检索结果-内蒙古大学图书馆

Heterogeneous computing Workshop

作者： Sung-Yong Park Joohan Lee S. Hariri High Performance Distributed Computing Laboratory Department of Electrical Engineering and Computer Science Syracuse University Syracuse NY USA

NYNET (ATM wide-area network testbed in New York state) Communication System (NCS) is a multithreaded message-passing tool developed at Syracuse University that provides low-latency and high-throughput communication services over Asynchronous Transfer Mode (ATM)-based high-performance distributed computing (HPDC) environments. NCS provides flexible and scalable group communication services based on dynamic grouping and tree-based multicasting. The NCS architecture, which separates the data and control functions, allows group operations to be implemented efficiently by utilizing the control connections when transferring status information (e.g. topology information, routing information). Furthermore, NCS provides several different algorithms for group communication and allows programmers to select an appropriate algorithm at runtime. The authors overview the general architecture of NCS and present the multicasting services provided by NCS. They analyze and compare the performance of NCS with that of other message-passing tools such as p4, PVM, and MPI in terms of primitive performance and performance.

关键词： Asynchronous transfer mode computer architecture Communication system control Multicast algorithms System testing distributed computing Topology Routing Programming profession Runtime

来源：评论

学校读者我要写书评

暂无评论

Reusable software components for performability tools, and their utilization for web-based configurable tools 10th

Reusable software components for performability tools, and t...

引用

10th International Conference on Modelling Techniques and Tools for computer performance Evaluation, Tools 1998

作者： Van Moorsel, Aad P. A. Huang, Yiqing Distributed Software Research Department Bell Laboratories Research Lucent Technologies 600 Mountain Ave Murray HillNJ07974 United States Center for Reliable and High-Performance Computing Coordinated Science Laboratory University of Illinois at Urbana-Champaign 1308 W. Main St UrbanaIL61801 United States

ISBN: (纸本)3540649492

This paper discusses software reusability strategies for performance and reliability modeling tools. Special emphasis is on web-embedded tools, and the potential interaction between such tools. We present the system analysis tools (SAT) application programming interface, which allows for quickly embedding existing tools in the web, and generally simplifies programming analysis tools by structured reuse. We also introduce the FREUD project, which has as primary aim to establish a single point of access to a variety of web-enabled tools. In addition, FREUD facilitates configurable web tools by allowing a user to select from the registered modeling formalisms, solvers and graphics tools, and by providing glue between the tools through scripting. We will argue that this form of reuse is particularly suitable for performability modeling tools because of their predictable usage pattern. © Springer-Verlag Berlin Heidelberg 1998.

关键词： computer software reusability

来源：评论

学校读者我要写书评

暂无评论

Transaction specification for object-oriented realtime systems in HRT-HOOD 2nd

Transaction specification for object-oriented realtime syste...

引用

2nd International Eurospace-Ada-Europe Symposium, Ada in Europe 1995

作者： Cornwell, Pete Wellings, Andy Distributed Objects Research Group Department of Computing Bournemouth University Dorset United Kingdom Real-Time Systems Research Group Department of Computer Science University of York York United Kingdom

ISBN: (纸本)3540607579

Object-oriented development is a structuring paradigm. It does not provide substantial support for the specification of ‘end to end’ control and data flow. This has implications for hard real-time system development, where applications may specify ‘end to end’ response times, also known as transactions. This paper introduces a graphical notation for the specification of transactions in HRT-HOOD, an object-oriented design technique for hard real-time system development. It provides an overview of the notation, and its practical application as part of a ‘top-down’ development process. This is illustrated through a case study, based on a naval command-control system. © Springer-Verlag Berlin Heidelberg 1996.

关键词： Specifications

来源：评论

学校读者我要写书评

暂无评论

Transaction integration for reusable hard real-time components

Transaction integration for reusable hard real-time componen...

引用

IEEE International Symposim on high Assurance Systems Engineering

作者： P. Cornwell A. Wellings Distributed Objects Research Group Department of Computing Bournemouth University UK Real-Time Systems Research Group Department of Computer Science University of York UK

The paper presents a strategy for the composition of components to meet 'end to end' timing requirements. This is a framework based approach known as transaction integration. A framework is constructed through the use of transactions. A transaction represents a control and data flow path across a set of component interfaces, or placeholders, that must be completed by a specific deadline. Reusable real time components are 'plugged' into each placeholder. The paper provides techniques to verify that the performance characteristics of each placeholder, and the 'end to end' deadlines associated with each transaction are met.

关键词： Real time systems Object oriented modeling Plugs Sockets distributed computing computer science Timing Weapons Bibliographies Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

An MPI library which uses polling, interrupts and remote copying for the Fujitsu AP1000+

An MPI library which uses polling, interrupts and remote cop...

