检索结果-内蒙古大学图书馆

2004 ACM/IEEE Conference on Supercomputing, SC 2004

作者： Schulz, Martin Bronevetsky, Greg Fernandes, Rohit Marques, Daniel Pingali, Keshav Stodghill, Paul Center for Applied Scientific Computing Lawrence Livermore National Laboratory LivermoreCA94551 United States Department of Computer Science Cornell University IthacaNY14853 United States

ISBN: (纸本)0769521533

The running times of many computational science applications are much longer than the mean-time-to-failure of current high-performance computing platforms. To run to completion, such applications must tolerate hardware failures. Checkpoint-and-restart (CPR) is the most commonly used scheme for accomplishing this - the state of the computation is saved periodically on stable storage, and when a hardware failure is detected, the computation is restarted from the most recently saved state. Most automatic CPR schemes in the literature can be classified as system-level checkpointing schemes because they take core-dump style snapshots of the computational state when all the processes are blocked at global barriers in the program. Unfortunately, a system that implements this style of checkpointing is tied to a particular platform;in addition, it cannot be used if there are no global barriers in the program. We are exploring an alternative called application-level, non-blocking checkpointing. In our approach, programs are transformed by a pre-processor so that they become self-checkpointing and self-restartable on any platform;there is also no assumption about the existence of global barriers in the code. In this paper, we describe our implementation of application-level, non-blocking checkpointing. We present experimental results on both a Windows cluster and a Compaq Alpha cluster, which show that the overheads introduced by our approach are small. © 2004 IEEE.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Estimate of error bounds in the improved support vector regression

引用

自然科学进展（英文版） 2004年第4期14卷 362-364页

作者： SUN Yanfeng LIANG Yanchun WU Chunguo Yang Xiaowei LEE Heow Pueh LIN Wu Zhong College of Computer Science and Technology Jilin UniversityChangchun 130012China College of Computer Science and Technology Jilin UniversityChangchun 130012China Institute of High Performance Computing Singapore 117528Singapore Department of Applied Mathematics South China University of TechnologyGuangzhou 510640China Centre for ACES Department of Mechanical EngineeringNational University of Singapore119260Singapore Institute of High Performance Computing Singapore 117528Singapore

An estimate of a generalization error bound of the improved support vector regression(SVR)is provided based on our previous *** boundedness of the error of the improved SVR is proved when the algorithm is applied to t... 详细信息

关键词： support vector regression(SVR) generalization error approximation error estimation error

来源：评论

学校读者我要写书评

暂无评论

Application-level checkpointing for shared memory programs

Application-level checkpointing for shared memory programs

引用

作者： Bronevetsky, Greg Marques, Daniel Pingali, Keshav Szwed, Peter Schulz, Martin Department of Computer Science Cornell University Ithaca NY 14853 School of Electrical and Computer Engineering Cornell University Ithaca NY 14853 Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore CA 94551

Trends in high-performance computing are making it necessary for long-running applications to tolerate hardware faults. The most commonly used approach is checkpoint and restart (CPR) - the state of the computation is saved periodically on disk, and when a failure occurs, the computation is restarted from the last saved state. At present, it is the responsibility of the programmer to instrument applications for CPR. Our group is investigating the use of compiler technology to instrument codes to make them self-checkpointing and self-restarting, thereby providing an automatic solution to the problem of making long-running scientific applications resilient to hardware faults. Our previous work focused on message-passing programs. In this paper, we describe such a system for shared-memory programs running on symmetric multiprocessors. This system has two components: (i) a pre-compiler for source-to-source modification of applications, and (ii) a runtime system that implements a protocol for coordinating CPR among the threads of the parallel application. For the sake of concreteness, we focus on a non-trivial subset of OpenMP that includes barriers and locks. One of the advantages of this approach is that the ability to tolerate faults becomes embedded within the application itself, so applications become self-checkpointing and self-restarting on any platform. We demonstrate this by showing that our transformed benchmarks can checkpoint and restart on three different platforms (Windows/x86, Linux/x86, and Tru64/Alpha). Our experiments show that the overhead introduced by this approach is usually quite small;they also suggest ways in which the current implementation can be tuned to reduced overheads further. Copyright 2004 ACM.

