检索结果-内蒙古大学图书馆

international symposium on High Performance distributed computing

作者： J. Leichtl P.E. Crandall M.J. Clement Computer Science & Engineering University of Connecticut Storrs CT USA University of Connecticut Storrs CT US Department of Computer Science Brigham Young University Provo UT USA

ISBN: (纸本)9780818681172

An important development in cluster computing is the availability of multiprocessor workstations. These are able to provide additional computational power to the cluster without increasing network overhead and allow multiparadigm parallelism, which we define to be the simultaneous application of both distributed and shared memory parallel processing techniques to a single problem. In this paper we compare execution times and speedup of parallel programs written in a pure message-passing paradigm with those that combine message passing and shared-memory primitives in the same application. We consider three basic applications that are common building blocks for many scientific and engineering problems: numerical integration, matrix multiplication and Jacobi iteration. Our results indicate that the added complexity of combining shared- and distributed-memory programming methods in the same program does not contribute sufficiently to performance to justify the added programming complexity.

关键词： Parallel programming Workstations Jacobian matrices Parallel processing Libraries Computer science Power engineering computing Computer networks Concurrent computing distributed computing

来源：评论

学校读者我要写书评

暂无评论

Predicting slowdown for networked workstations 97

Predicting slowdown for networked workstations

引用

international symposium on High Performance distributed computing

作者： S.M. Figueira F. Berman Department of Computer Science and Engineering University of California San Diego La Jolla CA USA

ISBN: (纸本)9780818681172

Most applications share the resources of networked workstations with other applications. Since system load can vary dramatically, allocation strategies that assume that resources have a constant availability and/or capability are unlikely to promote performance-efficient allocations in practice. In order to best allocate application tasks to machines, it is critical to provide a realistic model of the effects of contention on application performance. In this paper, we present a model that provides an estimate of the slowdown imposed by competing load on applications targeted to high-performance clusters and networks of workstations. The model provides a basis for predicting realistic communication and computation costs and is shown to achieve good accuracy for a set of scientific benchmarks commonly found in high-performance applications.

关键词： Workstations Predictive models Computational efficiency Costs Application software Resource management Availability Computer science Scholarships Contracts

来源：评论

学校读者我要写书评

暂无评论

A new data abstraction layer required for OODBMS

A new data abstraction layer required for OODBMS

引用

international symposium on Database engineering and applications (IDEAS)

作者： Eun-Sun Cho Sang-Yong Han Hyoung-Joo Kim Department of Computer Science Seoul National University Seoul South Korea

ISBN: (纸本)0818681144

A 'class' in the object-oriented paradigm represents both interface and implementation of the class. However, interface and implementation of a class are needed for different purposes, since class interface is shared among most users, while class implementation is used only by the implementers of the class. The authors introduce a new level of data abstraction, called the 'class-implementation level', which is based on separate management of interface and implementation of a class. They also describe a new model for OODBMS which provides users with the abstract view of the class implementation.

关键词： Object oriented modeling Computer science Object oriented databases Computer interfaces Data structures Relational databases Data models Computer architecture distributed computing Libraries

来源：评论

学校读者我要写书评

暂无评论

Fail-awareness: an approach to construct fail-safe applications

Fail-awareness: an approach to construct fail-safe applicati...

引用

international symposium on Fault-Tolerant computing (FTCS)

作者： C. Fetter F. Cristian Dept. of Comput. Sci. & Eng. California Univ. San Diego La Jolla CA USA Department of Computer Science & Engineering University of California San Diego CA USA

We present a framework for building fail-safe hard real-time applications on top of an asynchronous distributed system subject to communication partitions, i.e. using processors and communication facilities whose real-time delays cannot be guaranteed. The basic assumption behind our approach is that each processor has a local hardware clock that proceeds within a linear envelope of real-time. This allows to compute an upper bound on the actual delays incurred by a particular processing sequence or message transmission. Services and applications can use these computed bounds to detect when they cannot guarantee all their properties because of excessive delays. This allows an application to detect when to switch to a fail-safe mode.