引用

International Symposium on Parallel Architectures, Algorithms and Networks (ISPAN)

作者： D. Sitsky E. Hayashi Department of Computer Science CAP Research Program Australian National University Canberra ACT Australia High Performance Computing Group Fujitsu Laboratories Limited Kawasaki Japan

A complete implementation of MPI for the Fujitsu AP1000+ is presented. The library can employ a number of different mechanisms in implementing the send and receive message passing operations. The method of detecting the arrival of new messages can be realized through interrupt-driven and polling techniques. Transferring message data is achieved by either sending the message data directly to the receiver "in-place", or using a rendezvous method which allows the use of a fast noncopying nonblocking remote-fetching operation. The MPI library exhibits good performance compared to the native message passing library, and allows the user to decide at runtime which mechanisms will be used in order to achieve the best performance on a per-application basis.

关键词： Message passing Broadcasting Routing computer science Runtime library Parallel programming Standards organizations Workstations National electric code high performance computing

来源：评论

学校读者我要写书评

暂无评论

Operational systems, logistics engineering and technology insertion

引用

NAVAL ENGINEERS JOURNAL 1997年第3期109卷 205-220页

作者： Grubb, MJ Skolnick, A Michael J. Grubb:has perfomzed naval engineering at the Crane Division Naval Surface Warfare Center (NSWC) for more than twenty-five years. He currently the program manager for the Sustainable Hardware and Affordable Readiness Practice Program (SHARP). is a Navy-wide logistics research and development prgram aimed at the successful transiton of proved technologies inot the fleet. SHARP focuses on reducing acquisition performance capability reliability maintainability and readiness of these systems. Mr. Grubb has served as a department director for the past eleven years. His most recent assignment was as director of the Tactical Computer Resources and Test Equipment Department. This organization provided acquisition and in-service engineering support of the Navy's tectical embedded computers priphirals displays and mass memory storage devices. The organization also provided a complete range of product engineering and metrology engineerig services for SSP NevSeaSysCom and the Trident Program. Prior to this appointment Mr. Grubb was deptuy director psysical security programs department and site responsible for program management and system integration of ashore integrated security systems. Mr. Grubb holds a B.S. degree in industrial engineering from Iowa State Universtiy an M.S. degree in industrial engineering from Purdue University and has copleted the course work (Dec.'96 for a master's degree in public and environment affairs at Indiana University-Purdue Univeristy Indianapolis. Dr. Alfred Skolnick operates System Science Consultants (SSC) specilizing in strategic planning technical program definiton technology assessment and engineering analyses on selected matters of national interest. He has taught physics mathematics and management sciences at University of Virginia and Marymount University and is currently adjunct professor of mathematics at Northern Virginia Community College. Form 1985 to 1989 he was president of the American Society of Noval Engineers. Dr. Skolnick served at Applied Physic

In an era of fiscal austerity, downsizing and unforgiving pressure upon human and economic capital, it is an Augean task to identify resources for fresh and creative work. The realities of the day and the practical demands of more immediate fleet needs can often dictate higher priorities. Yet, the Navy must avoid eating its seed corn. Exercising both technical insight and management foresight, the fleet, the R&D community, the Office of the Chief of Naval Operations (OpNav) and the product engineering expertise of the Naval Surface Warfare Center (NSWC) are joined and underway with integrated efforts to marry new, fully demonstrated technologies and operational urgencies. Defense funding today cannot sponsor all work that can be mission-justified over the long term because budgets are insufficient to support product maturation within the classical development cycle. However, by rigorous technical filtering and astute engineering of both marketplace capabilities and currently available components, it is possible in a few select cases to compress and, in effect, integrate advanced development (6.3), engineering development (6.4), weapon procurement (WPN), ship construction (SCN), operation and maintenance (O&M,N) budgetary categories when fleet criticalities and technology opportunities can happily meet. In short, 6.3 funds can be applied directly to ''ripe gateways'' so modern technology is inserted into existing troubled or aging systems, sidestepping the lengthy, traditional development cycle and accelerating practical payoffs to recurrent fleet problems. To produce such constructive results has required a remarkable convergence of sponsor prescience and engineering workforce excellence. The paper describes, extensively, the philosophy of approach, transition strategy, polling of fleet needs, technology assessment, and management team requirements. The process for culling and selecting specific candidate tasks for SHARP sponsorship (matching operational need with t

关键词： Naval vessels

来源：评论

学校读者我要写书评

暂无评论

Virtual memory versus file interfaces for large, memory-intensive scientific applications 96

Virtual memory versus file interfaces for large, memory-inte...

