In this paper, analysis of non-preemptive priority queues with multiple servers and multiple priority classes is presented. It is assumed that the service times – for all priority classes – are identically and exponentially distributed to simplify the complexity of the residual service time mathematical expression to an extent will enable calculating the average customer waiting time. The paper...

AbsfructThe IEEE 802.6 standard for metropolitan area networks does not provide multiple priority traffic for connectionless data services. A priority mechanism that was considered in earlier versions of the standard showed to be not effective. As of now, there exists no protocol for multiple access dual bus networks that is able to implement preemptive priorities and, at the same time, can sat...

In this paper we present the problem of optimal priority assignments in fixed priority preemptive single processor systems where tasks have probabilistic arrivals and execution times. We show that Rate Monotic is not optimal for our problem.

Consider a two-class unobservable priority queue, with Poisson arrivals, generally distributed service, and strategic customers. Customers are charged fee when joining the premium class. We analyze maximum revenue achievable under non-preemptive (NP) preemptive-resume (PR) policies, show that provider is always better off implementing PR policy. Further, sometimes achieved only fraction of cust...

For the M/G/1 model, we look into a preemptive priority scheme in which the priority level is decided by a lottery. Such a scheme has no effect on the mean waiting time in the non-preemptive case (in comparison with the First Come First Served (FCFS) regime, for example). This is not the case when priority comes with preemption. We derived the resulting mean waiting time (which is invariant wit...

Although many multiprocessor resource sharing protocols have been proposed, their impacts on the schedulability of real-time tasks are largely ignored in most of the existing literature. Recently, work has been done to integrate queue locks (FIFO-queue-based non-preemptive spin locks) with multiprocessor schedulability analysis but the techniques used introduce a substantial amount of pessimism...

Fixed Priority Scheduling with Deferred Preemption (FPDS) offers a balance between Fixed Priority Non-preemptive Scheduling (FPNS) and Fixed Priority Preemptive Scheduling (FPPS), by allowing preemptions only at specified preemption points. It provides finer grained preemptions than FPNS, improving the schedulability of higher priority tasks, and a coarser grain preemptions than FPPS, reducing ...

We derive recursive formulas for computing the state probabilities in priority queueing systems (preemptive and non-preemptive). The derivation is based only on the general structure of the generating function involved, and thus is simpler and more general than previous methods. We also discuss the applications of our method to other queueing systems.