This paper presents a Probabilistic Preemptive (PP) scheduling discipline for service differentiation in Optical Burst Switching (OBS) networks. The main idea of PP is that a probabilistic parameter is added to the preemptive scheduling discipline. With PP, a high priority burst preempts a low priority burst in a probabilistic manner. By changing the probabilistic parameter, PP can achieve flex...

While it is widely believed that preemptibility is a necessary requirement for developing real-time software, there are additional costs involved with preemptive scheduling, as compared to non-preemptive scheduling. Furthermore, in the context of fixed-priority scheduling, feasibility of a task set with nonpreemptive scheduling does not imply feasibility with preemptive scheduling (and vice-ver...

In this paper, we analyze a discrete-time GI-G1, Geo2-1 preemptive resume priority queue. We consider two classes of packets which have to be served, where one class has preemptive resume priority over the other. The high-priority class contains packets with generally distributed service times while the low-priority packets are assumed to have geometrically distributed service times. We show th...

The integration of different types of traffic in packet-based networks spawns the need for traffic differentiation. In this tutorial paper, we present some analytical techniques to tackle discrete-time queueing systems with priority scheduling. We investigate both preemptive (resume and repeat) and non-preemptive priority scheduling disciplines. Two classes of traffic are considered, high-prior...

Real-time signal processing applications can be described naturally with dataflow graphs. The systems we consider have a mix of real-time and non-real-time processing, where independent dataflow graphs represent tasks and individual dataflow actors are subtasks. Rate-monotonic scheduling is optimal for fixed-priority, preemptive scheduling of periodic tasks. Priority inheritance protocols exten...

We give a closed-form expression for the discounted weighted queue length and switching costs of a two-class single-server queueing model under a preemptive priority rule. These expressions are used to do a single step of policy iteration in a polling model with a dynamically controlled switching rule, starting from the preemptive priority rule. Numerical experiments show that this leads to a p...

While it is widely believed that preemptability is a necessary requirement for developing real-time software, there are additional costs involved with preemptive scheduling, as compared to non-preemptive scheduling. Furthermore, in the context of fixed-priority scheduling, feasibility of a task set with nonpreemptive scheduling does not imply feasibility with preemptive scheduling (and vice-ver...

This paper extends previous research on congested service facilities to generalized service distributions, a significant extension given the limitations of exponential distributions for networked computer job modeling. Building on the framework first presented in Mendelson and Whang (1990), we present fundamental theorems for non-priority M/G/1 queues, nonpreemptive M/G/1 queues, and preemptive...

In this paper, we analyze a discrete-time GI-Geo-1 preemptive resume priority queue. We consider two classes of packets which have to be served, where one class has preemptive resume priority over the other. We show that the use of generating functions is beneficial for analyzing the system contents and packet delay of both classes. Moments and (approximate) tail probabilities of system content...

Resource-monopolizing Denial-of-Service (DoS) is one form of malicious code attack. An attacking code exclusively uses shared resources and attempts to drop the responsiveness of the machine so low that it is practically useless. Resource-monopolizing DoS is difficult to prevent in commodity operating systems, because they allow every process to compete for shared resources in an uncontrolled m...