关键词： computer systems programming

来源：评论

学校读者我要写书评

暂无评论

Towards a formal model of CyberRail

引用

IFIP 18th World computer Congress Topical Sessions

作者： Bjørner, Dines Chiang, Peter Jacobsen, Morten S.T. Hansen, Jens Kielsgaard Madsen, Michael P. Penicka, Martin Department of Computer Science School of Computing National University of Singapore 3 Science Drive 2 Singapore 117543 Singapore Section of Computer Science and Engineering Institute of Informatics and Computer Engineering Technical University of Denmark DK-2800 Kgs Lyngby Denmark Czech Technical University Department of Applied Mathematics Na Florenci 25 110 00 Praha 1 Czech Republic

ISBN: (纸本)1402081561

Based on a number of reports and publications, primarily by Takahiko Ogino [14], [15], [16] (in these proceedings), and [17], on the emerging concept of CyberRail, we attempt to show what a formal domain model of CyberRail might look like, and what benefits one might derive from establishing and having such a formal model. © 2004 Springer science + Business Media, Inc.

关键词： Abstract data types

来源：评论

学校读者我要写书评

暂无评论

SPLIT: A priority based TCP-friendly video protocol

SPLIT: A priority based TCP-friendly video protocol

引用

IEEE Symposium on computers and Communications (ISCC)

作者： N.K. Chilamkurti B. Soh A. Aitham Applied Computing Research Institute Department of Computer Science and Computer Engineering La Trobe University Melbourne Australia

In this paper, priority dropping mechanism (PDM) is applied to RED algorithm to protect the video layers selectively from performance and throughput deterioration. Moreover, to achieve fairness among different type of the network traffic with PDM, packet classification scheme (PCS) is implemented to assign the priority to each SPLIT layer or packet based upon the target rate.

关键词： Protocols Videoconference Internet Multicast algorithms Scalability Feedback Streaming media Forward error correction computer science Application software

来源：评论

学校读者我要写书评

暂无评论

Early user involvement in the development of information technology-related products for older people

引用

Universal Access in the Information Society 2004年第2期3卷 131-140页

作者： Eisma, R. Dickinson, A. Goodman, J. Syme, A. Tiwari, L. Newell, A. F. Division of Applied Computing University of Dundee Dundee UK University of Glasgow Department of Computer Science Glasgow UK ICCAVE University of Abertay Dundee Dundee UK

The UTOPIA project (Usable Technology for Older People—Inclusive and Appropriate) is focused on developing effective methods for the early involvement of older people in the development of information technology-related products for people aged 60 and over, and on providing industry with tools to assist in the development of information technology products for such older people. An essential part of this methodology is building a diverse user base, forming a long-lasting partnership with older people, and developing approaches for effective interaction with this target user group. Our experiences with eliciting information from groups of older people about technology is described, together with a report on seminars for Scottish industry designed to raise an awareness of these issues.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reversible n-bit to n-bit integer Haar-like transforms

Reversible n-bit to n-bit integer Haar-like transforms

引用

Data Compression Conference (DCC)

作者： J. Senecal M. Duchaineau K.I. Joy Center for Image Processing and Integrated Computing Computer Science Department University of California Davis USA Center for Applied Scientific Computing Lawrence Livemore National Laboratory USA

Methods of producing reversible n-bit to n-bit integer transforms is presented in this paper. Such methods are particularly suited for hardware-based implementations, as keeping the coefficients to n bits simplifies the design of custom hardware and makes it easier to use these approaches on preexisting hardware with limited channel width. One of our methods, called table-lookup Haar (TLHaar) is an approximation of the Haar integer wavelet transform (Haar IWT). The Haar IWT takes two integer data values A and B and using averaging and differencing produces a low-pass value L and a high-pass value H, both integers. The lookup tables are created by initializing each with an identity transform. TLHaar executes up to 44% faster. Compression results are mixed, and depend on the compression method used and the image type.

