Abstract
In this paper, we describes the design and implementation of Grid-based Parallel and Distributed Simulation environment(GPDS). GPDS not only addresses the problems that it is difficult for parallel and distributed application to achieve its expected performance, because of some obstacles such as deficient computing powers, weakness in fault and security problem, but also supports scalability using Grid technologies. GPDS supports a 3-tier architecture which consists of clients at front end, interaction servers at the middle, and a network of computing resources at back-end including DataBase, which provides three services: Automatic Distribution Service, Dynamic Migration Service and Security Service, designed by UML-based diagrams such like class diagram and interaction diagram. The GPDS has been implemented as Grid Agent(GA) and Simulation Agent(SA) using C++. The object-oriented design and implementation of GA and SA in GPDS provides users with modification, extensibility, flexibility through abstraction, encapsulation and inheritance.
| Original language | English |
|---|---|
| Pages (from-to) | 873-880 |
| Number of pages | 8 |
| Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Volume | 3038 |
| Publication status | Published - 2004 Dec 1 |
Fingerprint
ASJC Scopus subject areas
- Computer Science(all)
- Biochemistry, Genetics and Molecular Biology(all)
- Theoretical Computer Science