引用

Proceedings of the 1996 ACM/IEEE conference on Supercomputing

作者： Yoonho Park Ridgway Scott Stuart Sechrest High Performance Computing Center University of Houston Houston TX Software Systems Research Laboratory Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI

ISBN: (纸本)9780897918541

Scientific applications often require some strategy for temporary data storage to do the largest possible simulations. The use of virtual memory for temporary data storage has received criticism because of performance problems. However, modern virtual memory found in recent operating systems such as Cenju-3/DE give application writers control over virtual memory policies. We demonstrate that custom virtual memory policies can dramatically reduce virtual memory overhead and allow applications to run out-of-core efficiently. We also demonstrate that the main advantage of virtual memory, namely programming simplicity, is not lost.

关键词： custom policies

来源：评论

学校读者我要写书评

暂无评论

The parallel solution of nonsymmetric sparse linear systems using the H∗ reordering and an associated factorization 94

The parallel solution of nonsymmetric sparse linear systems ...

引用

8th International Conference on Supercomputing, ICS 1994

作者： Gallivan, K.A. Marsolf, B.A. Wijshoff, H.A.G. Center for Supercomputing Research and Development University of Illnois at Urbana-Champaign UrbanaIL61801 United States High Performance Computing Division Department of Computer Science Leiden University Leiden Netherlands

ISBN: (纸本)0897916654

In this paper, a nonsymmetric sparse linear system solver based on the exploitation of multilevel parallelism is proposed. One of the main issues addressed is the application of tearing techniques to enhance large grain parallelism in a manner that maintains reasonable stability. This is accomplished by a combination of a novel reordering technique (H∗) and pivoting strategy. The large grain parallelism exposed by the reordering is combined with medium (various parallel row updates strategies) and fine grain (vectorization) parallelism to allow adaptation to a wide range of multiprocessor architectures. Experimental results are presented which show the effectiveness of the reordering, as well as the stability and efficiency of the solver. © 1994 ACM.

关键词： Linear systems

来源：评论

学校读者我要写书评

暂无评论

distributed network computing over local ATM networks 94

Distributed network computing over local ATM networks

引用

Supercomputing Conference

作者： Mengjou Lin Jehwei Hsieh D.H.C. Du J.P. Thomas J.A. MacDonald Computer Science Department University of Minnesota USA Minnesota Supercomputer Center Inc. Minneapolis MN USA Army High Performance Computing Research Center University of Minnesota USA Computer Science Department University of Minnesota

ISBN: (纸本)9780818666056

Communication between processors has long been the bottleneck of distributed network computing. However recent progress in switch-based high-speed local area networks (LANs) may be changing the situation. Asynchronous transfer mode (ATM) is one of the most widely-accepted and emerging high-speed network standards which can potentially satisfy the communication needs of distributed network computing. We investigate distributed network computing over local ATM networks. We first study the performance characteristics involving end-to-end communication in an environment that includes several types of workstations interconnected via a Fore Systems' ASX-100 ATM Switch. We then compare the communication performance of four different application programming interfaces (APIs). The four APIs were Fore Systems ATM API, BSD socket programming interface, Sun's Remote Procedure Call (RPC), and the Parallel Virtual Machine (PVM) message passing library. Each API represents distributed programming at a different communication protocol layer. We evaluate parallel matrix multiplication over the local ATM network. The experimental results show that network computing is promising over local ATM networks.< >

关键词： computer networks distributed computing Asynchronous transfer mode Communication switching Switches Local area networks high-speed networks Communication standards Workstations Sockets

来源：评论

学校读者我要写书评

暂无评论

A standard simulation environment: a review of preliminary requirements 94

A standard simulation environment: a review of preliminary r...

引用

Simulation Winter Conference

作者： M.A.F. Wagner S. Sevinc O. Tanir P.L. Haigh J.D. Arthur R.E. Nance H.D. Schwetman SysTech Software Solutions Annandale VA USA Knowledge Systems Group Basser University of Sydney Australia Quality Engineering and Research (QER) Group Longueuil QUE Canada High Performance Software Inc. Dayton OH USA Systems Research Center Department of Computer Science Virginia Polytechnic Institute and State University Blacksburg VA USA Mesquite Software Inc. Austin TX USA

ISBN: (纸本)9780780321090

The concept of a simulation environment combines different aspects of the simulation process into one complete powerful tool. The environment should provide the necessary state-of-the-art advances and concepts that a modeler may require during the simulation development and execution process. The "environment" requirements must address the end user needs while minimizing the necessity of learning a new language. The requirements must specify the useful features, independent of different platforms and implementation languages. Hence the opportunity exists for standardization of environments. The paper opens a discussion on challenges and rewards awaiting the simulation community with the advent of a standard environment.

关键词： Power system modeling Computational modeling computer simulation Standardization Knowledge based systems Australia Knowledge engineering Software performance Drives computer science

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：