关键词： Image coding Table lookup Scientific computing Hardware Laboratories Image processing computer science Wavelet transforms Equations Convergence

来源：评论

学校读者我要写书评

暂无评论

The Healthgrid White Paper 3

The Healthgrid White Paper

引用

3rd Annual Conference of the Healthgrid Association, Healthgrid 2005

作者： Breton, Vincent Dean, Kevin Solomonides, Tony Aloisio, Giovanni Benkner, Siegfried Bilofsky, Howard Blanquer, Ignacio Brady, Michael Cannataro, Mario Chouvarda, Ioanna Claerhout, Brecht De Moor, Georges De Neve, Wilfried De Wagter, Carlos Fiore, Sandro Hassan, Kinda Heeren, Germaine Hernández, Vicente Herveg, Jean A.M. Hofmann, Martin Jones, Chris Koutkias, Vassilios Lloyd, Sharon Lonsdale, Guy Alonso, Victoria L. Maglaveras, Nicos Maigne, Lydia Malousi, Andigoni Martín-Sánchez, Fernando McClatchey, Richard Medico, Enzo Miguet, Serge Mirto, Maria Montagnat, Johan Nørager1, Sofie Nozaki, Kazunori Oliveira, Ilídio Castro Pennec, Xavier Poullet, Yves Merino, Juan Pedro Sánchez Strizh, Irina G. Taillet, Michel Tristram, Clive Tverdokhlebov, Nikolay Veltri, Pierangelo Ziegler, René CNRS-IN2P3 LPC Campus des Cézeaux AubiereCedex63177 France Internet Business Solutions Group Cisco Systems 9 New Square Bedfont Lakes Middlesex FelthamTW14 8HA United Kingdom University of the West of England Bristol CEMS Coldharbour Lane BristolBS6 6TH United Kingdom Center for Advanced Computational Technologies ISUFI Dept. of Innovation Engineering University of Lecce Via per Monteroni Lecce73100 Italy Institute of Scientific Computing University of Vienna Nordbergstrasse 15 ViennaA-1090 Austria University of Pennsylvania School of Engineering and Applied Science Computer and Information Science Center for Bioinformatics 1416 Blockley Hall 423 Guardian Drive PhiladelphiaPA19104-6021 United States Universidad Politecnica de Valencia Camino de Vera/n Valencia46022 Spain FRS FREng Dept. of Engineering Science Oxford University Parks Road OxfordOX1 3PJ United Kingdom Magna Graecia University of Catanzaro School of Bioinformatics and Biomedical Engineering Campus di Germaneto Viale Europa Germaneto Catanzaro88100 Italy Aristotle University The Medical School Lab of Medical Informatics Box 323 Thessaloniki54124 Greece Custodix NV Verlorenbroodstraat 120 Bus 14 MerelbekeB-9820 Belgium Research in Advanced Medical Informatics and Telematics UZ Gent 5K3 De Pintelaan 185 GentB9000 Belgium Department of Radiotherapy Ghent University Hospital Building P7 De Pintelaan 185 GentB-9000 Belgium Laboratoire LIRIS Université Lumière Lyon 2 Campus Porte des Alpes 5 avenue Pierre Mendès France Bron69676 France av. E.Mounierlaan 83 Brussels1200 Belgium Faculté de Droit de Namur - FUNDP Centre de Recherches Informatiques & Droit 5 rempart de la Vierge Namur5000 Belgium Schloss Birlinghoven Sankt Augustin53754 Germany CERN Geneva 231211 Switzerland Oxford University Computing Laboratory Wolfson Building OxfordOX1 3QD United Kingdom C&C Research Laboratories NEC Europe Ltd. Rathausallee 10 Sankt AugustinD-53757 Germany Medic

