检索结果-内蒙古大学图书馆

International Conference on Parallel Processing (ICPP)

作者： Laércio L. Pilla Christiane Pousa Ribeiro Daniel Cordeiro Chao Mei Abhinav Bhatele Philippe O.A. Navaux François Broquedis Jean-François Méhaut Laxmikant V. Kale Institute of Informatics Federal University of Rio Grande do Sul Porto Alegre Brazil LIG Laboratory-CEA / INRIA Grenoble University Grenoble France Department of Computer Science University of Illinois Urbana-Champaign Urbana IL USA Center for Applied Scientific Computing Lawrence Livemore National Laboratory Livermore CA USA

Multi-core compute nodes with non-uniform memory access (NUMA) are now a common architecture in the assembly of large-scale parallel machines. On these machines, in addition to the network communication costs, the memory access costs within a compute node are also asymmetric. Ignoring this can lead to an increase in the data movement costs. Therefore, to fully exploit the potential of these nodes and reduce data access costs, it becomes crucial to have a complete view of the machine topology (i.e. the compute node topology and the interconnection network among the nodes). Furthermore, the parallel application behavior has an important role in determining how to utilize the machine efficiently. In this paper, we propose a hierarchical load balancing approach to improve the performance of applications on parallel multi-core systems. We introduce NucoLB, a topology-aware load balancer that focuses on redistributing work while reducing communication costs among and within compute nodes. NucoLB takes the asymmetric memory access costs present on NUMA multi-core compute nodes, the interconnection network overheads, and the application communication patterns into account in its balancing decisions. We have implemented NucoLB using the Charm++ parallel runtime system and evaluated its performance. Results show that our load balancer improves performance up to 20% when compared to state-of-the-art load balancers on three different NUMA parallel machines.

关键词： Load management Topology Parallel machines Benchmark testing Runtime Network topology Multiprocessor interconnection

来源：评论

学校读者我要写书评

暂无评论

A scalable eigensolver for large scale-free graphs using 2D graph partitioning

A scalable eigensolver for large scale-free graphs using 2D ...

引用

2011 International Conference for High Performance computing, Networking, Storage and Analysis, SC11

作者： Yoo, Andy Baker, Allison H. Pearce, Roger Henson, Van Emden Lawrence Livermore National Laboratory Center for Applied Scientific Computing United States Department of Computer Science and Engineering Texas A and M University United States

ISBN: (纸本)9781450307710

Eigensolvers are important tools for analyzing and mining useful information from scale-free graphs. Such graphs are used in many applications and can be extremely large. Unfortunately, existing parallel eigensolvers do not scale well for these graphs due to the high communication overhead in the parallel matrix-vector multiplication (MatVec). We develop a MatVec algorithm based on 2D edge partitioning that significantly reduces the communication costs and embed it into a popular eigensolver library. We demonstrate that the enhanced eigensolver can attain two orders of magnitude performance improvement compared to the original on a state-of-art massively parallel machine. We illustrate the performance of the embedded MatVec by computing eigenvalues of a scale-free graph with 300 million vertices and 5 billion edges, the largest scale-free graph analyzed by any in-memory parallel eigensolver, to the best of our knowledge. Copyright 2011 ACM.

关键词： Eigenvalues and eigenfunctions

来源：评论

学校读者我要写书评

暂无评论

Evaluation of simple causal message logging for large-scale fault tolerant HPC systems

Evaluation of simple causal message logging for large-scale ...

引用

25th IEEE International Parallel and Distributed Processing Symposium, Workshops and Phd Forum, IPDPSW 2011

作者： Meneses, Esteban Bronevetsky, Greg Kalé, Laxmikant V. Department of Computer Science University of Illinois Urbana IL United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore CA United States

ISBN: (纸本)9780769543857

The era of petascale computing brought machines with hundreds of thousands of processors. The next generation of exascale supercomputers will make available clusters with millions of processors. In those machines, mean time between failures will range from a few minutes to few tens of minutes, making the crash of a processor the common case, instead of a rarity. Parallel applications running on those large machines will need to simultaneously survive crashes and maintain high productivity. To achieve that, fault tolerance techniques will have to go beyond checkpoint/restart, which requires all processors to roll back in case of a failure. Incorporating some form of message logging will provide a framework where only a subset of processors are rolled back after a crash. In this paper, we discuss why a simple causal message logging protocol seems a promising alternative to provide fault tolerance in large supercomputers. As opposed to pessimistic message logging, it has low latency overhead, especially in collective communication operations. Besides, it saves messages when more than one thread is running per processor. Finally, we demonstrate that a simple causal message logging protocol has a faster recovery and a low performance penalty when compared to checkpoint/restart. Running NAS Parallel Benchmarks (CG, MG, BT and DT) on 1024 processors, simple causal message logging has a latency overhead below 5%. © 2011 IEEE.

