检索结果-内蒙古大学图书馆

6th International Conference on Parallel Interconnects, PI 1999

作者： Carrera, Enrique V. Bianchini, Ricardo COPPE Systems Engineering Federal University of Rio de Janeiro Rio de Janeiro21945-970 Brazil Department of Computer Science University of Rochester RochesterNY14627 United States

ISBN: (纸本)076950440X

In this paper we propose a simple extension to the optical network of a scalable multiprocessor that optimizes page swap-outs significantly. More specifically, we propose to extend the network with an optical ring that not only transfers swapped-out pages between the local memories and the disks of the multiprocessor but also acts as a system-wide write cache for these pages. This extended optical network has several performance benefits: it provides a staging area where swapped-out pages can reside until the disk is free;it increases the possibility of combining several writes to disk;and it acts as a victim cache for pages that are swapped out and subsequently accessed by the same or a different processor. In order to evaluate the extent to which these benefits affect performance, we use detailed execution-driven simulations of several out-of-core parallel applications running on an 8-node scalable multiprocessor. Our results demonstrate that our optical ring provides consistent performance improvements, coming mostly from faster page swap-outs and victim caching. Based on these results and on our parameter space study, our main conclusion is that our optical ring is highly efficient under several architectural assumptions and for most out-of-core parallel applications. © 1999 IEEE.

关键词： Fiber optic networks

来源：评论

学校读者我要写书评

暂无评论

A comparison of training algorithms for DHP adaptive critic neurocontrol

A comparison of training algorithms for DHP adaptive critic ...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： G.G. Lendaris T.T. Shannon A. Rustan Systems Science Ph.D. Program and Electrical & Computer Engineering Department Portland State University USA

A variety of alternate training strategies for implementing the dual heuristic programming (DHP) method of approximate dynamic programming in the neurocontrol context are explored. The DHP method of controller training has been successfully demonstrated by a number of authors on a variety of control problems in recent years, but no unified view of the implementation details of the method has yet emerged. A number of options are described for sequencing the training of the controller and critic networks in DHP implementations. Results are given about their relative efficiency and the quality of the resulting controllers for two benchmark control problems.

关键词： Neural networks Shape control Dynamic programming Equations Design optimization Amorphous materials Design methodology Feedback loop Supervised learning Process design

来源：评论

学校读者我要写书评

暂无评论

Nomad: a scalable operating system for clusters of uni- and multiprocessors

Nomad: a scalable operating system for clusters of uni- and ...

引用

IEEE International Conference on Cluster Computing

作者： E. Pinheiro R. Bianchini COPPE Systems Engineering Federal University of Rio de Janeiro Rio de Janeiro Brazil Department of Computer Science University of Rochester Rochester NY USA

The recent improvements in workstation and interconnection network performance have popularized the clusters of off-the-shelf workstations. However, the usefulness of these clusters is yet to be fully exploited, mostly due to the inadequate management of cluster resources implemented by current distributed operating systems. In order to eliminate this problem and approach the computational power of large clusters of workstations, in this paper we propose Nomad, an efficient operating system for clusters of uni and/or multiprocessors. Nomad includes several important characteristics for modern cluster-oriented operating systems: scalability, efficient resource management across the cluster, efficient scheduling of parallel and distributed applications, distributed I/O, fault detection and recovery, protection, and backward compatibility. Some of the mechanisms used by Nomad, such as process checkpointing and migration, can be found in previously proposed systems. However, our system stands out for its strategy for disseminating information across the cluster and its efficient management of all cluster resources. In addition, Nomad is highly scalable as it uses neither centralized control nor extra messages to implement its functionality, taking advantage of the I/O traffic associated with its distributed file system. Our preliminary evaluation of the load balancing aspect of Nomad shows that the pattern of file accesses in our distributed Ale system allows for efficient and scalable load balancing. Our main conclusion is that the complete implementation of Nomad will most likely be efficient and will be a nice platform for future research on operating systems for clusters of workstations.

