Anomalous behavior for some reference strings has been encountered during program execution where a working set (WS) policy has been invoked. Under the WS policy, a parameter value specifies the number of pages previo...
详细信息
Anomalous behavior for some reference strings has been encountered during program execution where a working set (WS) policy has been invoked. Under the WS policy, a parameter value specifies the number of pages previously referenced that are to be kept in memory, and the WS policy then keeps those pages. Four numerical programs were written to investigate the incidence of the parameter-real memory anomaly and the real memory-fault rate anomaly and the possible relatedness of this behavior to the parameter value. In 2 of the 4 programs, it was indeed observed that as the value was increased, average real-time memory decreased, as did the number of page faults, even though the opposite effect would have been expected. The results throw into question any attempt to use window size as a device for controlling the execution space and time required in a multiprogramming environment.
A demand prepaging algorithm DPMIN is defined and proved to be an optimal demand prepaging algorithm. However, it cannot be used in practice since it requires that the future refreence string be completely known in ad...
详细信息
A demand prepaging algorithm DPMIN is defined and proved to be an optimal demand prepaging algorithm. However, it cannot be used in practice since it requires that the future refreence string be completely known in advance. Several practical prepaging algorithms are also defined which require only a partial knowledge of the future reference string. Finally, we show that these prepaging algorithms reduce the paging problems of array algorithms operating on large arrays.
This paper presents, using queuing theory and optimization techniques, an approach for estimating the optimal capacities and speeds of the memory levels in a memory hierarchy operating in the multiprogrammed environme...
详细信息
This paper presents, using queuing theory and optimization techniques, an approach for estimating the optimal capacities and speeds of the memory levels in a memory hierarchy operating in the multiprogrammed environment. Optimality is defined with respect to mean system response time under a fixed cost constraint. It is assumed that the number of levels in the hierarchy as well as the capacity of the lowest level are known. The effect of the storage management strategy is characterized by the miss ratio function which, together with the device technology cost functions, is assumed to be representable by power functions. It is shown that as the arrival rate of processes and/or the number of active processes in the system increase, the optimal solution deviates considerably from that under a uniprogrammed environment. It is also shown that the solution obtained is the global optimum.
An important topic in automatically parallelizing nested loop programs (for the implementation on a parallel machine or for the mapping onto VLSI circuits) is the computation of linear data dependencies. Recently, Fea...
详细信息
An important topic in automatically parallelizing nested loop programs (for the implementation on a parallel machine or for the mapping onto VLSI circuits) is the computation of linear data dependencies. Recently, Feautrier has developed an algorithm which is based on the solution of a parametric integer linear program. In this paper another method is presented. This method only works for special cases but it should be simpler and more efficient. Apart of these special cases the extensions suggested in this paper provide a system of fewer inequalities and variables in the general case compared to the original algorithm of Feautrier. The key idea is the solution of a system of linear diophantine equations.
A simple nucleus for a single-processor computer is considered. The entire nucleus program is written in the high-level programming language Modula-2 1 without assembly coding. The concepts of the nucleus and some sim...
详细信息
A simple nucleus for a single-processor computer is considered. The entire nucleus program is written in the high-level programming language Modula-2 1 without assembly coding. The concepts of the nucleus and some simple applications are described. A complete listing of the nucleus is included. We conclude that programming of a nucleus in an appropriate high-level language is simpler than in an assembly language. The resulting nucleus has a comparable performance with other nuclei written in assembly code.
The technique of windowing has often been used in the implementation of the waveform relaxation method for solving ordinary differential equations and time-dependent partial differential equations. Its efficiency depe...
详细信息
The technique of windowing has often been used in the implementation of the waveform relaxation method for solving ordinary differential equations and time-dependent partial differential equations. Its efficiency depends upon the operator splitting and the coupling of the subsystems involved. For two model problems, estimates for the window length and convergence rate for the early phase of the iteration are derived. The effectiveness of windowing is then investigated, leading to some suggestions for the use of windowing. The discussions are supported by the computational results.
The performance of multiprogrammed shared-memory multiprocessors suffers often from scheduler interventions that neglect data locality. On cache-coherent distributed shared-memory (DSM) multiprocessors, such scheduler...
详细信息
The performance of multiprogrammed shared-memory multiprocessors suffers often from scheduler interventions that neglect data locality. On cache-coherent distributed shared-memory (DSM) multiprocessors, such scheduler interventions tend to increase the rate of remote memory accesses. This paper presents a novel dynamic page migration algorithm that remedies this problem in iterative parallel programs, The purpose of the algorithm is the early detection of pages that will most likely be accessed remotely by threads associated with them via a thread-to-memory affinity relation. The key mechanism that enables timely identification of these pages is a communication interface between the page migration engine and the operating system scheduler. The algorithm improves previously proposed competitive page migration algorithms in many aspects. including accuracy, timeliness and cost amortization. Most notably, the algorithm is not biased by obsolete memory access history that may be accumulated in the page access counters at runtime. Experiments on the SGI Origin-2000 show that the algorithm outperforms by far the best static page placement algorithm and a customized page migration engine implemented in IRIX. the Origin-2000 operating system. The algorithm is implemented at user-level and its functionality is orthogonal to the scheduling police of the operating system. (C) 2002 Elsevier Science (USA).
Communication by message passing or by procedure calls in one of the key issues when discussing languages for multiprogramming. The two languages Platon and Concurrent Pascal represent the different approaches which a...
详细信息
Communication by message passing or by procedure calls in one of the key issues when discussing languages for multiprogramming. The two languages Platon and Concurrent Pascal represent the different approaches which are contrasted by presenting a few programs written in both languages.
The process of transporting Brinch Hansen's implementation of Concurrent PASCAL to another minicomputer is described. Applicable porting strategies are discussed with emphasis on the design decisions made for a sp...
详细信息
The process of transporting Brinch Hansen's implementation of Concurrent PASCAL to another minicomputer is described. Applicable porting strategies are discussed with emphasis on the design decisions made for a specific transportation. Important design decisions include the use of a virtual code interpreter and implementation in an operating system environment. The problems of this transportation are illustrated with accompanying suggestions for a more portable system.
This paper highlights the salient features of the real- time operating system/360 (RTOS) which is being used as the ground support system for the Apollo space program.
This paper highlights the salient features of the real- time operating system/360 (RTOS) which is being used as the ground support system for the Apollo space program.
暂无评论