A network computer is a collection of computers designed to function as one machine. On a network computer, as opposed to a multiprocessor, constituent subcomputers are memory-disjoint and communicate only by some for...
ISBN:
(纸本)9780897911696
A network computer is a collection of computers designed to function as one machine. On a network computer, as opposed to a multiprocessor, constituent subcomputers are memory-disjoint and communicate only by some form of message exchange. Ensemble architectures like multiprocessors and network computers are of growing interest because of their capacity to support parallel programs, where a parallel program is one that is made up of many simultaneously-active, communicating processes. Parallel programs should, on an appropriate architecture, run faster than sequential programs, and, indeed, good speed-ups have been reported in parallel programming experiments in several domains, amongst which are AI, numerical problems, and system simulation. Our interest lies in network computers, particularly ones that range in size from several hundred nodes to several *** computers may be organized in either of two basic ways: their nodes may communicate over a shared bus (or series of buses), as in S/Net; or over point-to-point links, as in Cosmic Cube and the Transputer Network. The work to be presented deals with the point-to-point class, the elements of which we shall refer to as “linked networks”.Linked networks face a fundamental communication problem. Unless they are completely connected (which is rarely possible), two communicating nodes will not necessarily be connected by a single link. Messages between nodes must therefore, in general, travel over several links and be processed by several intermediate nodes. Communication delays increase with the length of the traveled path. Network computer designers therefore provide networks the diameters of which are small relative to their size, and network operating systems will attempt to place communicating processes as close to each other as *** present a communication protocol for linked networks that was designed specifically for network computers. Staged Circuit Switching is a communication protocol th
The importance of performance measurement and evaluation has long been recognized in the field of computer science. It is used in the analysis of existing systems, to make projections on the performance of new or modi...
ISBN:
(纸本)9780897911504
The importance of performance measurement and evaluation has long been recognized in the field of computer science. It is used in the analysis of existing systems, to make projections on the performance of new or modified systems, and in the design and selection of new hardware and software. The tools and methods used in performance measurement and evaluation are varied. They include such things as timings, benchmarks, simulations, analytic modeling, and both hardware and software monitors. The ones used in a given instance depend upon the goals of the investigator and the system being *** “Proposed Automated Information Management at NASA: Its Performance measurement and Evaluation” Rebecca R. Bogart presents performance measurements that are being considered for use in evaluating an information system that is still in the planning stages. This system will be an automated information system for a network that includes NASA headquarters and ten NASA centers. The paper describes the goals and objectives of this system and an evaluation technique that can be used to determine how well the goals and objectives are met. The primary tools proposed are software monitors embedded in the system which will gather system performance statistics. The use of user questionnaires is planned to augment the evaluation *** E. Tolle used transaction log analysis and stochastic processes in the study described in his paper, “Performance measurement and Evaluation of Online Information systems.” Transaction logs, records of user commands and system responses, were gathered from several different online information systems and analyzed. This paper describes how desired data were obtained from these logs and how they were used in the stochastic processes. The primary objective of this study was to discover the extent to which online systems were used and to determine the patterns of user commands when conducting information searches. This information could then be used to d
This conferenceproceedings contains 22 papers. They present a view of performance evaluation of computersystems as a viable experimental science that continues to be relevant to the real-world problems of systems an...
详细信息
ISBN:
(纸本)0897911415
This conferenceproceedings contains 22 papers. They present a view of performance evaluation of computersystems as a viable experimental science that continues to be relevant to the real-world problems of systems and software effectiveness. The main subjects presented are computersystems performance evaluation, queueing network models, computer simulation, and probabilistic and approximate models of performance evaluation. Some applications of performance evaluation techniques on some well-known operating systems are also discussed.
This paper describes a hybrid model, combining both analytical and simulation techniques, which was developed to study the performance of a netted computer based system. The computer based system that was modeled is t...
