检索结果-内蒙古大学图书馆

International conference on parallel Processing (ICPP)

作者： D. Krishnaswamy P. Banerjeer Center for Reliable & High Performance Comput. Illinois Univ. Urbana IL USA

Edge detection and shape detection in digital images are very computationally intensive problems. parallel algorithms can potentially provide significant speedups while preserving the quality of the result obtained. Hough and Radon Transforms are projection-based transforms which are commonly used for edge detection and shape detection respectively. We propose in this paper various new parallel algorithms which exploit both task and data parallelism available in Hough and Radon transforms algorithms. A memory scalable aggressive task parallel algorithm is shown to be the most optimal algorithm in terms of memory scalability and performance on an IBM SP2.

关键词： Image edge detection Pixel parallel processing Shape parallel algorithms Concurrent computing distributed computing Digital images Scalability Contracts

来源：评论

学校读者我要写书评

暂无评论

9th International conference on Modeling Techniques and Tools for Computer Performance Evaluation, 1997

9th International Conference on Modeling Techniques and Tool...

引用

9th International conferences on Modeling Techniques and Tools for Computer Performance Evaluation, 1997

ISBN: (纸本)9783540631019

The proceedings contain 18 papers. The special focus in this conference is on Modeling Techniques and Tools for Computer Performance Evaluation. The topics include: A performability modeling environment tool;dependability evaluation and the optimization of performability;design and implementation of a network computing platform using JAVA;storage alternatives for large structured state spaces;an efficient disk-based tool for solving very large markov models;efficient transient overload tests for real-time systems;towards an analytical tool for performance modeling of ATM networks by decomposition;an embedded network simulator to support network protocols’ development;synchronized two-way voice simulation for internet phone performance analysis and evaluation;processes as language-oriented building blocks of stochastic petri nets;measurement tools and modeling techniques for evaluating WEB server performance;workload characterization of input/output intensive parallel applications;interval based workload characterization for distributed systems;bounding the loss rates in a multistage ATM switch;simple bounds for queues fed by markovian sources and on queue length moments in fork and join queuing networks with general service times.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A fault tolerant protocol as an extension to a distributed mutual exclusion algorithm

A fault tolerant protocol as an extension to a distributed m...

引用

International conference on parallel and distributed Systems (ICPADS)

作者： V. Revannaswamy P.C.P. Bhatt Greystone Software Technology Woburn MA USA Kochi University of Technology Japan

In this paper, we present an extension to Kerry Raymond's tree based distributed mutual exclusion algorithm to make it fault tolerant. The algorithm uses a spanning tree of a computer network. With N nodes in the network the number of messages exchanged per critical section entry is typically O(log N). The novelty of the algorithm lies in each node holding information only about its immediate neighbors in the spanning tree. We basically present a protocol to make the algorithm tolerant to single node/link failure and associated network partition. This protocol assumes that the graph of the underlying network is biconnected. Basically we attempt reconfiguration of the tree after isolating the failed components. With this enhancement a significant number of cases in which the algorithm would have halted due to lack of communication paths, can be effectively handled by utilizing alternative communication paths now available, thus continuing to provide the mutual exclusion service.

关键词： Fault tolerance Protocols Partitioning algorithms Tree graphs Computer networks Software algorithms Drives distributed algorithms distributed computing Protection

来源：评论

学校读者我要写书评

暂无评论

Fault recovery for distributed shared memory systems

Fault recovery for distributed shared memory systems

引用

IEEE conference on Aerospace

作者： W.R. Dieter J.E. Lumpp Department of Electrical Engineering University of Kentucky Lexington KY USA

distributed Shared Memory (DSM) offers programmers a shared memory abstraction on top of an underlying network of distributed memory machines. advances in network technology and price/performance of workstations suggest that DSM will be the dominant paradigm for future high-performance computing. However, as long running DSM applications scale to hundreds or even thousands of machines, the probability of a node or network link failing increases. Fault tolerance is typically achieved via "checkpointing" techniques that allow applications to "roll back" to a recent checkpoint rather than restarting. High-performance DSM systems using relaxed memory consistency are significantly more difficult to checkpoint than uniprocessor or message passing architectures. This paper describes previous approaches to checkpointing message passing parallel programs along with extensions to DSM systems.

关键词： Workstations Message passing Checkpointing Programming profession Large-scale systems Computer networks High performance computing Fault tolerance parallel architectures Scalability

来源：评论

学校读者我要写书评

暂无评论

A multiprocessor scheduling heuristic for functional parallelism and its performance measure

A multiprocessor scheduling heuristic for functional paralle...

引用

International conference on parallel and distributed Systems (ICPADS)

作者： D. Li A. Mizuno Y. Iwahori N. Ishii Educational Center of Information Processing Nagoya Institute of Technology Nagoya Japan Department of Intelligence and Computer Science Nagoya Institute of Technology Nagoya Japan Nagoya Kogyo Daigaku Nagoya Aichi JP

This paper addresses the following scheduling problem: given a precedence graph with communication costs and a machine architecture with different types of processors, construct a schedule that runs on the given architecture at the minimum possible execution time. The main contributions are: Firstly, we present a static scheduling algorithm that keeps processors idle for future important tasks and fills idle time slots incurred due to interprocessor communication. Secondly, to evaluate the effectiveness of the algorithm, we develop a lower bound on the length of a optimal schedule as a yardstick. Experiments show that this new approach produces better schedules and takes much less compile time.

