检索结果-内蒙古大学图书馆

PARALLEL-PROCESSING ON THE MULTI-PSI COMPUTER AND ITS EVALUATION - A programming PARADIGM-BASED ON A SMALL-GRAIN HIGHLY CONCURRENT OBJECT MODEL

引用

SYSTEMS AND COMPUTERS IN JAPAN 1994年第1期25卷 17-36页

作者： TAKI, K ICHIYOSHI, N Member Institute for New Generation Computer Technology Tokyo Japan 108 Nomember

Multi-PSI is a distributed-memory MIMD (multiple-instruction, multiple-data) computer that has 64 processors. Problems of knowledge processing, which are the object of the Multi-PSI, must be divided into many partial problems in order to achieve a favorable load balance because a characteristic of the computation is the dynamic loss of uniformity. This, however, has the danger of increasing the communication overhead. Therefore, in order to constrain the communication overhead and achieve a favorable load balance, in this paper, we propose a programming method oriented to the above problem. In this methodology, the problem is formulated as multiple communicating objects and the degrees of freedom in the load allocation are maintained. On the other hand, when the load is allocated, the communication objects are estimated from the system performance and the objects' characteristics, and the processing granularity is adjusted so that it is within the tolerance. This program methodology was applied to the three program developments of the shortest path problem, the LSI routing problem, and logic simulation;each was implemented with high efficiency. For example, in the shortest path program, 260,000 small-grained objects were distributed over 64 processors and reached about 75% efficiency. In addition, in the logic simulation composed of 12,000 objects (gates), a high absolute efficiency of 99 thousand events/second and a 48-fold increase in speed with 64 processors were obtained. Along with confirming the effectiveness of this programming methodology, these results demonstrated that high efficiency can be realized in parallel processing for some degree of granularity even in a distributed-memory parallel computer.

关键词： MULTI-PSI MIMD COMPUTER programming methodology CONCURRENT OBJECT MODEL LOAD BALANCE COMMUNICATION OVERHEAD PERFORMANCE EVALUATION

来源：评论

学校读者我要写书评

暂无评论

REASONING ABOUT PROBABILISTIC PARALLEL PROGRAMS

引用

ACM TRANSACTIONS ON programming LANGUAGES AND SYSTEMS 1994年第3期16卷 798-842页

作者： RAO, JR IBM Thomas J. Watson Research Center Yorktown Heights NY

The use of randomization in the design and analysis of algorithms promises simple and efficient algorithms to difficult problems, some of which may not have a deterministic solution. This gain in simplicity, efficiency, and solvability results in a trade-off of the traditional notion of absolute correctness of algorithms for a more quantitative notion: correctness with a probability between 0 and 1. The addition of the notion of parallelism to the already unintuitive idea of randomization makes reasoning about probabilistic parallel programs all the more tortuous and difficult. In this paper we address the problem of specifying and deriving properties of probabilistic parallel programs that either hold deterministically or with probability 1. We present a proof methodology based on existing proof systems for probabilistic algorithms, the theory of the predicate transformer, and the theory of UNITY. Although the proofs of probabilistic programs are slippery at best, we show that such programs can be derived with the same rigor and elegance that we have seen in the derivation of sequential and parallel programs. By applying this methodology to derive probabilistic programs, we hope to develop tools and techniques that would make randomization a useful paradigm in algorithm design.

关键词： ALGORITHMS VERIFICATION CORRECTNESS PROOFS PARALLEL programming PROBABILISTIC ALGORITHMS programming methodology SPECIFICATION TECHNIQUES VERIFICATION

来源：评论

学校读者我要写书评

暂无评论

Formal, model-oriented software development methods: From VDM to ProCoS & from RAISE to LaCoS

引用

Future Generation Computer Systems 1992年第2-3期7卷 111-138页

作者： Bjørner, Dines Haxthausen, Anne Elisabeth Havelund, Klaus Department of Computer Science Technical University of Denmark DK-2800 Lyngby Buildings 344-345 Denmark CRI: Computer Resources Intl. DK-3460 Birkerød Bregnerødvej 144 Denmark