关键词： Real time systems Hardware Switches Delay Application software Upper bound Force measurement Computer science Buildings Clocks

来源：评论

学校读者我要写书评

暂无评论

Checkpointing Message-Passing Interface (MPI) parallel programs

Checkpointing Message-Passing Interface (MPI) parallel progr...

引用

Pacific Rim international symposium on Fault Tolerant Systems

作者： Wei-Jih Li Jyh-Jong Tsay Department of Computer Science and Information Engineering National Chung Cheng University Chiayi Taiwan

ISBN: (纸本)0818682124

Many scientific problems can be distributed on a large number of processes to take advantage of low cost workstations. In a parallel systems, a failure on any processor can halt the computation and requires restarting all applications. Checkpointing is a simple technique to recover the failed execution. Message Passing Interface (MPI) is a standard proposed for writing portable message-passing parallel programs. In this paper, we present a checkpointing implementation for MPI programs, which is transparent, and requires no changes to the application programs. Our implementation combines coordinated, uncoordinated and message logging techniques.

关键词： Checkpointing Message passing Degradation Costs Workstations Concurrent computing Writing Computer science Availability Communications technology

来源：评论

学校读者我要写书评

暂无评论

A formal specification strategy for electronic commerce

A formal specification strategy for electronic commerce

引用

international symposium on Database engineering and applications (IDEAS)

作者： S.A. Ehikioya K. Barker Advanced Database Systems Laboratory Department of Computer Science University of Manitoba Winnipeg MAN Canada

Most current Web applications perform operations that are basically read-only (retrieval of data/documents), so conflicts between transactions do not arise. However as Web applications increasingly become more sophisticated, permitting both concurrent data retrieval and update operations to remote databases, the potentials for conflicts between transactions becomes very high. Therefore, we must ensure correct transaction operation and the correct synchronization of concurrent transactions to avoid deadlock and ensure correctness. The paper presents a formal specification methodology based on the concept of concurrent execution that provides automatic communication and synchronization mechanisms between concurrent processes. The paper explores the issues of correctness, reliability, and recovery of data distributed across the enterprise, describes techniques for guaranteeing and enhancing correctness and procedures for recovering transactions and data in case of any transaction failures to guarantee a high degree of system dependability. Formal methods simplify the writing of concurrent programs by taking most of the burden of synchronization away from the programmer. Further specifying transaction systems formally, enhances application quality and reliability by verifying the functionalities of applications and the correctness of data.

关键词： Formal specifications Electronic commerce Internet Transaction databases Information retrieval World Wide Web business Web server Database systems Laboratories

来源：评论

学校读者我要写书评

暂无评论

A parallel pipelined renderer for time-varying volume data

A parallel pipelined renderer for time-varying volume data

引用

international symposium on Parallel Architectures, Algorithms and Networks (ISPAN)

作者： Tzi-Cier Chiueh Kwan-Liu Ma Computer Science Department State University of New York Stony Brook Stony Brook NY USA NASA Langley Research Center Institute for Computer Applications in Science and Engineering Hampton VA USA

ISBN: (纸本)0818682596

This paper presents a strategy for efficiently rendering time-varying volume data on a distributed-memory parallel computer. Visualizing time-varying volume data take both large storage space and long computation time. Instead of employing all processors to render one volume at a time, a pipelined rendering approach partitions processors into groups so that multiple volumes can be rendered concurrently. The overall rendering time is greatly minimized because rendering is overlapped with I/O required to load the volume data sets. Moreover, parallelization overhead may be reduced as a result of partitioning the processors. We modify an existing parallel volume renderer to exploit various levels of rendering parallelism and to study how the partitioning of processors may lead to optimal rendering performance. We find that two factors affecting the overall execution time are resource utilization efficiency and pipeline startup latency. The optimal partitioning configuration is the one that balances these two factors. Tests on Intel Paragon computers show that in general optimal partitionings do exist for a given rendering task and result in 40-50% saving in overall rendering time.

