检索结果-内蒙古大学图书馆

IEEE international Conference on Cluster Computing

作者： P. Berenbrink T. Friedetzky L.A. Goldberg Department of Computer Science University of Warwick Coventry UK

In this paper we analyse a very simple dynamic work-stealing algorithm. In the work-generation model, there are n generators which are arbitrarily distributed among a set of n processors. During each time-step, with probability /spl lambda/, each generator generates a unit-time task which it inserts into the queue of its host processor. After the new tasks are generated, each processor removes one task from its queue and services it. Clearly, the work-generation model allows the load to grow more and more imbalanced, so, even when /spl lambda/<1, the system load can be unbounded. The natural work-stealing algorithm that we analyse works as follows. During each time step, each empty processor sends a request to a randomly selected other processor. Any non-empty processor having received at least one such request in turn decides (again randomly) in favour of one of the requests. The number of tasks which are transferred from the non-empty processor to the empty one is determined by the so-called work-stealing function f. We analyse the long-term behaviour of the system as a function of /spl lambda/ and f. We show that the system is stable for any constant generation rate /spl lambda/<1 and for a wide class of functions f. We give a quantitative description of the functions f which lead to stable systems. Furthermore, we give upper bounds on the average system load (as a function of f and n).

关键词： Algorithm design and analysis Load management Computer science Heuristic algorithms Load modeling Upper bound distributed computing Contracts parallel programming Kernel

来源：评论

学校读者我要写书评

暂无评论

Performance evaluation of VBR traffic in InfiniBand

Performance evaluation of VBR traffic in InfiniBand

引用

Canadian Conference on Electrical and Computer Engineering (CCECE)

作者： F.J. Alfaro J.L. Sanchez L. Orozco J. Duato Department de Informática Escuela Politécnica Superior Uni Albacete Spain School of Information Technology and Engineering University of Ottawa Ottawa ONT Canada Department de Information de Sistemas y Computadores University Politéchnica de Valencia Valencia Spain

The InfiniBand architecture (IBA) is becoming an industry standard both for communication between processing nodes and I/O devices and for interprocessor communication. It replaces the traditional I/O bus with a switch-based interconnect for connecting processing nodes and I/O devices. It is being developed by the InfiniBand/sup SM/ Trade Association (IBTA) to provide the levels of reliability, availability, performance, scalability, and quality of service (QoS) necessary for present and future server systems. For this, IBA provides a series of mechanisms that are able to guarantee QoS to the applications. Alfaro, Sanchez and Das (see proceedings of international parallel and distributed Processing symposium, April 2002) proposed a strategy to compute the InfiniBand arbitration tables. We only evaluated our proposal for CBR traffic with fixed mean bandwidth requirements. We evaluate our strategy to compute the InfiniBand arbitration tables with VBR traffic. Performance results show that, this class of traffic also gets their QoS requirements.

关键词： Quality of service Communication industry Communication standards Communication switching Joining processes Samarium Availability Scalability distributed processing Concurrent computing

来源：评论

学校读者我要写书评

暂无评论

An Architecture Interface and Offload Model for Low-Overhead, Near-Data, distributed Accelerators 22

An Architecture Interface and Offload Model for Low-Overhead...

引用

proceedings of the 55th Annual IEEE/ACM international symposium on Microarchitecture

作者： Saambhavi Baskaran Mahmut Taylan Kandemir Jack Sampson Pennsylvania State University

ISBN: (纸本)9781665462723

The performance and energy costs of coordinating and performing data movement have led to proposals adding compute units and/or specialized access units to the memory hierarchy. However, current on-chip offload models are restricted to fixed compute and access pattern types, which limits software-driven optimizations and the applicability of such an offload interface to heterogeneous accelerator resources. This paper presents a computation offload interface for multi-core systems augmented with distributed on-chip accelerators. With energy-efficiency as the primary goal, we define mechanisms to identify offload partitioning, create a low-overhead execution model to sequence these fine-grained operations, and evaluate a set of workloads to identify the complexity needed to achieve distributed near-data *** demonstrate that our model and interface, combining features of dataflow in parallel with near-data processing engines, can be profitably applied to memory hierarchies augmented with either specialized compute substrates or lightweight near-memory cores. We differentiate the benefits stemming from each of elevating data access semantics, near-data computation, inter-accelerator coordination, and compute/access logic specialization. Experimental results indicate a geometric mean (energy efficiency improvement; speedup; data movement reduction) of (3.3; 1.59; 2.4)×, (2.46; 1.43; 3.5)× and (1.46; 1.65; 1.48)× compared to an out-of-order processor, monolithic accelerator with centralized accesses and monolithic accelerator with decentralized accesses, respectively. Evaluating both lightweight core and CGRA fabric implementations highlights model flexibility and quantifies the benefits of compute specialization for energy efficiency and speedup at 1.23× and 1.43×, respectively.