The first author has been invited to reminisce over almost 20 years of his research into and application of formal methods in software development. The paper deals with model-oriented methods. The title of this paper mentions two methods: VDM (Vienna Development Method) and RAISE (Rigorous Approach to Industrial Software Engineering), the latter derived from the former, and two projects: ProCoS (Provably Correct Systems) and LaCoS (Large-scale Correct Systems). LaCoS is a follow-up on RAISE. ProCoS adheres to the same principles as VDM as VDM and RAISE, but with a twist! The paper will outline the technical/scientific essence of the VDM and RAISE methods, and explain the ProCos and LaCoS project structures. The paper will first outline five central rôles of software developers: problem domain specialists, programmers, software engineers, resident computation scientists and managers. We will briefly illustrate examples of specifications in VDM and RAISE. The aim of the paper is to record that formal methods in software development are now well-developed and broadly accepted in Europe. © 1992.

关键词： abstract data types algebraic semantics concurrency formal method model-oriented semantics modules non-determinism parallellism programming methodology RAISE VDM

来源：评论

学校读者我要写书评

暂无评论

ADAPTIVE programming

引用

IEEE TRANSACTIONS ON SOFTWARE ENGINEERING 1991年第9期17卷 911-921页

作者： GOUDA, MG HERMAN, T Department of Computer Sciences University of Texas at Austin Austin TX 78712

An adaptive program is one that changes its behavior based on the current state of its environment. This notion of adaptivity is formalized and a logic for reasoning about adaptive programs is presented. The logic includes several composition operators that can be used to define an adaptive program in terms of given constituent programs;programs resulting from these compositions retain the adaptive properties of their constituent programs.

关键词： DISTRIBUTED COMPUTING ADAPTIVE SYSTEMS programming methodology PROGRAM COMPOSITION SELF-STABILIZATION

来源：评论

学校读者我要写书评

暂无评论

EMERALD - A GENERAL-PURPOSE programming LANGUAGE

引用

SOFTWARE-PRACTICE & EXPERIENCE 1991年第1期21卷 91-118页

作者： RAJ, RK TEMPERO, E LEVY, HM BLACK, AP HUTCHINSON, NC JUL, E DIGITAL EQUIPMENT CORP CAMBRIDGE RES LABCAMBRIDGEMA 02139 UNIV ARIZONA DEPT COMP SCITUCSONAZ 85721 UNIV COPENHAGEN DIKUDK-2100 COPENHAGENDENMARK

Emerald is a general-purpose language with aspects of traditional object-oriented languages, such as Smalltalk, and abstract data type languages, such as Modula-2 and Ada. It is strongly typed with a nontraditional object model and type system that emphasize abstract types, allow separation of typing and implementation, and provide the flexibility of polymorphism and subtyping with compile-time checking. This paper describes the Emerald language and its programming methodology. We give examples that demonstrate Emerald's features, and compare and contrast the Emerald approach to programming with the approaches used in other similar languages.

关键词： programming LANGUAGES programming methodology OBJECT-ORIENTED programming ABSTRACT DATA TYPES INHERITANCE OBJECT-BASED CONCURRENCY

来源：评论

学校读者我要写书评

暂无评论

Improving the curriculum through the teaching of calculation and discrimination

引用

Education and Computing 1991年第1-2期7卷 61-72页

作者： Gries, David Cornell University Ithaca NY 14853 United States

The field of computing-including its application by programmers, systems analysts and others-suffers tremendously from a lack of use of formal reasoning in everyday work. The problem can be solved, at least partly, by teaching calculational methods in mathematics and programming in the freshman year, and in such a way that students develop a skill with the methods, and not just a passing understanding. Freshman courses should instil the idea that mathematical techniques can help and should encourage a sense of discrimination, of making judgements on technical matters based on technical concerns. A serious overhaul of the conventional second programming course and the discrete mathematics course along these lines could have a profound impact on the whole computer science curriculum. Moreover, a properly designed course would be beneficial for mathematicians and engineers as well. © 1991 Elsevier Science Publishers B.V. All rights reserved.

