Browsing by Author "Teo Yong Meng"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Conservative Parallel Simulation of Finite Buffered Multistage Interconnection Networks
    (1994-06-01T00:00:00Z) Tay Seng Chuan; Teo Yong Meng
    Multistage interconnection networks are used in a number of application areas such as parallel computers and high-speed communication systems. As the performance of these systems lies on an efficient design of the interconnection network, a thorough analysis of the network's performance is important. Mathematical analysis so far provides inadequate results and simulation analysis using a uniprocessor usually requires long hours to evaluate large networks. In this paper, parallel simulation technique is used to speedup the execution. The conventional null-message approach to resolving deadlock problem in conservative simulation is based on a lookahead mechanism. For some application domains, unfortunately, the lookahead information is not available. Consequently, parallel simulation using null messages can result in livelock. We propose a deadlock/livelock free scheme using null messages, but without the guaranteed lookahead, to coordinate the simulation. In addition, we investigate different partitioning and transformation techniques for mapping a simulation program onto multicomputers. A flushing mechanism to address the combinatoric explosion of using null-message in conservative simulation is also discussed. Our analysis shows that the proposed flushing mechanism effectively reduces the number of null messages from exponential to linear.
  • No Thumbnail Available
    Item
    Parallel Discrete-Event Simulation on Distributed-Memory Multicomputers
    (1994-03-01T00:00:00Z) Tay S C; Teo Yong Meng
    As computers become more powerful and their use expands, the need to simulate larger and more complex systems in reasonable computing times becomes more important. Parallel simulation can significantly speedup the process. Nevertheless, simulating complex systems on high-speed computers with multiprocessor capabilities is not a trivial task. This paper describes the parallel simulation of multi-station queueing networks on distributed-memory multicomputers. A deadlock-free conservative scheme for organizing parallel simulation is proposed, and implemented on a network of workstations running the PVM software. Preliminary results show that close-to-linear performance speedup can be achieved when the simulated system can be partitioned into processes of sufficiently large grain-size.