关键词： Operating systems Workstations Resource management Load management Multiprocessor interconnection networks Power system management Scalability Processor scheduling Fault detection Protection

来源：评论

学校读者我要写书评

暂无评论

Recurrent neural gas in electric load forecasting

Recurrent neural gas in electric load forecasting

引用

International Joint Conference on Neural Networks (IJCNN)

作者： M.A. Teixeira G. Zaverucha V.N.A.L. da Silva G.F. Ribeiro Systems Engineering and Computer Science Program-Graduate Programs in Engineering Federal University of Rio de Janeiro Brazil Electric Power Research Center Brazil

We have proposed for the task of hourly electric load forecasting a hybrid neural system combining unsupervised and supervised learning. The system consists of a recurrent neural gas (RNG) network and many Elman neural networks (ENs). RNG is a modification we introduced in the neural gas (NG) network in order to enable it to do clustering using a sequence of input data. For verifying the RNG's performance, many architectures are compared in the learning of global and local models. In a global model only one supervised network is trained and in a local model the training examples are grouped by a clustering algorithm and each one of these groups is sent to different supervised networks. These architectures use different clustering algorithms (NG and RNG) or different supervised networks for prediction (ENs that are trained by backpropagation or backpropagation through time, and feedforward networks).

关键词： Load forecasting Clustering algorithms Neural networks Predictive models Power engineering and energy Backpropagation algorithms Supervised learning systems engineering and theory computer science Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Definition of evaluation criteria for system development acquisition modeling and simulation

引用

NAVAL ENGINEERS JOURNAL 1999年第1期111卷 55-64页

作者： Leite, MJ Mensh, DR Michael J. Leite:is a Principal Engineer with PRC Inc. a division of Litton Industries. He supports combat system engineering for theater air and missile defense. His other tasks have included the command and control for the AEGIS shipbuilding program systems engineering for the 21st Century Surface Combatant combat system survivability and the development of NATO standardization agreements for naval ordnance. He was previously a Senior Engineer with San Diego Gas & Electric with responsibility for its energy application and lighting programs. Prior to joining SDG&E Mr. Leite was a commissioned officer in the U.S. Navy where he served in operations and engineering assignments. Following active duty he accepted a Naval Reserve commission and has retired with the rank of Captain. His assignments included command operational and engineering tours. Mr. Leite has also served as an expert witness in admiralty and engineering matters. He is a gradate of the University of California Berkeley with a Bachelor of Science Degree in Engineering and also holds a Masters Degree in Business Administration from National University in San Diego. Mr. Leite is a Registered Professional Engineer in the States of California and Minnesota. Mr. Leite is a member of ASNE ASCE MORS the Illuminating Engineering Society and the U.S. Naval institute. Dennis Roy Mensh:is a Senior Engineer with PRC Inc. a division of Litton Industries in Crystal City VA where he supports modeling and simulation tasking for combat systems. He received BS and MS degrees in applied Physics from Lopola College in Baltimore MD and the American University in Washington DC. He has also completed the course work towards a Ph.D. degree in computer science specializing in the fields of Operations Reseurch Anabsis Systems Analysis and Computer Modeling and Simulation. Previously he was employed at the White Oak Laboratory of the Naval Surface Warfare Carter in Silver Spring MD where he worked in the areas of naval sensor and weapon system analysis

This paper defines, develops and examines a set of generic analysis tools that can be applied to Models and Simulations at the systems engineering level of fidelity. The tools examine the performance and effectiveness of Sensors;Weapons;and Battle Management, Command, Control, Communications, computers, and Intelligence ((BMCI)-I-4) systems and equipment. The Measures of Performance (MOPs), Measures of Effectiveness (MOEs) and Measures of Force Effectiveness (MOFEs) were extracted from the Modular Command and Control Structure Paradigm which was developed at the Naval Postgraduate School. The paradigm provides for the development of evaluation criteria (MOPs, MOEs, and MOFEs) in a framework that ensures the traceability of system performance and effectiveness to the system operational requirements as specified in the Operational Requirements Document (ORD). Also, the analysis tools provide insight and valid estimates of numerical measures of the defined system functionality threads, which represent the system's operational requirements as specified in the ORD. The tools are directly transferrable and applicable to test and evaluation exercise events which are conducted in support of the development and acquisition of systems and equipment. Once the levels of system performance have been defined, the Paradigm generates a quantitative database that becomes a useful tool in system tradeoffs and selection. Once the alternative system suites have been defined, the suites can be analyzed in terms of system functionality threads and their corresponding performance capabilities versus cost.