关键词： Supercomputers

来源：评论

学校读者我要写书评

暂无评论

Weighted locality-sensitive scheduling for mitigating noise on multi-core clusters

Weighted locality-sensitive scheduling for mitigating noise ...

引用

18th International Conference on High Performance computing, HiPC 2011

作者： Kale, Vivek Bhatele, Abhinav Gropp, William D. Department of Computer Science University of Illinois at Urbana-Champaign Urbana IL 61801 United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore CA 94551 United States

ISBN: (纸本)9781457719516

Recent studies have shown that operating system (OS) interference, popularly called OS noise can be a significant problem as we scale to a large number of processors. One solution for mitigating noise is to turn off certain OS services on the machine. However, this is typically infeasible because full-scale OS services may be required for some applications. Furthermore, it is not a choice that an end user can make. Thus, we need an application-level solution. Building upon previous work that demonstrated the utility of within-node light-weight load balancing, we discuss the technique of weighted micro-scheduling and provide insights based on experimentation for two different machines with very different noise signatures. Through careful enumeration of the search space of scheduler parameters, we allow our weighted micro-scheduler to be dynamic, adaptive and tunable for a specific application running on a specific architecture. By doing this, we show how we can enable running scientific applications efficiently on a very large number of processors, even in the presence of noise. © 2011 IEEE.

关键词： Scheduling

来源：评论

学校读者我要写书评

暂无评论

Avoiding hot-spots on two-level direct networks

Avoiding hot-spots on two-level direct networks

引用

2011 International Conference for High Performance computing, Networking, Storage and Analysis, SC11

作者： Bhatele, Abhinav Jain, Nikhil Gropp, William D. Kale, Laxmikant V. Lawrence Livermore National Laboratory Center for Applied Scientific Computing Livermore CA 94551 United States Department of Computer Science University of Illinois at Urbana-Champaign Urbana IL 61801 United States

ISBN: (纸本)9781450307710

A low-diameter, fast interconnection network is going to be a prerequisite for building exascale machines. A two-level direct network has been proposed by several groups as a scalable design for future machines. IBM's PERCS topology and the dragonfly network discussed in the DARPA exascale hardware study are examples of this design. The presence of multiple levels in this design leads to hot-spots on a few links when processes are grouped together at the lowest level to minimize total communication volume. This is especially true for communication graphs with a small number of neighbors per task. Routing and mapping choices can impact the communication performance of parallel applications running on a machine with a two-level direct topology. This paper explores intelligent topology aware mappings of different communication patterns to the physical topology to identify cases that minimize link utilization. We also analyze the trade-offs between using direct and indirect routing with different mappings. We use simulations to study communication and overall performance of applications since there are no installations of two-level direct networks yet. This study raises interesting issues regarding the choice of job scheduling, routing and mapping for future machines. Copyright 2011 ACM.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Improving communication performance in dense linear algebra via topology aware collectives

Improving communication performance in dense linear algebra ...

引用

2011 International Conference for High Performance computing, Networking, Storage and Analysis, SC11

作者： Solomonik, Edgar Bhatele, Abhinav Demmel, James Dept. of Electric Engineering and Computer Science University of California at Berkeley Berkeley CA 94720 United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore CA 94551 United States

ISBN: (纸本)9781450307710

Recent results have shown that topology aware mapping reduces network contention in communication-intensive kernels on massively parallel machines. We demonstrate that on mesh interconnects, topology aware mapping also allows for the utilization of highly-efficient topology aware collectives. We map novel 2.5D dense linear algebra algorithms to exploit rectangular collectives on cuboid partitions allocated by a Blue Gene/P supercomputer. Our mappings allow the algorithms to exploit optimized line multicasts and reductions. Commonly used 2D algorithms cannot be mapped in this fashion. On 16,384 nodes (65,536 cores) of Blue Gene/P, 2.5D algorithms that exploit rectangular collectives are sig- nificantly faster than 2D matrix multiplication (MM) and LU factorization, up to 8.7x and 2.1x, respectively. These speed-ups are due to communication reduction (up to 95.6% for 2.5D MM with respect to 2D MM). We also derive LogP- based novel performance models for rectangular broadcasts and reductions. Using those, we model the performance of matrix multiplication and LU factorization on a hypothetical exascale architecture. Copyright 2011 ACM.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Support for the evolution of C++ generic functions