关键词： distributed accelerator near-data offload energy efficiency heterogeneous architecture interface

来源：评论

学校读者我要写书评

暂无评论

parallel simulation of stochastic reward nets using theatre 21

Parallel simulation of stochastic reward nets using theatre

引用

proceedings of the 2021 IEEE/ACM 25th international symposium on distributed Simulation and Real Time Applications

作者： Franco Cicirelli Libero Nigro CNR - National Research Council of Italy Italy University of Calabria Italy

ISBN: (纸本)9781665433266

This work aims at the development of tools for supporting modelling and analysis of timed systems by Stochastic Reward Nets (SRN). In a first approach it was proposed and experimented a formal reduction of SRN over Timed Automata (TA) in the context of the Uppaal popular toolbox. The reduction has the merit to allow both exhaustive model checking of an SRN model, useful for the assessment of qualitative properties (e.g., absence of deadlocks, occurrence of particular event sequences etc.), and quantitative analysis through the statistical model checker, which is based on simulations. However, although Uppaal enabled formal reasoning on the semantics of SRN, its practical usage suffers of scalability problems, that is it can introduce severe limitations in time and space when studying complex models. To cope with this problem, this paper describes a Java implementation of the SRN operational core engine, using the lock-free and efficient Theatre actor system which permits the parallel simulation of large models. The realization can be used for functional property checking on an untimed version of a source SRN model, and quantitative estimation of measurables through simulations. The paper discusses the design and implementation of the core engine of SRN on top of Theatre, together with supported intuitive configuration process of an SRN model, and reports some experimental results using a scalable grid computing model. The experiments confirm Theatre/SRN are capable of exploiting the potential of modern multi-core machines and can deliver good execution performances on large models.

关键词： theatre Java stochastic reward nets actors high-performance computing performability analysis

来源：评论

学校读者我要写书评

暂无评论

A framework for auto-tuning HDF5 applications 13

A framework for auto-tuning HDF5 applications

引用

proceedings of the 22nd international symposium on High-performance parallel and distributed computing

作者： Babak Behzad Joseph Huchette Huong Vu Thanh Luu Ruth Aydt Surendra Byna Yushu Yao Quincey Koziol Prabhat University of Illinois at Urbana-Champaign Urbana IL USA Rice University Houston TX USA The HDF Group Champaign IL USA Lawrence Berkeley National Laboratory Berkeley CA USA

ISBN: (纸本)9781450319102

The modern parallel I/O stack consists of several software layers with complex inter-dependencies and performance characteristics. While each layer exposes tunable parameters, it is often unclear to users how different parameter settings interact with each other and affect overall I/O performance. As a result, users often resort to default system settings, which typically obtain poor I/O bandwidth. In this research, we develop a benchmark guided auto-tuning framework for tuning the HDF5, MPI-IO, and Lustre layers on production supercomputing facilities. Our framework consists of three main components. H5Tuner uses a control file to adjust I/O parameters without modifying or recompiling the application. H5PerfCapture records performance metrics for HDF5 and MPI-IO. H5Evolve uses a genetic algorithm to explore the parameter space to determine well-performing configurations. We demonstrate I/O performance results for three HDF5 application-based benchmarks on a Sun HPC system. All the benchmarks running on 512 MPI processes perform 3X to 5.5X faster with the auto-tuned I/O parameters compared to a configuration with default system parameters.

关键词： auto-tuning H5perfcapture parallel I/O H5tuner H5evolve HDF5 auto-tuning parallel file systems

来源：评论

学校读者我要写书评

暂无评论

7th international Workshop on Performance Modeling, Evaluation, and Optimization of Ubiquitous Computing and Networked systems (PMEO-UCNS'08)

IPDPS Miami 2008 - Proceedings of the 22nd IEEE Internationa...

引用

IPDPS Miami 2008 - proceedings of the 22nd IEEE international parallel and distributed Processing symposium, Program and CD-ROM 2008年