关键词： Naval vessels

来源：评论

学校读者我要写书评

暂无评论

Modelling combat system manning and manpower reduction

引用

NAVAL ENGINEERS JOURNAL 1999年第3期111卷 153-171页

作者： Roos, CH Carl H. Roos:is a Senior Engineer with Logicon-Syscon. A graduate of the University of Pittsburgh with a BSEE degree he has over 35 years experience in functional operational combat system fire control interface and computer program design. As technology changed and the combat system was upgraded Mr. Roos maintained his level of technical expertise by taking graduate-level courses in computer science modelling & structured analysis networking & fiber optics. He has worked in various capacities on the LHD LHA DDG 993 DD 963 LPD 17 LCC LPD 13 CGN 38 CGN 9 CG 26 and the DDG Class Combat Systems. In recent years Mr. Roos has been responsible for managing directing and performing engineering design and analysis efforts associated with Battle Management Organization (BMO) functional analysis and operational analysis. These efforts were used in defining combat system operational requirements shipboard space requirements and integration requirements. His paper “Configuration Management of Digital Programs” was published at the 1972 IEEE Southeastern Conrence.

NAVSEA 03K41 is responsible for generating Combat System Rattle Management Organizations (BMO) and Functional Flow Diagrams (FFD). Several years ago, NAVSEA provided the resources to conduct a functional analysis that would support the development and validation of the BMOs and FFDs. The major obstacle in performing the analysis was obtaining a consensus on how the functional hierarchy was to be structured. The non-optimum organization of the hierarchy was selected;as a result, the functions were difficult to define, find, use, and validate. Recognizing the shortcomings of this effort, research was conducted to evaluate state-of-the-art structured modelling techniques, concepts, and methodologies. Two modelling concepts by James Martin were found to be applicable for the combat system functional analysis: Enterprise Modelling Concept and Functional Decomposition Modelling Concept. The Structure Modelling definitions of Whitten, Bently, and Barlow provided the guidelines for using the Martin concepts. During the ensuing BMO and FFD development efforts, a Ship's Combat System (SCS) Modelling concept evolved and a SCS Model was developed. This paper addresses how the modelling concepts and tools are used in the BMO and FFD development and validation process. Data from the SCS Model provides the basis for defining combat system requirements (e.g., software, data display, database, networking, etc.).

关键词： Naval vessels

来源：评论

学校读者我要写书评

暂无评论

Acquisition of the enhanced Maritime Prepositioning Force, MPF(E), Ship

引用

NAVAL ENGINEERS JOURNAL 1999年第1期111卷 15-26页

作者： Mazumdar, T Sandison, J DuFlocq, R Thpan Mazurndar:is a Deputy Ship Design Manager from Naval Surface Warfare Center Carderock Division. He was the Design Manager for the Engineering and Design Phase ofthe Enhanced Maritime Prepositioning Fme Ship (MPF(E)). Cuwently Mt: M2.zumdar is Deputy Ship Design Manager in NAVSEA 030 for the LHD and LSD classes of ships. Mr. Mazumdar was awarded a BSE in Naval Architecture and Marine Engineering from the University of Michigan in 1981 and an MS in Computer Science fiom Virginia Tech in 1988. He also possesses an International ChiejEngineer License (Motot unlimited). Mr. Mancmdar has been an Associate Member of WAME since 1981. James C. Sandison:is a Senior Ship Design Manager (SDM) for surface ships in the Nawl Sea Systems Command (NAVSEA03D3C). He has a long tem practical background in design construction and testing of surface ships. Starting in NAVSEC as a diving and salvage systems engineer for the Bathyscaph TRIESTE he has been involved with most ofthe Nays Mine Warfare and Auxiliary ships from naval architect to the present Senior Ship Design Manager for the Strategic Sealift Program. Mr: Sandison entered the U.S. Navy in 1956 in the (diesel) Submarine Service. Afterward he “went to sea in sailing ships and completed an apprenticeship as a wooden shipbuildex. In 1971 he graduated from the University of Michigan with a BSE in Naval Architecture and Marine Engineering and was employed by NAVSEC. His latest additional duties have been as engineering coordinator on board the USS CONSTITUTION for her 200th anniversary sail. Mr. Sandison is a member of SNAME ASNE and the U.S. Naval Institute. Rick DuFlocq has over 24 years of combined project management mechanical system and general awangmmt design and design documentation experience the last 14 years urith Advance Marine Enterprises (AME). He has fmjmned this wurk on a wide variety of ships and auxiliary crafl including submarines primarily for the U.S. Navy and the Military Sealift Command (MSC). He has worked for both shiprards and marine eng

