检索结果-内蒙古大学图书馆

Introducing multi-level parallelism, at coarse, fine and instruction level to enhance the performance of iterative solvers for large sparse linear systems on multi- and Many-core architecture 6

Introducing multi-level parallelism, at coarse, fine and ins...

引用

IEEE/ACM 6th Workshop on the LLVM Compiler Infrastructure in HPC / Workshop on Hierarchical parallelism for Exascale Computing (LLVM-HPC and HiPar)

作者： Gratien, Jean-Marc IFP Energies Nouvelles Comp Sci Dept Rueil Malmaison France

ISBN: (纸本)9780738110424

With the evolution of High Performance Computing, multi-core and many-core systems are now a common feature of new hardware architectures. The introduction of very large number of cores at the processor level is challenging because it requires to handle multi level parallelism at various levels either coarse or fine to fully take advantage of the offered computing power. The induced programming effort can be fixed with parallel programming models based on the data flow model and the task programming paradigm [1]. To do so many of the standard numerical algorithms must be revisited as they cannot be easily parallelized at the finest levels. Iterative linear solvers are a key part of petroleum reservoir simulation as they can represent up to 80% of the total computing time. In these algorithms, the standard preconditioning methods for large, sparse and unstructured matrices - such as Incomplete LU Factorization (ILU) or Algebraic multigrid (AMG) - fail to scale on shared-memory architectures with large number of cores. In this paper we reconsider preconditioning algorithms to better introduce multi-level parallelism at both coarse level with MPI, fine level with threads and at the instruction level to enable SIMD optimizations. This paper illustrates how we enhance the implementation of preconditioners like the multi-level domain decomposition (DDML) preconditioners [2], based on the popular Additive Schwartz Method (ASM), or the classical ILU0 preconditioner with the fine grained parallel fixed point variant presented in [3]. Our approach is validated on linear systems extracted from realistic petroleum reservoir simulations. The robustness of the preconditioners is tested with respect to the data heterogeneities of the study cases. We evaluate the extensibility of our implementation regarding the model sizes and its scalability regarding the large number of cores provided by new KNL processors or multi-nodes clusters.

关键词： HPC Task Programming Many-Cores multi-level parallelism Parallel Algorithms

来源：评论

学校读者我要写书评

暂无评论

Exploiting multi-level parallel metaheuristics and heterogeneous computing to boost phylogenetics

引用

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 2022年 127卷 208-224页

作者： Santander-Jimenez, Sergio Vega-Rodriguez, Miguel A. Sousa, Leonel Univ Extremadura Escuela Politecn Dept Comp & Commun Technol Campus Univ S-N Caceres 10003 Spain Inst Engn Sistemas & Comp Invest & Desenvolvimen Rua Alves Redol 9 P-1000029 Lisbon Portugal

Optimization problems are becoming increasingly difficult challenges as a result of the definition of more realistic formulations and the availability of larger input data. Fortunately, the computing capabilities of state-of-the-art heterogeneous systems represent an opportunity to deal with the main complexity factors of these problems. These platforms open the door to the definition of robust metaheuristic solvers, in which parallel computations of different nature can be efficiently mapped to the most suitable architectures and hardware resources. This work investigates the combination of multi-level parallelism and heterogeneous computing to address an important multiobjective problem in bioinformatics: phylogenetics. A parallel metaheuristic approach, based on the joint exploitation of parallel tasks at the algorithm, iteration, and solution levels, is proposed to tackle computationally intensive inferences on CPU+GPU systems. Different heterogeneous design alternatives are also discussed, in accordance with the way the interactions between CPU and GPU are handled. The experimental evaluation of the proposal on real-world biological datasets points out the benefits of using multi-level, heterogeneous strategies, reporting accelerations up to 396x over the baseline metaheuristic as well as significant energy savings with regard to other parallel approaches, without impacting multiobjective solution quality. (C) 2021 The Author(s). Published by Elsevier B.V.

关键词： Heterogeneous computing multi-level parallelism Evolutionary computation High performance computing Bioinformatics

来源：评论

学校读者我要写书评

暂无评论

Optimum parallelism in Spark Framework on Hadoop YARN for Maximum Cluster Resource Utilization 1st

Optimum Parallelism in Spark Framework on Hadoop YARN for Ma...

