检索结果-内蒙古大学图书馆

Parallel Algorithms and Applications 1997年第1-3期12卷 5-20页

作者： Balaram Sinharoy[a] Boleslaw K. Szymanski[b] [a] Systems Technology and Architecture IBM Corporation Poughkcepsie NY [b] Department of Computer Science Rensselaer Polytechnic Institute Troy NY

In recent years, high performance computing underwent a deep transformation. In this paper, we review the state of parallel computation with detailed discussion of the current and future research issues in the area of parallel architectures and compilation methods, instruction level parallelism and optimization methods to improve the performance of the memory hierarchy.

关键词： compile-time scheduling instruction level parallelism parallelising compiler parallel programming multiprocessors

来源：评论

学校读者我要写书评

暂无评论

INTRODUCTION: SPECIAL ISSUE ON OPTIMISING COMPILERS FOR PARALLEL LANGUAGES

引用

Parallel Algorithms and Applications 1997年第1-3期12卷 1-4页

作者： Balaram Sinharoy[a] Boleslaw K. Szymanski[b] [a] Systems Technology and Architecture IBM Corporation Potighkeepsie NY [b] Department of Computer Science Rensselaer Polytechnic Institute Troy NY

Click to increase image sizeClick to decrease image size

关键词：

来源：评论

学校读者我要写书评

暂无评论

Using synthetic M-mode imaging to measure temporal differences in left ventricular contraction patterns during ischemia

Using synthetic M-mode imaging to measure temporal differenc...

引用

computers in Cardiology (CinC)

作者： S.H. Lee M.T. Saltzberg A. Neumann J. Wolfer J.S. Soble J. Roberge Department of Computer Science Illinois Institute of Technology Chicago USA Section of Cardiology Rush Medical College Chicago USA

Myocardial ischemia is usually identified by echocardiography using qualitative analysis of regional left ventricular (LV) endocardial excursion. This is often hampered by sub-optimal endocardial visualization and lack of quantitative standards and measurement techniques Ischemia changes the temporal pattern of LV contraction, and a method for quantifying changes in this temporal sequence would have clinical applicability. The authors developed and tested a computer-assisted technique using Synthetic M-mode (SMM) to measure the temporal variations in regional LV contraction. Using a LV short axis image, 6 SMM images were created through a user-defined LV center. The time to peak (t) contraction in 6 LV segments was then measured using the R-wave as a timing reference point. In the authors' study of 15 closed chest dogs with induced regional ischemia, t was significantly prolonged only in those LV segments subtended by the occluded artery. Temporal SMM is a promising technique for defining temporal contraction patterns indicative of regional ischemia.

关键词： Ischemic pain Image segmentation Myocardium Echocardiography Visualization Measurement standards Measurement techniques Testing Time measurement Timing

来源：评论

学校读者我要写书评

暂无评论

The agree predictor: a mechanism for reducing negative branch history interference 97

The agree predictor: a mechanism for reducing negative branc...

引用

Proceedings of the 24th annual international symposium on computer architecture

作者： Eric Sprangle Robert S. Chappell Mitch Alsup Yale N. Patt ROSS Technology Inc. 5316 Hwy 290 West Austin Texas and Advanced Computer Architecture Laboratory Department of Electrical Engineering and Computer Science The University of Michigan Ann Arbor Michigan Advanced Computer Architecture Laboratory Department of Electrical Engineering and Computer Science The University of Michigan Ann Arbor Michigan and ROSS Technology Inc. 5316 Hwy 290 West Austin Texas ROSS Technology Inc. 5316 Hwy 290 West Austin Texas Advanced Computer Architecture Laboratory Department of Electrical Engineering and Computer Science The University of Michigan Ann Arbor Michigan

ISBN: (纸本)9780897919012

Deeply pipelined, superscalar processors require accurate branch prediction to achieve high performance. Two-level branch predictors have been shown to achieve high prediction accuracy. It has also been shown that branch interference is a major contributor to the number of branches mispredicted by two-level *** paper presents a new method to reduce the interference problem called agree prediction, which reduces the chance that two branches aliasing the same PHT entry will interfere negatively. We evaluate the performance of this scheme using full traces (both user and supervisor) of the SPECint95 benchmarks. The result is a reduction in the misprediction rate of gcc ranging from 8.62% with a 64K-entry PHT up to 33.3% with a 1K-entry PHT.

