Conditional Critical Regions (CCR) are a very good instrument for describing and designing process-synchronisation in real time systems. A straightforward implementation is not a feasible solution because of its huge ...
详细信息
Conditional Critical Regions (CCR) are a very good instrument for describing and designing process-synchronisation in real time systems. A straightforward implementation is not a feasible solution because of its huge runtime-overhead, but certain properties of the CCR allow the separation of the synchronizing operations on the shared data from the non - synchronizing ones. This allows to use standardized conditions (“normalized CCR”) instead of general conditional expressions and to extend the monitor-concept by integrating the normalized CCR. Thus a very efficient implementation has been found. The feasibility of this solution in multitasking systems on a single processor has been proven by an industrial project, a large data-acquisition system in a rolling mill, where many difficult problems of event-, timeout- and exception-handling in a very rough process environment had to be solved. Then some ideas concerning the applicability of the CCR in multiprocessor environments will be discussed, especially in distributed systems where the shared data concept in the original meaning is no longer valid and is replaced by other process-communication techniques (signals, mailboxes, rendezvous). It is shown, that the normalized CCR as synchronizing system-control element need not necessarily be embedded in the shared data-concept and the concept is suited for distributed systems also, preserving its advantages such as transparent programming of process control on user level, functional separation of long- and medium term scheduling, simplicity and generality.
Three principles of optimality for multiprogramming are derived from a general model of a virtual memory computer system. They state the existence of both an optimal multiprogramming degree and an optimal program mixt...
详细信息
Simple queueing models are used to study the performance tradeoffs of multiple processor systems. Issues considered include the impact of CPU service disciplines and distributions, level of multiprogramming, multitask...
详细信息
In a Paged Virtual Memory system, scheduling and multiprogramming control must consider factors such as the page management policy, system resource utilization and page thrashing. This paper investigates algorithms fo...
详细信息
Computer evaluation, comparison, and selection is essentially a decision process. The decision making is based on a number of worth indicators, including various computer performance indicators. The performance indica...
详细信息
ISBN:
(纸本)9781450374880
Computer evaluation, comparison, and selection is essentially a decision process. The decision making is based on a number of worth indicators, including various computer performance indicators. The performance indicators are obtained through the computer performance measurement procedure. Consequently, this procedure should he completely conditioned by the decision process. This paper investigates various aspects of computer performance measurement and evaluation procedure within the context of computer evaluation, comparison and selection process based on the Logic Scoring of Preference method. The set of elementary criteria for performance evaluation is proposed and the corresponding set of performance indicators is defined. The necessary performance measurements are based on the standardized set of synthetic benchmark programs and include three separate measurements: monoprogramming performance measurement, multiprogramming performance measurement, and multiprogramming efficiency measurement. Using the proposed elementary criteria, the measured performance indicators can be transformed into elementary preferences and aggregated with other non-performance elementary preferences obtained through the evaluation process. The applicability of presented elementary criteria is illustrated by numerical examples. � ACM. All rights reserved.
An operating system supporting laboratory automation and interactive use of a computer has been designed and implemented for the multiprogrammed RC4000 computer. The emphasis in the design has been to give the user in...
详细信息
An operating system supporting laboratory automation and interactive use of a computer has been designed and implemented for the multiprogrammed RC4000 computer. The emphasis in the design has been to give the user interactive access to as many system resources as possible, and to facilitate writing of process control programs in a high-level language, i.e. ALGOL. The operating system is used to run control and data collection programs for a variety of experiments in physical chemistry. Laboratory automation programs can be started and removed dynamically, and may even be restarted automatically after a system failure. The operating system is a 1500 line ALGOL program. The first version was designed, implemented and debugged in about 4 person-months. The operating system has been running day and night for approximately 5 years without errors despite heavy use.
The study of operating systems principles is greatly enhanced if processes can be expressed in language constructs that explicitly allow for multiprogramming. The enhancement is increased if such a system actually exi...
详细信息
The study of operating systems principles is greatly enhanced if processes can be expressed in language constructs that explicitly allow for multiprogramming. The enhancement is increased if such a system actually exists for practical use. This paper describes an implementation of such a system. The language used (a forerunner of Concurrent Pascal) is described briefly. The body of the paper describes the implementation method in some detail. In order to achieve multiprogramming with a single processor, a number of process descriptors are maintained, each associated with the execution point in the program of a parallel process. Control is distributed among the processes in a manner determined by the program environment. The design of the implementation, in itself, provides an interesting example of operating systems principles at work.
A simple MP system consisting of an input-output facility and a central processor is modeled as a two-parameter Markov chain. The conditions for stability are demonstrated, and the steady-state joint probabilities are...
详细信息
A simple MP system consisting of an input-output facility and a central processor is modeled as a two-parameter Markov chain. The conditions for stability are demonstrated, and the steady-state joint probabilities are calculated explicitly. Various priority and capacity assignments result in radically different analytical situations, some of which have been considered in the literature. The present work treats a version that was considered for a time intractable. This paper emphasizes the analytical properties of the probability-generating functions and a method to solve a resultant functional equation. The numerical results display the importance of dependence between variables in the model.
We propose a new method for the control of a multiprogrammed virtual memory computer system. A mathematical model solved by decomposition permits us to justify that the method avoids thrashing. Simulation experiments ...
详细信息
We propose a new method for the control of a multiprogrammed virtual memory computer system. A mathematical model solved by decomposition permits us to justify that the method avoids thrashing. Simulation experiments are used to test the robustness of the predictions of the mathematical model when certain simplifying assumptions are relaxed and when a slightly simpler control technique based on the same principle is used. Comparisons are given with the case where an "optimal" control is used and with that with no control. We also provide a simulation evaluating the estimators used in an implementation of the control, as well as the responsiveness of the controlled system to transients in the workload.
Utilization of a uniprocessor system in a multiprogramming environment can be optimized by maximizing the overlap of processor and input-output operations. A computational process can be modeled by a directed graph ea...
详细信息
Utilization of a uniprocessor system in a multiprogramming environment can be optimized by maximizing the overlap of processor and input-output operations. A computational process can be modeled by a directed graph each node of which represents a task comprising processor and input-output segments. Any optimal schedulng algorithm for the model cannot be polynomially bounded, but the optimal criteria can be used to develop a hierarchy of dispatching heuristics based upon selecting an optimal partial task schedule. These heuristics are analyzed and evaluated by a simulation study and are shown to be more effective than those previously proposed. The dispatching heuristics developed have a wide range of potential applications to systems requiring dynamic task scheduling.
暂无评论