引用

1st International Conference on Sustainable Technologies for Computational Intelligence (ICTSCI)

作者： Janardhanan, P. S. Samuel, Philip SunTec Business Solut Pvt Ltd TechnoPk Trivandrum Kerala India Cochin Univ Sci & Technol Dept Comp Sci Kochi Kerala India

ISBN: (纸本)9789811500299;9789811500282

Spark iswidely used as a distributed computing framework for in-memory parallel processing. It implements distributed computing by splitting the jobs into tasks and deploying them on executors on the nodes of a cluster. Executors are JVMs with dedicated allocation of CPU cores and memory. The number of tasks depends on the partitions of input data. Depending on the number of CPU cores allocated to executors, one or more cores get allocated to one task. Tasks run as independent threads on executors hosted on JVMs dedicated exclusively to the executor. One or more executors are deployed on the nodes of the cluster depending on the resource availability. The performance advantage provided by distributed computing on Spark framework depends on the level of parallelism configured at 3 levels, namely node level, executor level, and task level. The parallelism at each of these levels should be configured to fully utilize the available computing resources. This paper recommends optimum parallelism configuration for Apache Spark framework deployed on Hadoop YARN cluster. The recommendations are based on the results of the experiments conducted to evaluate the dependency of parallelism at each of these levels on the performance of Spark applications. For the purpose of the evaluation, a CPU-intensive job and an I/O-intensive job are used. The performance is measured by varying the parallelism at each of the 3 levels. The results presented in this paper help Spark users in selecting optimum parallelism at each of these levels for achieving maximum performance for Spark jobs by maximum resource utilization.

关键词： Distributed computing Apache Spark Hadoop YARN SparkBench Spark configuration multi-level parallelism Resource optimization

来源：评论

学校读者我要写书评

暂无评论

TIMEDELN: A programme for the detection and parametrization of overlapping resonances using the time-delay method

引用

COMPUTER PHYSICS COMMUNICATIONS 2017年 215卷 137-148页

作者： Little, Duncan A. Tennyson, Jonathan Plummer, Martin Noble, Clifford J. Sunderland, Andrew G. UCL Dept Phys & Astron Gower St London WC1E 6BT England Sri Tech Daresbury STEC Daresbury Lab Sci Comp Dept Daresbury WA4 4AD Cheshire England

TIMEDELN implements the time-delay method of determining resonance parameters from the characteristic Lorentzian form displayed by the largest eigenvalues of the time-delay matrix. TIMEDELN constructs the time-delay matrix from input K-matrices and analyses its eigenvalues. This new version implements multi-resonance fitting and may be run serially or as a high performance parallel code with three levels of parallelism. TIMEDELN takes K-matrices from a scattering calculation, either read from a file or calculated on a dynamically adjusted grid, and calculates the time-delay matrix. This is then diagonalized, with the largest eigenvalue representing the longest time-delay experienced by the scattering particle. A resonance shows up as a characteristic Lorentzian form in the time-delay: the programme searches the time-delay eigenvalues for maxima and traces resonances when they pass through different eigenvalues, separating overlapping resonances. It also performs the fitting of the calculated data to the Lorentzian form and outputs resonance positions and widths. Any remaining overlapping resonances can be fitted jointly. The branching ratios of decay into the open channels can also be found. The programme may be run serially or in parallel with three levels of parallelism. The parallel code modules are abstracted from the main physics code and can be used independently. New version programme summary Programme Title: TIMEDELN Programme Files doi: http://***/10.17632/wmv4f42xnz.1 Licencing provisions: MIT Programming language: FORTRAN Journal reference of previous version: Computer Phys. Comms., 114, 236-242 (1998). Does the new version supersede the previous version?: Yes Nature of problem: TIMEDELN detects and parametrizes resonances, including overlapping resonances when provided with the K-matrix of the scattering problem. Solution method: Resonances are identified by peaks in the largest few eigenvalues of the time-delay matrix. Reasons for the new version:

关键词： (Overlapping) resonances Scattering Time-delay K-matrix Branching ratio Eigenphase multi-level parallelism

来源：评论

学校读者我要写书评

暂无评论

RTL Design of a Dynamically Reconfigurable Cell Array for multimedia Processing 4

RTL Design of a Dynamically Reconfigurable Cell Array for Mu...