The paper will describe the streamlined acquisition process involved in the procurement, and conversion, of the first two of three Enhanced Maritime Prepositioning Force (MPF(E)) ships. This program was one of the first programs undertaken within the Government's new policy of Acquisition Reform, which resulted in the development of "performance based" requirements for these ships. This program is notable in that one prime contractor is responsible for the accomplishment of all phases, and that the contractors participating were not shipyards as is usually the fashion for Government ship acquisition programs. Also of note, was that after the conversion contracts were awarded, responsibility for the conduct of detail design, conversion, and operation and maintenance of the ship was transferred from the NAVSEA Sealift program Office (PMS 385) to the Military Sealift Command (MSC). The first part of the paper will describe the basic mission of the MPF(E) ships, and a description of the origin of the program requirements. The second part of the paper will chronicle in detail the portions of the engineering design and specification development process, which will include descriptions of the unique digital data recording and tracking systems developed by the Government MPF(E) Design Support Team to support the acquisition phases of the procurement. The third and final part of the paper will elaborate on the conversion contract awards and the transition of the program from PMS 385 to MSC.

关键词： Naval vessels

来源：评论

学校读者我要写书评

暂无评论

Understanding the performance of DSM applications 1st

引用

1st Workshop on Communication and Architectural Support for Network-based Parallel Computing, CANPC 1997

作者： Meira, Wagner Leblanc, Thomas J. Hardavellas, Nikolaos Amorim, Clgudio Department of Computer Science University of Rochester RochesterNY14627 United States COPPE Systems Engineering UFRJ Rio de Janeiro21945-970 Brazil

ISBN: (纸本)3540625739

Carnival is a performance measurement and analysis tool that assists users in understanding the performance of DSM applications and protocols. Using traces of program executions, Carnival presents performance data as a hierarchy of execution profiles. During analysis, Carnival automates the inference process that relates performance phenomena to specific causes in the source code or DSM protocol using techniques that focus on the two most important sources of overhead in DSM systems: waiting time analysis identifies the causes of synchronization overhead, and produces an explanation for each source of waiting time in the program;communication analysis identifies the sequence of requests that result in invalidations, and produces an explanation for each source of communication. We describe these techniques and their implementation in TreadMarks, and show how to use waiting time analysis and communication analysis to improve the running time of two programs froIn the SPLASH application snite when executed on DEC Alphas connected by a DEC Memory Channel network. © Springer-Verlag Berlin Heidelberg 1997.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Incorporating contextual information in white blood cell identification 97

Incorporating contextual information in white blood cell ide...

引用

Proceedings of the 11th International Conference on Neural Information Processing systems

作者： Xubo Song Yaser Abu-Mostafa Joseph Sill Department of Electrical Engineering California Institute of Technology Pasadena CA Dept. of Electrical Engineering and Dept. of Computer Science California Institute of Technology Pasadena CA Computation and Neural Systems Program California Institute of Technology Pasadena CA

In this paper we propose a technique to incorporate contextual information into object classification. In the real world there are cases where the identity of an object is ambiguous due to the noise in the measurements based on which the classification should be made. It is helpful to reduce the ambiguity by utilizing extra information referred to as context, which in our case is the identities of the accompanying objects. This technique is applied to white blood cell classification. Comparisons are made against "no context" approach, which demonstrates the superior classification performance achieved by using context. In our particular application, it significantly reduces false alarm rate and thus greatly reduces the cost due to expensive clinical tests.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Empirical testing: The prototyper's evaluation mechanism

引用

NAVAL ENGINEERS JOURNAL 1997年第1期109卷 41-46页

作者： Hafner, AN Arnold N. Hafner Ph.D.:is founder and president of Information Systems Research (ISR). He has twenty-five years of experience in systems development and is published in the field of systems development management. He served as corporate research scientist at Systems Exploration Inc. from 1988 to 1991 program director at Computer Science Corporation from 1983 to 1988 director of operations at Republic Management Systems Corporation from 1981 to 1983 and program manager at Computer Science Corporation from 1972 to 1981. A 1962 graduate of the US. Naval Academy he holds a doctoral degree in human behavior and engineering degrees in electronics and communications. He has taught courses on information systems and systems management at most of the colleges in the San Diego area. Dr. Hafner has presented fourteen refereed research papers while publishing sixteen articles and a book A Manager's Guide to Software System Development.

Evaluating complex systems is the subject of this paper, the third in a series investigating prototyping. It provides an interesting and helpful overview of how to evaluate systems prototypes and outlines the iterative stream of developer-user interactions that is replacing older approaches to testing and evaluating new military systems, which promise to reduce the time required to develop and field future military capabilities. Changes to the acquisition process, such as those the paper sketches, will facilitate the nation's rapid transit through its current revolution in military affairs.

关键词： systems analysis

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：