关键词： branch prediction two-level branch prediction speculative execution superscalar

来源：评论

学校读者我要写书评

暂无评论

Detecting common mode failures in N-version software using weakest precondition analysis

Detecting common mode failures in N-version software using w...

引用

Asia-Pacific Conference on Software Engineering

作者： Gwang Sik Yoon Sung Deok Cha Yong Rae Kwon Chan Hyoung Yoo Department of Computer Science Korea Advanced Institute of Science and Technology Taejon South Korea ATM M and A Software Section Electronics and Telecommunications Research Institute Taejon South Korea

ISBN: (纸本)081868271X

An underlying assumption for N-version programming technique is that independently developed versions would fail in a statistically independent manner However empirical studies have demonstrated that common mode failures can occur even for independently developed versions, and that common mode failures degrade system reliability. In this paper, we demonstrate that the weakest precondition analysis is effective in determining input spaces leading to common mode failures. We applied the weakest precondition to the Launch Interceptor Programs which were used in several other experiments related to the N-version programming technique. We detected 13 out of 18 fault pairs which have been known to cause common mode failure. These faults were due to logical flaws in program design. Although the weakest precondition analysis may be labor-intensive since they are applied manually our results convincingly demonstrate that it is effective for identifying input spaces causing common mode failures and further improving the reliability of N-version software.

关键词： Failure analysis Fault detection Software safety computer science Asynchronous transfer mode Degradation Reliability Fault tolerant systems Communication industry computer industry

来源：评论

学校读者我要写书评

暂无评论

A loop transformation for maximizing parallelism from single loops with nonuniform dependencies

A loop transformation for maximizing parallelism from single...

引用

International Conference on High Performance Computing in the Asia-Pacific Region

作者： Chul-Kwon Cho Jae-Chan Shim Mann-Ho Lee ADD Daejeon Taejon South Korea Service Architecture Section Electronics and Telecommunications Research Institute Taejon South Korea Department of Computer Science Chungnam National University Taejon South Korea

This paper describes several loop splitting methods for exploiting parallelism from single loops, and also proposes a generalized and optimal loop transformation technique for exploiting parallelism from single loops with nonuniform dependencies. The proposed algorithm is based on partitioning a serial loop by using the size of dependence distance such that it varies between different instances of the dependence. It outperforms the two methods proposed by C.D. Polychronopoulos (see Compiler optimizations for enhancing parallelism and their impact on architecture design, IEEE Trans. Comput., vol.37, no.8, p. 991-1004, 1988).

关键词： Parallel processing Testing Equations computer architecture computer science Partitioning algorithms Data mining Algorithm design and analysis

来源：评论

学校读者我要写书评

暂无评论

Improved lexicon modeling for continuous speech recognition

Improved lexicon modeling for continuous speech recognition

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Seong Jin Yun Yung Hwan Oh Gyung Chul Shin Department of Computer Science Korea Advanced Institute of Science and Technology Taejon South Korea ATM M&A SIW Section Electronics and Telecommnications Research Institute Taejon South Korea

ISBN: (纸本)0818679190

We propose the stochastic lexicon model which represents the pronunciation variations to optimally cope with the continuous speech recognizer. In this lexicon model, the baseform of words are represented by subword states and the probability distribution of subwords as a hidden Markov model. Also, the proposed approach can be applied to a system employing non-linguistic recognition units and the lexicon is automatically trained from training utterances. In speaker independent speech recognition tests using a 3000 word continuous speech database, the proposed system improves the word accuracy by about 27.8% and the sentence accuracy by about 22.4%.

关键词： Speech recognition Hidden Markov models Stochastic processes Probability distribution Automatic speech recognition Stochastic systems computer science Electronic mail System testing Databases

来源：评论

学校读者我要写书评

暂无评论

Recovery in a shared disk parallel database system using DFeRAM

Recovery in a shared disk parallel database system using DFe...

