The problem of multiprogram scheduling on a single processor is studied from the viewpoint...

Traditional processor scheduling mechanisms in operating systems are fairly rigid, often supportingonly one fixed scheduling policy, or, at most, a few "scheduling classes" whose implementations are closely tied together in the OS kernel. This paper presents CPU inheritance scheduling, a novel processor scheduling framework in which arbitrary threads can act as schedulers for other threads. Widely different scheduling policies can be implemented under the framework, and many different policies can coexist in a single system, providing much greater scheduling flexibility. Modular, hierarchical control can be provided over the processor utilization of arbitrary administrative domains, such as processes, jobs, users, and groups, and the CPU resources consumed can be accounted for and attributed accurately. Applications, as well as the OS, can implement customized local scheduling policies; the framework ensures that all the different policies work together logically and predictably. As a ...

We describe extensions to lottery scheduling, a proportional -share resource management algorithm, to provide the performance assurances present in traditional nonreal time process schedulers. Lottery scheduling enables flexible control over relative process execution rates with a ticket abstraction and provides load insulation among groups of processes using currencies. We first show that a straightforward implementation of lottery scheduling does not provide the responsiveness for a mixed interactive and CPU-bound workload offered by the decay usage priority scheduler of the FreeBSD operating system. Moreover, standard lottery scheduling ignores kernel priorities used in the FreeBSD scheduler to reduce kernel lock contention. In this paper, we show how to use dynamic ticket adjustments to incorporate into a lottery scheduler the specializations present in the FreeBSD scheduler to improve interactive response time and reduce kernel lock contention. We achieve this while maintaining lo...

Computer systems researchers have begun to apply the Foreground-Background (FB) schedul-ing discipline to a variety of applications, and as a result, there has been a resurgence in theo-retical research studying FB. In this paper, we bring together results from both of these research streams to provide a survey of state-of-the-art theoretical results characterizing the performance of FB. Our emphasis throughout is on the impact of these results on computer systems.

Abstract not found

normalized queuing delay due to buffering is equal to 1.25 character-service times. Since each service time equals 1/u = 1/240 = 4.16 ms, the waiting time of each character is 5.06 ms. Now suppose the number of terminals increases from 48 to 96, so that the traffic intensity is less than unity, two transmission lines are needed, and the traffic intensity is still equal to 0.6. From Fig. 2, the buffer length corresponding to the desired overflow probability for two transmission lines is about 14 characters. The waiting time is about 0.8 characterservice times which is equal to 3.33 ms. Although the difference between 5.06 and 3.33 ms may not be detected by a user at a terminal, a common buffer of the same size operating with two output lines can handle twice the number of input lines as with one output line. Thus, the common buffer approach permits handling a wide range of traffic without substantial variation in buffer size. REFERENCES [1] W. W. Chu, "A study of asynchronous time division multiplexing for time-sharing computer communications, " presented at the 2nd Hawaii

Abstract not found

We give an overview of the results in the literature on single-server queues with the FB discipline. The FB discipline gives service to the customer that has received the least amount of service. This not so well-known discipline has some appealing features, and performs well for heavy-tailed service times. We describe results on the queue length, sojourn time, and the influence of variability in the service times.