Abstract Many workstation‐based distributed systems allow programs to be executed on remote idling machines for effective utilization of system resources. Usually, the control policies in these systems force a remote job be discontinued by the arrival of local jobs to guarantee the autonomy of individual workstations. Therefore, one special concern in the design of such systems is the fault‐tolerant aspects for the execution of remote jobs. In the paper we discuss two control policies of workstation‐based distributed systems, checkpointing and non‐checkpointing policy, which support fault‐tolerant execution of remote jobs on idling workstations. An analytical analysis on the reliability and mean turnaround time of the execution of remote jobs are conducted for both control policies. The optimal time interval between checkpoints in the checkpointing policy is formulated based on the given reliability and overhead of the system. In addition, several sample results derived from these analyses are compared with the outcome of corresponding simulation programs. Some observations of fault‐tolerant features of each control policy are thereupon presented as guidelines for the future development of such workstation‐based distributed systems.

In a ring LAN, performance characteristics such as message transfer time, throughput, and bandwidth utilization depend on its station permutation. As ring LANs adopt fiber optics or other high speed transmission media, extend their geographical coverage, and hence contain more stations, such a dependency becomes more significant. The paper formulates the ring performance optimization problem associated with such a dependency, proves the NP-completeness of the problem, and establishes an integer programming model for solving the problem.>

A token ring local area network, the parallelring, is introduced. The ability to concurrently support multiple communication paths on a single loop in an efficient and fair manner differentiates the parallelring from earlier ring architectures. This is accomplished by using a message destination removal scheme and, hence, allowing more than one station to transmit a variable-length frame at a time. Simulation studies show that the parallelring exhibits better performance in terms of throughput-delay characteristics than the other token rings. In addition, a priority scheme is used in the parallelring to provide messages of different time constraints with different media access priorities so that it can support media access fairness and real-time services. In the parallelring, error detection and recovery are completely decentralized.>