关键词： Parallel processing Animation NASA Delay Computer applications Data engineering Concurrent computing Data visualization Rendering (computer graphics) Computer science

来源：评论

学校读者我要写书评

暂无评论

Matching user requirements in parallel programming

引用

FUTURE GENERATION COMPUTER SYSTEMS-THE international JOURNAL OF GRID computing AND Escience 1996年第4期12卷 291-305页

作者： Decker, KM Dvorak, JJ Rehmann, RM Ruhl, R Swiss Center for Scientific Computing Via Cantonale CH-6928 Manno Switzerland

While productive use of parallel systems for business applications is quickly growing, their usage for scientific applications is far below expectations. The main reason is that programming of parallel systems is still far too complex. Starting from an analysis of the user requirements, this paper explains our strategy towards next-generation parallel programming environments. We first discuss the problem areas of parallel programming from a user's point of view and give an overview on existing low-, high- and intermediate-level tools. We summarize our objectives and the status of our research. Then we discuss three enabling technologies to overcome current-days limitations of tool environments for parallel computing: formal specification languages, artificial intelligence, and compiler technology. The impact of these technologies on the components proposed for a next-generation programming environment is discussed in detail.

关键词： parallel computing development environment compiler technology specification language artificial intelligence user interface distributed memory high performance

来源：评论

学校读者我要写书评

暂无评论

Implementation and performance of a parallel file system for high performance distributed applications

Implementation and performance of a parallel file system for...

引用

5th IEEE international symposium on High Performance distributed computing

作者： Ligon, WB Ross, RB CLEMSON UNIV CLEMSONSC 29631

ISBN: (纸本)0818675829

Dedicated Cluster Parallel Computers (DCPCs) are emerging as low-cost high performance environments for many important applications in science and engineering. A significant class of applications that perform well on a DCPC are coarse-grain applications that involve large amounts of file I/O. Current research in parallel file systems for distributed systems is providing a mechanism for adapting these applications to the DCPC environment. We present the Parallel Virtual File System (PVFS), a system that provides disk striping across multiple nodes in a distributed parallel computer and file partitioning among tasks in a parallel program. PVFS is unique among similar systems in that it uses a streams-based approach that represents each file access with a single set of request parameters and decouples the number of network messages from details of the files striping and partitioning. PVFS also provides support for efficient collective file accesses and allows overlapping file partitions. We present results of early performance experiments that show PVFS achieves excellent speedups in accessing moderately sized file segments.

关键词： Parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Implementation and performance of a parallel file system for high performance distributed applications

Implementation and performance of a parallel file system for...

引用

international symposium on High Performance distributed computing

作者： W.B. Ligon R.B. Ross Clemson University USA

Dedicated cluster parallel computers (DCPCs) are emerging as low-cost high performance environments for many important applications in science and engineering. A significant class of applications that perform well on a DCPC are coarse-grain applications that involve large amounts of file I/O. Current research in parallel file systems for distributed systems is providing a mechanism for adapting these applications to the DCPC environment. We present the Parallel Virtual File System (PVFS), a system that provides disk striping across multiple nodes in a distributed parallel computer and file partitioning among tasks in a parallel program. PVFS is unique among similar systems in that it uses a stream-based approach that represents each file access with a single set of request parameters and decouples the number of network messages from details of the file striping and partitioning. PVFS also provides support for efficient collective file accesses and allows overlapping file partitions. We present results of early performance experiments that show PVFS achieves excellent speedups in accessing moderately sized file segments.

关键词： File systems Application software Concurrent computing distributed computing Workstations Parallel processing Computer networks Bandwidth High performance computing NASA

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：