第23页 |
| ·Network Performance | 第23-24页 |
| ·Network Performance Affects the Performance of PVM Parallel Programs | 第23-24页 |
| ·How to Improve Network Performance | 第24页 |
| ·Task Granularity and the Number of Processors | 第24-25页 |
| ·Task Granularity and the Number of Processors Affect Speedup | 第24页 |
| ·A Strategy for Optimizing the Number of Processors | 第24-25页 |
| ·Memory Problems | 第25-26页 |
| 5 Network Parallel Computing Practices | 第26-38页 |
| ·A Parallel Algorithm for Solving Dense Linear Equations | 第26-30页 |
| ·Basic Idea | 第26-27页 |
| ·Numerical Experiment | 第27-29页 |
| ·Analysis and Enlightenment of the Experimental Results | 第29-30页 |
| ·A Piecewise Gauss-Seidel Parallel Iterative Algorithm for Solving Linear Equations | 第30-34页 |
| ·Basic Idea | 第30-32页 |
| ·Numerical Experiment | 第32-33页 |
| ·Analysis and Enlightenment of the Experimental Results | 第33-34页 |
| ·A Parallel Subspace Iterative Algorithm for Eigenvalue Problems and Its Implementation on LAN | 第34-38页 |
| ·Serial Subspace Iterative Algorithm | 第34-35页 |
| ·Basic Idea of Parallel Algorithm | 第35-36页 |
| ·Numerical Experiment | 第36-38页 |
| ·Analysis and Enlightenment of the Experimental Results | 第38页 |
| 6 Conclusion | 第38-40页 |
| Bibliography | 第40-43页 |
| Appendix: Published Papers of the Author | 第43-44页 |
| Acknowledgements | 第44-45页 |
| 原创性声明 | 第45页 |
| 关于学位论文使用授权的声明 | 第45页 |