关键词： computer science education programming methodology software engineering teaching logic

来源：评论

学校读者我要写书评

暂无评论

USING DYNAMIC PREDICATES IN AN OR-PARALLEL PROLOG SYSTEM

USING DYNAMIC PREDICATES IN AN OR-PARALLEL PROLOG SYSTEM

引用

INTERNATIONAL SYMP ON LOGIC programming

作者： SZEREDI, P SZKI Intelligent Software Ltd Budapest Hungary

ISBN: (纸本)0262691477

Aurora is a prototype or-parallel implementation of Prolog for shared memory multiprocessors. It supports the full Prolog language, thus being able to execute existing Prolog programs without any change. For some programs, however, typically those relying on dynamic database handling, full compatibility with Prolog may cause unnecessary sequencing delays. Aurora therefore supports a number of extensions to Prolog, including asynchronous versions of all side-effect predicates. Programs often rely on dynamic predicates because of bad programming style. There are, however, applications where dynamic predicates are the most natural way of expressing a solution. In this paper we look at a simple, yet interesting such application: a program for playing the game of mastermind. This is a typical example of a search using a continually changing knowledge base. We first look at sequential search strategies for playing the game of mastermind. We then proceed to discuss the problems arising when these strategies are executed in parallel, with asynchronous database handling. We present several versions of the mastermind program, discussing various synchronization techniques and outlining proposals for higher level synchronization primitives to be incorporated into Aurora. We discuss performance results for the presented programs using an experimental implementation of the synchronization primitives.

关键词： LOGIC programming programming methodology PARALLEL EXECUTION SYNCHRONIZATION

来源：评论

学校读者我要写书评

暂无评论

programming ABSTRACTIONS FOR DYNAMICALLY PARTITIONING AND COORDINATING LOCALIZED SCIENTIFIC CALCULATIONS RUNNING ON MULTIPROCESSORS

引用

SIAM JOURNAL ON SCIENTIFIC AND STATISTICAL COMPUTING 1991年第1期12卷 145-157页

作者： BADEN, SB UNIV CALIF BERKELEY LAWRENCE BERKELEY LAB DEPT MATHBERKELEYCA 94720 UNIV CALIF BERKELEY DIV COMP SCIBERKELEYCA 94720

Certain software abstractions help to automate load balancing during various math-physics calculations on a team of concurrently executing processors. These abstractions have been tested on a vortex method for computational fluid dynamics. Experiments exhibited good parallel speedups of 24 and 3.6, respectively, on 32 processors of the Intel iPSC-1-a message-passing hypercube architecture-and on 4 processors of a Cray X-MP-a shared-memory vector architecture. The abstractions should apply to diverse applications, including finite difference methods, and to diverse architectures without requiring that the application be reprogrammed extensively for each new architecture.

关键词： MULTIPROCESSORS DYNAMIC LOAD BALANCING PARALLEL NUMERICAL COMPUTATION SOFTWARE PORTABILITY programming methodology ABSTRACTIONS SOFTWARE TOOLS SCIENTIFIC COMPUTING

来源：评论

学校读者我要写书评

暂无评论

Recent evolution of the introductory curriculum in computing

引用

Education and Computing 1991年第1-2期7卷 43-60页

作者： Tucker, Allen B. Garnick, David K. Department of Computer Science Bowdoin College Brunswick ME 04011 United States

This paper traces the evolution of the first courses in undergraduate programs over the last twelve years. We identify fundamental weaknesses in the present sequence, and propose strengthening the introductory curriculum, using a breadth-first approach that integrates theoretical material. © 1991 Elsevier Science Publishers B.V. All rights reserved.

关键词： discrete mathematics introductory curriculum programming methodology scheduled laboratories social and ethical context

来源：评论

学校读者我要写书评

暂无评论

Introductory education in computer science

引用

Education and Computing 1991年第1-2期7卷 73-86页

作者： Habermann, A. Nico Carnegie-Mellon University Pittsburgh PA 15213 United States

There is room for improvement in the way introductory computer science is taught. In general, most of the available time is spent on teaching programming language syntax and very little on studying examples of good programs. The assertion approach to program verification has in practice resulted in a technique that resembles assembly code programming of more than 30 years ago. This practice has also encouraged the view that each new program is an invention that must be developed from scratch. In addition, the student's attention is largely directed towards the how of the technique and hardly towards learning the proven facts. In this paper, it is argued that the introduction to programming can be improved in several ways: by starting with a systems view that encourages students to study good examples and to reuse existing programs, by judging program quality relative to similar programs and program abstractions, and by building up a coherent body of knowledge that students can apply in program design and implementation. © 1991 Elsevier Science Publishers B.V. All rights reserved.

关键词： introductory programming program synthesis program verification programming methodology reusable programs

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：