ISBN: (纸本)158603510X

Over the last four years, a community of researchers working on Grid and High Performance computing technologies started discussing the barriers and opportunities that grid technologies must face and exploit for the development of health-related applications. This interest lead to the first Healthgrid conference, held in Lyon, France, on January 16th–17th, 2003, with the focus of creating increased awareness about the possibilities and advantages linked to the deployment of grid technologies in health, ultimately targeting the creation of a European/international grid infrastructure for health. The topics of this conference converged with the position of the eHealth division of the European Commission, whose mandate from the Lisbon Meeting was "To develop an intelligent environment that enables ubiquitous management of citizens’ health status, and to assist health professionals in coping with some major challenges, risk management and the integration into clinical practice of advances in health knowledge." In this context "Health" involves not only clinical procedures but covers the whole range of information from molecular level (genetic and proteomic information) over cells and tissues, to the individual and finally the population level (social healthcare). Grid technology offers the opportunity to create a common working backbone for all different members of this large "health family" and will hopefully lead to an increased awareness and interoperability among disciplines. The first HealthGrid conference led to the creation of the Healthgrid association, a non-profit research association legally incorporated in France but formed from the broad community of European researchers and institutions sharing expertise in health grids. After the second Healthgrid conference, held in Clermont-Ferrand on January 29th–30th, 2004, the need for a "white paper" on the current status and prospective of health grids was raised. Over fifty experts from different areas of grid tec

关键词： Health risks

来源：评论

学校读者我要写书评

暂无评论

Length-limited variable-to-variable length codes for high-performance entropy coding

Length-limited variable-to-variable length codes for high-pe...

引用

Data Compression Conference (DCC)

作者： J. Senecal M. Duchaineau K.I. Joy Institute for Scientific Computing Research Lawrence Livemore National Laboratory USA Center for Image Processing and Integrated Computing Computer Science Department University of California Davis USA Center for Applied Scientific Computing Lawrence Livemore National Laboratory USA

Arithmetic coding achieves a superior coding rate when encoding a binary source, but its lack of speed makes it an inferior choice when true high-performance encoding is needed. This paper presents the practical implementation of fast entropy coders for binary messages utilizing only bit shifts and table lookups. To limit code table size the proposed code lengths is limited with a type of variable-to-variable (VV) length code created from source string merging. This is referred to as "merged codes". With merged codes it is possible to achieve a desired level of speed by adjusting the number of bits read from the source at each step. The most efficient merged codes yield a coder with a worst-case inefficiency of 0.4%, relative to the Shannon entropy. Using a hybrid Golomb-VV bin coder the compression ratio that is competitive with other state-of-the-art coders, at a superior throughput is achieved.

关键词： Entropy coding Image coding Table lookup Throughput Computational modeling Scientific computing Laboratories Degradation Arithmetic Image processing

来源：评论

学校读者我要写书评

暂无评论

Range image registration via probability field

Range image registration via probability field

引用

computer Graphics International

作者： H. Zhang O. Hall-Holt A. Kaufman Center for Visual Computing (CVC) Computer Science Department Stony Brook University Stony Brook NY USA Center for Visual Computing (CVC) Department of Applied Mathematics and Statistics Stony Brook University Stony Brook NY USA

This paper presents a powerful variant of the ICP (iterative closest point) algorithm for registering range images using a probability field. The probability field (p-field) represents the probability distribution of the surface position. By capitalizing on the properties of the range image, fast construction, compact representation, and efficient query of the p-field can be achieved. Different sensor models are supported by the p-field according to the properties of the range image. Range images can be precisely aligned by maximizing the probability of overlapping surfaces via the p-field

关键词： Image registration Iterative closest point algorithm Iterative algorithms Image reconstruction Iterative methods computer science Mathematics Statistical distributions Probability distribution Image sensors

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：