作者： Min, Geyong Ould-Khaoua, Mohamed Al-Dubai, Ahmed Y. Jin, Xiaolong Notare, Mirela Sechi Moretti Annoni Al-Begain, K. Al-Hashimi, B. Romdhani, I. Arabnia, H.R. Arvind, D.K. Awan, I. Boukerche, A. Bradley, J. Cockshott, P. Colajanni, M. Day, K. Djemame, K. El-Ghazawi, T. Fatoohi, R. Gueroui, M. Helal, S. Hassanein, H. Jarvis, S. Karatza, H. Katangur, A. Khonsari, A. Knottenbelt, W. Li, K. Liu, H. Loucif, S. Mackenzie, L.M. Pan, Y. Pradhan, D.K. Sarbazi-Azad, H. Shahrabi, A. Song, E. Thomas, N. Touzene, A. Woodward, M. Wu, J. Xiao, L. Xie, T. Yang, L.T. Vanderbauwhede, W. Buchanan, W. Zhou, X. Zomaya, A. Department of Computing University of Bradford Bradford BD7 1DP United Kingdom Department of Computing Science University of Glasgow Glasgow G12 8RZ United Kingdom School of Computing Napier University Edinburgh EH10 5DT United Kingdom Univ. of Glamorgan United Kingdom Univ. of Southampton United Kingdom Napier University United Kingdom Univ. of Georgia United States Edinburgh University United Kingdom Univ. of Bradford United Kingdom Univ. of North Texas United States Imperial College London United Kingdom Univ. of Glasgow United Kingdom Univ. of Modena Italy Sultan Qaboos Univ. Oman Univ. of Leeds United Kingdom George Washington University United States San Jose State University United States University of Cergy-Pontoise France University of Florida United States Queen's University Canada Univ. of Warwick United Kingdom Univ. of Thessaloniki Greece Texas A and M Univ. United States IPM Iran State Univ. of New York New Paltz NY United States Huazhong Univ. of Science and Technology China Moncton University Canada Georgia State Univ. United States Univ. of Bristol United Kingdom Sharif Univ. and IPM Iran Glasgow Caledonian Univ. United Kingdom Univ. of Newcastle United Kingdom Florida Atlantic Univ. United States Michigan State Univ. United States San Diego State University United States St Francis Xavier Univ. Canada University of Glasgow United Kingdom University of Colorado Colorado Springs CO United States Univ. of Sydney Australia Barddal University Florianopolis SC Brazil

No abstract available

ISBN: (纸本)9781424416943

No abstract available

关键词：

来源：评论

学校读者我要写书评

暂无评论

6th international workshop on performance modeling, evaluation, and optimization of parallel and distributed systems (PMEO-PDS'07)

Proceedings - 21st International Parallel and Distributed Pr...

引用

proceedings - 21st international parallel and distributed Processing symposium, IPDPS 2007;Abstracts and CD-ROM 2007年

作者： Min, Geyong Ould-Khaoua, Mohamed Jin, Xiaolong Notare, Mirela Sechi Moretti Annoni Al-Begain, K. Al-Dubai, A. Arabnia, H.R. Awan, I. Boukerche, A. Bradley, J. Cockshott, P. Colajanni, M. Day, K. Djemame, K. El-Ghazawi, T. Fatoohi, R. Gelenbe, E. Gueroui, M. He, X. Ibbett, R. Jarvis, S. Jin, X. Karatza, H. Katangur, A. Khonsari, A. Knottenbelt, W. Li, K. Liu, H. Loucif, S. Mackenzie, L.M. Pan, Y. Pradhan, D.K. Qin, X. Sarbazi-Azad, H. Shahrabi, A. Song, E. Sun, X.H. Thomas, N. Touzene, A. Wang, X. Woodward, M. Wu, J. Xiao, L. Xie, T. Xu, C.Z. Xu, Z. Yan, S. Yang, L.T. Zhou, X. Zomaya, A. Department of Computing University of Bradford Bradford BD7 1DP United Kingdom Department of Computing Science University of Glasgow Glasgow G12 8RZ United Kingdom Barddal University Florianópolis SC Brazil Univ. of Glamorgan United Kingdom Napier Univ. United Kingdom Univ. of Georgia United States Univ. of Bredford United Kingdom Univ. of North Texas United States Imperial College London United Kingdom Univ. of Glasegow United Kingdom Univ. of Modena Italy Sultan Qaboas Univ. Oman Univ. of Leeds United Kingdom George Washington University United States San Jose State University United States University of Cergy-Pontoise France Tennessee Technological Univ. United States Univ. of Edinburgh United Kingdom Univ. of Warwick United Kingdom Univ. of Thessaloniki Greece Texas AandM Univ. United States IPM Iran Imperiel College London United Kingdom State Univ. of New York New Paltz NY United States Huazhoang Univ. of Science and Technology China Emirates University United Arab Emirates Univ. of Glasgow United Kingdom Georgia State Univ. United States Univ. of Bristol United Kingdom New Mexico Inst. of Maining and Technology United States Sharif Univ. and IPM Iran Glasgow Caledonian Univ. United Kingdom Huazhong Univ. of Science and Technology China Illinois Institute of Technology United States Univ. of Newcastle United Kingdom Sultan Qaboes Univ. Oman Villanova Univ. United States Univ. of Bradford United Kingdom Florida Atlantic Univ. United States Michigan State Univ. United States San Diego State University United States Wayne State Univ. United States Suffolk Univ. United States St Francis Xavier Univ. Canada University of Colorado Colorado Springs CO United States Univ. of Sydney Australia

No abstract available

ISBN: (纸本)1424409101

No abstract available

关键词：

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：