引用

4th National-Foundation-for-Science-and-Technology-Development (NAFOSTED) Conference on Information and Computer Science (NICS)

作者： Nguyen, Hung K. Phan, Minh T. VNU Univ Technol & Engn VNU Key Lab Smart Integrated Syst Hanoi Vietnam

ISBN: (纸本)9781538632109

This paper presents the design of a Coarse-grained Reconfigurable Architecture (CGRA), called MUSRA (multimedia Specific Reconfigurable Architecture). The MUSRA is proposed to exploit multi-level parallelism of the computation-intensive loops in multimedia processing applications. To solve the huge bandwidth requirement of parallel processing arrays, the proposed architecture focuses on the exploitation of data locality to reduce data access bandwidth and increase efficiency of pipelined execution of the kernel loops. The MUSRA also supports the capability of dynamic reconfiguration by enabling the hardware fabrics to be reconfigured into different functions even if the system is working. The proposed architecture has been modeled at Register Transfer level (RTL) by using VHDL language. Some benchmark applications have been mapped onto the MUSRA in order to validate the high flexibility and performance of the architecture that is suitable for a wide range of multimedia processing applications. The proposed CGRA can be applied as a reconfigurable hardware IP (Intellectual Property) core in reconfigurable high-performance System-on-Chips.

关键词： CGRA Data locality multi-level parallelism Reconfigurable computing parallel processing MUSRA

来源：评论

学校读者我要写书评

暂无评论

GridPACK^TM: A framework for developing power grid simulations on high-performance computing platforms

引用

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS 2016年第2期30卷 223-240页

作者： Palmer, Bruce Perkins, William Chen, Yousu Jin, Shuangshuang Callahan, David Glass, Kevin Diao, Ruisheng Rice, Mark Elbert, Stephen Vallem, Mallikarjuna Huang, Zhenyu Pacific NW Natl Lab Box 999 Richland WA 99352 USA Pacific NW Natl Lab Seattle WA USA Facebook Seattle WA USA

This paper describes the GridPACK(TM) framework, which is designed to help power grid engineers develop software capable of running on high-performance computers. The framework makes extensive use of software templates to provide high-level functionality while still providing flexibility to easily implement a broad range of models and algorithms. GridPACK(TM) contains modules for setting up distributed power grid networks, supporting application-specific bus and branch models, creating distributed matrices and vectors and using parallel linear and non-linear solvers. It also provides mappers to create matrices and vectors based on properties of the network and functionality to support Input/Output (IO) and to manage errors. The goal of GridPACK(TM) is to substantially reduce the complexity of writing software for parallel computers while still providing efficient and scalable software solutions. The use of GridPACK(TM) is illustrated for a simple powerflow example and performance results for the powerflow and dynamic simulations are discussed.

关键词： Electric power grid high-performance computing software frameworks distributed graphs multi-level parallelism

来源：评论

学校读者我要写书评

暂无评论

Architecture, implementation and parallelization of the software to search for periodic gravitational wave signals

引用

COMPUTER PHYSICS COMMUNICATIONS 2015年 188卷 167-176页

作者： Poghosyan, G. Matta, S. Streit, A. Bejger, M. Krolak, A. Karlsruhe Inst Technol Steinbuch Ctr Comp D-76131 Karlsruhe Germany PEC Univ Technol Chandigarh 160012 India Polish Acad Sci N Copernicus Astron Ctr PL-00716 Warsaw Poland Polish Acad Sci Inst Math PL-00956 Warsaw Poland