关键词： Processor scheduling Optimal scheduling parallel processing Scheduling algorithm Costs Information processing Machine intelligence Computer science Computer architecture Concurrent computing

来源：评论

学校读者我要写书评

暂无评论

A Study of Performance on SMP and distributed Memory Architectures Using a Shared Memory Programming Model

A Study of Performance on SMP and Distributed Memory Archite...

引用

Supercomputing conference

作者： E.D. Brooks K.H. Warren Lawrence Livermore National Laboratory University of California Livermore CA USA

We examine the use of a shared memory programming model to address the problem of portability between distributed memory and shared memory architectures. We conduct this evaluation by extending an existing programming model, the parallel C Preprocessor, with a type qualifier interpretation of the data sharing keywords borrowed from the Split-C and AC compilers. We evaluate the performance of the resulting programming model on a wide range of shared memory and distributed memory computing platforms using several numerical algorithms as benchmarks. We find the type-qualifier-based programming model capable of efficient execution on distributed memory and shared memory architectures.

关键词： Memory architecture Message passing Hardware Vehicles Delay Laboratories Fast Fourier transforms Bandwidth parallel programming Programming profession

来源：评论

学校读者我要写书评

暂无评论

Efficient personalized communication on wormhole networks

Efficient personalized communication on wormhole networks

引用

International conference on parallel Architecture and Compilation Techniques (PACT)

作者： F. Petrini R. Vanneschi International Computer Science Institute Berkeley CA USA Dipartimento di Informatica Pisa Italy

Bridging models, such as the BSP (bulk synchronous parallel) model, tend to abstract the characteristics of interconnection networks using a small set of parameters, by dividing the computation into supersteps and organizing the communication into global patterns called h-relations. In this paper, we evaluate (through experimental results conducted on a wormhole-routed 2D torus and a quaternary fat-tree with 256 processing nodes) the execution time of three families of h-relations with variable degree of imbalance. We also prove a strong result that links the communication performance of the fat-tree with the BSP abstraction of the interconnection network. Given a generic h-relation, we can provide a value of g (the gap) that, in the worst case, slightly overestimates the completion time and is very close to optimality.

关键词： Computer networks Delay Multiprocessor interconnection networks Random access memory Concurrent computing Costs Computer science Organizing distributed computing High performance computing

来源：评论

学校读者我要写书评

暂无评论

Dycos-a customizable kernel architecture supporting distributed operating environments

Dycos-a customizable kernel architecture supporting distribu...

引用

IEEE International conference on Algorithms and Architectures for parallel Processing (ICAP)

作者： C.B. Czech Department of Computer Science Technische Universitat Munchen Munich Germany

Techniques of customizing and extending operating systems (OSs) have a growing impact on system architectures in the field of distributed computing and parallel programming. Even if traditional methods of adaption have been limited to the user-level, modern OSs cannot do without kernel support. Hence concepts and structures of microkernel architectures must be redefined to meet the requirements of today's and future applications. We propose a new customizable low-level OS architecture-the Dycos kernel. We discuss customization demands on microkernels and describe the basic kernel concept. Dycos is an object-based approach providing a toolbox of operations to build user-definable compositions of kernel structures. The Dycos approach has been evaluated on a Solaris 2.5.1 platform.

关键词： Kernel Computer architecture Resource management Software architecture Computer science Electronic mail Operating systems distributed computing parallel programming Solar power generation

来源：评论

学校读者我要写书评

暂无评论

Implementing LOGFLOW on a workstation cluster

Proceedings of the Conference on High Performance Computing ...

引用

proceedings of the conference on High Performance computing on the Information Superhighway, HPC Asia'97 1997年 145-150页

作者： Kacsuk, Peter Puskas, Zsolt Nemeth, Zsolt KFKI Research Inst for Measurement and Computing Techniques Budapest Hungary

LOGFLOW is a distributed Prolog implementation running on transputer networks, developed at KFKI-MSZKI. To improve capabilities and the power of LOGFLOW, the system is ported onto workstation-clusters under the name WS-LOGFLOW. The new platform requires modification in the architecture of the system, in token transportation and in the work-distribution. This paper presents the modified architecture of WS-LOGFLOW, the new token distribution scheme and some details of the portation: processes, intra-node shared-memory communication, inter-node PVM-based message-passing communication.

关键词： parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Collaboration and visual steering of simulations

Collaboration and visual steering of simulations

引用

conference on Visual Data Exploration and Analysis IV

作者： Zou, S Ribarsky, W Jean, Y Heiner, J Schwan, K Sumner, B Okuma, O Georgia Institute of Technology (United States) Lucent Technologies Bell Labs. (United States)

ISBN: (纸本)0819424285

This paper presents a structure and set of tools to address the needs of groups of scientists working on large, time-dependent simulations. It describes a direct manipulation, 3D steering environment that is integrated with a controller for instrumenting parallel computations, collecting output, and passing it in binary made between heterogeneous machines, The instrumentation allows collection of data at chosen points in the model and control of the computation through changes of parameters or insertion of alternate data. The steering interface and controller are joined with a library of collaborative communication tools, With these tools a user may steer a simulation and share both the steering input and visual output with one or several colleagues, The tools permit the trading of steering and visualization tools within the group. In addition a lime-dependent steering interface has been introduced. Here time is treated on exactly the same basis as the spatial dimensions so there is a 4D environment, 3 shown spatially and one through animation. The steering interface is built upon a flexible visualization/analysis system. This permits the immediate display of time-dependent results from the dynamic simulations and refined interaction with the results to bring out the character and correlations of multivariate data. The user can then launch new simulations at any stage in this exploration using the visualization to define and focus the simulation parameters, region of interest, and time frame.

关键词： computational steering computer simulation visualization collaboration collaborative work distributed computing 3D graphical interface

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：