Support for the evolution of C++ generic functions

引用

Lecture Notes in computer science

作者： Pirkelbauer, Peter Dechev, Damian Stroustrup, Bjarne Lawrence Livermore National Laboratory Center for Applied Scientific Computing Livermore CA 94550-9698 United States University of Central Florida Department of Electrical Engineering and Computer Science Orlando FL 32816-2362 United States Texas AandM University Department of Computer Science and Engineering College Station TX 77843-3112 United States

ISBN: (纸本)9783642194399

The choice of requirements for an argument of a generic type or algorithm is a central design issue in generic programming. In the context of C++, a specification of requirements for a template argument or a set of template arguments is called a concept. In this paper, we present a novel tool, TACE (template analysis and concept extraction), designed to help programmers understand the requirements that their code de facto imposes on arguments and help simplify and generalize those through comparisons with libraries of well-defined and precisely-specified concepts. TACE automatically extracts requirements from the body of function templates. These requirements are expressed using the notation and semantics developed by the ISO C++ standards committee. TACE converts implied requirements into concept definitions and compares them against concepts from a repository. Components of a well-defined library exhibit commonalities that allow us to detect problems by comparing requirements from many components: Design and implementation problems manifest themselves as minor variations in requirements. TACE points to source code that cannot be constrained by concepts and to code where small modifications would allow the use of less constraining concepts. For people who use a version of C++ with concept support, TACE can serve as a core engine for automated source code rejuvenation. © 2011 Springer-Verlag Berlin Heidelberg.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Erratum: Search for new phenomena in final states with large jet multiplicities and missing transverse momentum at = 8 TeV proton-proton collisions using the ATLAS experiment