The parallelization, design and scalability of the PolGrawAllSky code to search for periodic gravitational waves from rotating neutron stars is discussed. The code is based on an efficient implementation of the F-statistic using the Fast Fourier Transform algorithm. To perform an analysis of data from the advanced LIGO and Virgo gravitational wave detectors' network, which will start operating in 2015, hundreds of millions of CPU hours will be required the code utilizing the potential of massively parallel supercomputers is therefore mandatory. We have parallelized the code using the Message Passing Interface standard, implemented a mechanism for combining the searches at different sky-positions and frequency bands into one extremely scalable program. The parallel I/O interface is used to escape bottlenecks, when writing the generated data into file system. This allowed to develop a highly scalable computation code, which would enable the data analysis at large scales on acceptable time scales. Benchmarking of the code on a Cray XE6 system was performed to show efficiency of our parallelization concept and to demonstrate scaling up to 50 thousand cores in parallel. Program summary Program title: parallel PolGrawAllSky Catalogue identifier: AEUX_v1_0 Program summary URL: http://***/summaries/AEUX_v1_*** Program obtainable from: CPC Program Library, Queen's University, Belfast, N. Ireland Licensing provisions: Standard CPC licence, http://***/licence/*** No. of lines in distributed program, including test data, etc.: 163747 No. of bytes in distributed program, including test data, etc.: 28989030 Distribution format: *** Programming language: C. Computer: Any parallel computing platform supporting MPI standard. Operating system: Linux as well any other supporting MPI standard. Has the code been vectorized or parallelized?: Yes, using MPI. Tested with up to 50208 processors RAM: 1 Gigabyte per parallel task Classification: 1.5. E

关键词： Parallelization MPI MPI I/O HPC Gravitational waves F-statistic multi-level parallelism Farm skeletons

来源：评论

学校读者我要写书评

暂无评论

Fast iterative method in solving eikonal equations : a multi-level parallel approach

Fast iterative method in solving eikonal equations : a multi...

引用

14th Annual International Conference on Computational Science

作者： Dang, Florian Emad, Nahid Silkan Meudon France Univ Versailles Laboratoire PRiSM Versailles France CNRS Maison Simulat Saclay France

The fast marching method is widely used to solve the eikonal equation. By introducing a new way of managing propagation interfaces which avoid the use of expensive data structures, the fast iterative method reveals to be a faster variant with a higher parallel potential compared to the fast marching method. We investigate in this paper a multi-level parallel approach for the fast iterative method which is well fitted for today heterogenous and hierarchical architectures. We show experiment results which focus on the fine-grained parallel level of the algorithm and we give a performance analysis.

关键词： eikonal equation fast iterative method multi-level parallelism Hamilton-Jacobi equations fast marching method parallel upwind finite difference

来源：评论

学校读者我要写书评

暂无评论

Fast Iterative Method in Solving Eikonal Equations: A multi-level Parallel Approach

引用

Procedia Computer Science 2014年 29卷 1859-1869页

作者： Florian Dang Nahid Emad Silkan. Meudon-la-Forêt France Laboratoire PRiSM. Université de Versailles France Maison de la Simulation. CNRS Saclay France

关键词： eikonal equation fast iterative method multi-level parallelism Hamilton-Jacobi equations fast marching method parallel upwind finite difference

来源：评论

学校读者我要写书评

暂无评论

multi-level Parallel Computing of Reverse Time Migration for Seismic Imaging on Blue Gene/Q 13

Multi-Level Parallel Computing of Reverse Time Migration for...

引用

18th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming

作者： Lu, Ligang Magerlein, Karen IBM Watson Res Ctr Yorktown Hts NY USA

ISBN: (纸本)9781450319225

Blue Gene/Q (BG/Q) is an early representative of increasing scale and thread count that will characterize future HPC systems: large counts of nodes, cores, and threads; and a rich programming environment with many degrees of freedom in parallel computing optimization. So it is both a challenge and an opportunity to it to accelerate the seismic imaging applications to the unprecedented levels that will significantly advance the technologies for the oil and gas industry. In this work we aim to address two important questions: how HPC systems with high levels of scale and thread count will perform in real applications; and how systems with many degrees of freedom in parallel programming can be calibrated to achieve optimal performance. Based on BG/Q's architecture features and RTM workload characteristics, we developed massive domain partition, MPI , and SIMD Our detailed deep analyses in various aspects of optimization also provide valuable experience and insights into how can be utilized to facilitate the advance of seismic imaging technologies. Our BG/Q RTM solution achieved a 14.93x speedup over the BG/P implementation. Our multi-level parallelism strategies for Reverse Time Migration (RTM) seismic imaging computing on BG/Q provides an example of how HPC systems like BG/Q can accelerate applications to a new level.

关键词： multi-level parallelism Parallel Computing High Performance Computing Performance Optimization Reverse Time Migration Seismic Imaging

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：