引用

International Conference on High Performance Computing in the Asia-Pacific Region

作者： Yong Keol Kim Young Sik Baek Seong II Jin Sung Hyun Cho Department of Office Automation Taejon Medical Junior College South Korea System Technology Section Electronics and Telecommunications Research Institute South Korea Department of Computer Science Chungnam National University South Korea Division of Electronics and Computer Engineering Hongik University South Korea

A nonvolatile memory called FeRAM (Ferroelectronic Random Access Memory) is one of the promising memory devices. This paper presents a dual plane FeRAM (DFeRAM) architecture for recovery in the shared disk parallel database system. This paper also presents a recovery algorithm based on a shadow paging method using a DFeRAM architecture. We compare the performance of the log based method and that of our method based on simulation results, which show that our method is better than the log based method when both methods use a nonvolatile memory.

关键词： Database systems Nonvolatile memory Random access memory Ferroelectric films Transaction databases Object oriented modeling Maintenance Discrete event simulation Communication system control Flash memory

来源：评论

学校读者我要写书评

暂无评论

Performance of the H.263 Video Compression Standard

引用

Journal of VLSI Signal Processing Systems for Signal, Image, and Video technology 1997年第2-3期17卷 101-111页

作者： Girod, Bernd Steinbach, Eckehard Färber, Nikolaus Telecommunications Institute University of Erlangen-Nuremberg 91058 Erlangen Cauerstrasse 7 Germany Georgia Institute of Technology United States University of Hannover Germany Inst. F. Theor. N. University of Hannover Germany Massachussetts Inst. of Technology Cambridge MA United States Research Laboratory of Electronics United States Department of Media Technology Media Laboratory Germany Department of Computer Graphics Germany Academy of Media Arts Cologne Germany Computer Science Section Cologne University Germany Digital Signal Processing Group Georgia Institute of Technology Atlanta GA United States Dept. of Elec. Eng./Telecom. University of Erlangen-Nuremberg Germany Telecommunications Institute Germany Vivo Software Inc. Waltham MA United States University of Karlsruhe Germany University of Essex United Kingdom ESIEE Paris France Image Communication Group Telecommunications Institute University of Erlangen-Nuremberg Germany University of Munich Germany Telecommunications Institute University of Erlangen-Nuremberg Germany Image Communication Group

The draft international standard ITU-T H.263 is closely related to the well known and widely used ITU-T Recommendation H.261. However, H.263 does provide the same subjective image quality at less than half the bit-rate. In this paper we investigate to what extend single enhancements of H.263 contribute to this performance gain, and consider the trade-off quality vs. complexity. Based on typical test sequences, H.263 with its various modes is compared to H.261 on the basis of rate distortion curves at bit-rates up to 128 kbps. At 64 kbps, the performance gain of H.263 in its default mode compared to H.261 is approximately 2 dB. This improvement is achieved with only little increase of complexity, and is mainly due to more accurate motion compensation with half-pel accuracy. Considering the trade-off quality vs. complexity, the combination of the optional coding-modes "Advanced prediction mode" and "PB-frames mode" is a good compromise, resulting in an additional performance gain of 1.5 dB PSNR at 64 kbps. The "Syntax-based arithmetic coding mode" on the other hand, offers only a very small performance gain (0.1 dB at 64 kbps) for its increased computational complexity. Results from profiling an H.263 software codec are presented in order to support complexity considerations of the optional coding-modes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Scaling in diffusive transport through membranes

引用

CHEMICAL ENGINEERING science 1996年第17期51卷 4115-4125页

作者： Grzywna, ZJ Department of Physical Chemistry and Technology of Polymers Section of Physics Mathematics and Computer Science Silesian Technical University M. Strzody 9 44-100 Gliwice Poland

Scaling accounts for a sort of dependence between local (microscopic, or very small) and global (macroscopic) properties of a system in question. Fractal dimension and its generalizations, as critical exponents in a power-law scaling, have been chosen as a tool in the reconsideration of a mass transport through membranes, and in the process of casting (DLA membranes). The passage from the diffusion on the prototype structures (e.g. Sierpinski gasket) to the diffusion on real structures has also been shown. Copyright (C) 1996 Elsevier science Ltd

关键词： diffusion sorption membrane scaling fractals generalized dimension

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：