ISBN:
(纸本)9780897911412
This paper describes a hybrid model, combining both analytical and simulation techniques, which was developed to study the performance of a netted computer based system. The computer based system that was modeled is the Facility Assignment and Control System (FACS). This system is presently being deployed within several Bell Operating Companies to inventory and assign central office and outside plant facilities. A key feature of the model is its ability to characterize the dynamic nature of FACS. An understanding of this dynamic nature is necessary in establishing important operational guidelines such as allowable CPU utilization, levels of multiprogramming and priority of transaction processing. In addition, the model allows the user to investigate the sensitivity of the system to a wide range of conditions. Typical study items could include the effect of various load scenarios, ability of the system to meet performance objectives, and different hardware configurations. As part of this paper, both the practical aspects of modeling a netted computer based system and the theoretical development of the hybrid model are considered.
The quality of placement and routing on gate arrays is commonly measured by average wire length. With regard to wire length, placement and routing are mutually competing tasks and the solution space for both is expone...
详细信息
ISBN:
(纸本)9780818605420
The quality of placement and routing on gate arrays is commonly measured by average wire length. With regard to wire length, placement and routing are mutually competing tasks and the solution space for both is exponential. Estimates of measures of placement such as average wire length or, total wiring tracks prior to routing give some indication of the routability of the placement and, can be used to select another placement and *** nature of these problems necessitates a probabilistic approach to the wirability analysis of integrated circuits. Stochastic models for wiring space estimation and the relation between wire length distribution and placement optimization have received attention recently [1], [6], [2], [5], [7], [3] and [4]. Much of the reported work on wire length distributions and placement of logic rests on empirical evidence that indicates that “well placed” chips exhibit Rent's Rule between the number of components and the number of corresponding external connections. Rent's Rule has been the basis of the heuristic arguments used to derive upper bounds on the average wire length and the form of the wire length distribution. Rent's Rule has the formT = KC p (1)where T is the average number of external connections, C is the average number of components, K is number of connections per component and p is a positive constant. In [1], [2] and [4] the effect of placement on wire length distribution was introduced by assuming that a hierarchical partitioning scheme aimed at minimizing the average wire length results in a configuration that exhibits Rent's Rule. In [2] an upper bound r-k for the average wire length between elements of different subsets of components of size k was derived, and using Rent's Rule to obtain the number of connections between such subsets, an upper bound on the average wire length was derived. In [3] the Pareto distribution is proposed for the distribution of wire lengths. Similar results were presented in [4].In this paper
The proceedings contain 25 papers. The topics discussed include: A Study of dynamic reconnection;on the BLI-model of program behavior;HOPT: a myopic version of the STOCHOPT automatic file migration policy;paging on an...
ISBN:
(纸本)0897911121
The proceedings contain 25 papers. The topics discussed include: A Study of dynamic reconnection;on the BLI-model of program behavior;HOPT: a myopic version of the STOCHOPT automatic file migration policy;paging on an object-oriented personal computer;diffusion approximations for storage processes in computersystems;transient diffusion approximation for some queuing systems;performance analysis of checkpointing strategies;and construction of a validated simulator for performance prediction of DECnet-based computer networks.
In this paper we focus on the storage resource. A basic model of the space time requirements of jobs in a computer system is described, and a number of its variations analyzed by means of diffusion approxmiations. Sub...
ISBN:
(纸本)0897911121
In this paper we focus on the storage resource. A basic model of the space time requirements of jobs in a computer system is described, and a number of its variations analyzed by means of diffusion approxmiations. Subject to the usual heavy traffic assumptions, the result of this analysis enable one to quantify the effects of limitations on both storage capacity and processing rates.
Recent developments in very large scale integration have made it feasible to construct a highly parallel computer composed of large numbers of interconnected microcomputers. The modeling problems posed by this approac...
详细信息
This paper presents an approximate analysis of a multiprocessor system consisting of P processors, M memory modules, and B buses. The model assumes constant memory access times, arbitrary memory access patterns, and b...
详细信息
暂无评论