引用

Journal of High Energy Physics 2014年第1期2014卷 1-18页

作者： G. Aad T. Abajyan B. Abbott J. Abdallah S. Abdel Khalek O. Abdinov R. Aben B. Abi M. Abolins O. S. AbouZeid H. Abramowicz H. Abreu Y. Abulaiti B. S. Acharya L. Adamczyk D. L. Adams T. N. Addy J. Adelman S. Adomeit T. Adye S. Aefsky T. Agatonovic-Jovin J. A. Aguilar-Saavedra M. Agustoni S. P. Ahlen A. Ahmad M. Ahsan G. Aielli T. P. A. Åkesson G. Akimoto A. V. Akimov M. A. Alam J. Albert S. Albrand M. J. Alconada Verzini M. Aleksa I. N. Aleksandrov F. Alessandria C. Alexa G. Alexander G. Alexandre T. Alexopoulos M. Alhroob M. Aliev G. Alimonti L. Alio J. Alison B. M. M. Allbrooke L. J. Allison P. P. Allport S. E. Allwood-Spiers J. Almond A. Aloisio R. Alon A. Alonso F. Alonso A. Altheimer B. Alvarez Gonzalez M. G. Alviggi K. Amako Y. Amaral Coutinho C. Amelung V. V. Ammosov S. P. Amor Dos Santos A. Amorim S. Amoroso N. Amram C. Anastopoulos L. S. Ancu N. Andari T. Andeen C. F. Anders G. Anders K. J. Anderson A. Andreazza V. Andrei X. S. Anduaga S. Angelidakis P. Anger A. Angerami F. Anghinolfi A. V. Anisenkov N. Anjos A. Annovi A. Antonaki M. Antonelli A. Antonov J. Antos F. Anulli M. Aoki L. Aperio Bella R. Apolle G. Arabidze I. Aracena Y. Arai A. T. H. Arce S. Arfaoui J.-F. Arguin S. Argyropoulos E. Arik M. Arik A. J. Armbruster O. Arnaez V. Arnal O. Arslan A. Artamonov G. Artoni D. Arutinov S. Asai N. Asbah S. Ask B. Åsman L. Asquith K. Assamagan R. Astalos A. Astbury M. Atkinson N. B. Atlay B. Auerbach E. Auge K. Augsten M. Aurousseau G. Avolio D. Axen G. Azuelos Y. Azuma M. A. Baak C. Bacci A. M. Bach H. Bachacou K. Bachas M. Backes M. Backhaus J. Backus Mayes E. Badescu P. Bagiacchi P. Bagnaia Y. Bai D. C. Bailey T. Bain J. T. Baines O. K. Baker S. Baker P. Balek F. Balli E. Banas Sw. Banerjee D. Banfi A. Bangert V. Bansal H. S. Bansil L. Barak S. P. Baranov T. Barber E. L. Barberio D. Barberis M. Barbero D. Y. Bardin T. Barillari M. Barisonzi T. Barklow N. Barlow B. M. Barnett R. M. Barnett A. Baroncelli G. Barone A. J. Barr F. Barreiro J. Barreiro Guimarães da Costa R. Bartoldus A. E. Barton V. Bartsch A. B Fakultät für Mathematik und Physik Albert-Ludwigs-Universität Freiburg Germany Physikalisches Institut University of Bonn Bonn Germany Homer L. Dodge Department of Physics and Astronomy University of Oklahoma Norman United States of America Institut de Física d’Altes Energies and Departament de Física de la Universitat Autònoma de Barcelona Barcelona Spain LAL Université Paris-Sud and CNRS/IN2P3 Orsay France Institute of Physics Azerbaijan Academy of Sciences Baku Azerbaijan Nikhef National Institute for Subatomic Physics and University of Amsterdam Amsterdam Netherlands Department of Physics Oklahoma State University Stillwater United States of America Department of Physics and Astronomy Michigan State University East Lansing United States of America Department of Physics University of Toronto Toronto Canada Raymond and Beverly Sackler School of Physics and Astronomy Tel Aviv University Tel Aviv Israel DSM/IRFU (Institut de Recherches sur les Lois Fondamentales de l’Univers) CEA Saclay (Commissariat à l’Energie Atomique et aux Energies Alternatives) Gif-sur-Yvette France Department of Physics Stockholm University Stockholm Sweden INFN Gruppo Collegato di Udine Udine Italy Department of Physics King’s College London London United Kingdom AGH University of Science and Technology Faculty of Physics and Applied Computer Science Krakow Poland Physics Department Brookhaven National Laboratory Upton United States of America Department of Physics Hampton University Hampton United States of America Department of Physics Yale University New Haven United States of America Fakultät für Physik Ludwig-Maximilians-Universität München München Germany Particle Physics Department Rutherford Appleton Laboratory Didcot United Kingdom Department of Physics Brandeis University Waltham United States of America Vinca Institute of Nuclear Sciences University of Belgrade Belgrade Serbia Laboratorio de Instrumentacao e Fisica Experimental de Particulas - LIP Lisboa Portug

来源：评论

学校读者我要写书评

暂无评论

A scalable eigensolver for large scale-free graphs using 2D graph partitioning

A scalable eigensolver for large scale-free graphs using 2D ...

引用

Supercomputing Conference

作者： Andy Yoo Allison H. Baker Roger Pearce Van Emden Henson Center for Applied Scientific Computing Lawrence Livemore National Laboratory USA Department of Computer Science and Engineering Texas A and M University USA

关键词： Program processors Sparse matrices Partitioning algorithms Vectors Libraries Scalability Generators

来源：评论

学校读者我要写书评

暂无评论

Modeling the performance of an algebraic multigrid cycle on HPC platforms 11

Modeling the performance of an algebraic multigrid cycle on ...

引用

Proceedings of the international conference on Supercomputing

作者： Gahvari, Hormozd Baker, Allison H. Schulz, Martin Yang, Ulrike Meier Jordan, Kirk E. Gropp, William Computer Science Department University of Illinois at Urbana-Champaign Urbana IL 61801 United States Center for Applied Scientific Computing Lawrence Livermore National Laboratory Livermore CA 94551 United States IBM TJ Watson Research Center Cambridge MA 02142 United States

ISBN: (纸本)9781450301022

Now that the performance of individual cores has plateaued, future supercomputers will depend upon increasing parallelism for performance. Processor counts are now in the hundreds of thousands for the largest machines and will soon be in the millions. There is an urgent need to model application performance at these scales and to understand what changes need to be made to ensure continued scalability. This paper considers algebraic multigrid (AMG), a popular and highly efficient iterative solver for large sparse linear systems that is used in many applications. We discuss the challenges for AMG on current parallel computers and future exascale architectures, and we present a performance model for an AMG solve cycle as well as performance measurements on several massively-parallel platforms. © 2011 ACM.

关键词： Parallel architectures

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：