Performance evaluation of the networks of workstations for parallel processing applications

The computing power of the local area network (LAN) has been explored for distributed task execution. An application program represented in a task graph format is scheduled by the user workstation (master) for parallel execution in the idle or lightly loaded processors (slaves) in a LAN. The performance of this operation is characterized by the network speedup considering scheduling time, network load and size, communication time, TCP/IP communication overhead, task execution, and synchronization time. The sequential and parallel execution times and the performance degradation factors were measured during the implementation for various network loads. The measurement and theoretical results have shown that the system performance is degraded mostly by the heavily loaded nodes, the TCP/IP overhead (0.2 s), and the network size.<
>