检索结果-内蒙古大学图书馆

shared memory parallelization of data mining algorithms: Techniques, programming interface, and performance

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2005年第1期17卷 71-89页

作者： Jin, RM Yang, G Agrawal, G Ohio State Univ Dept Comp & Informat Sci Columbus OH 43210 USA

With recent technological advances, shared memory parallel machines have become more scalable, and offer large main memories and high bus bandwidths. They are emerging as good platforms for data warehousing and data mining. In this paper, we focus on shared memory parallelization of data mining algorithms. We have developed a series of techniques for parallelization of data mining algorithms, including full replication, full locking, fixed locking, optimized full locking, and cache-sensitive locking. Unlike previous work on shared memory parallelization of specific data mining algorithms, all of our techniques apply to a large number of popular data mining algorithms. In addition, we propose a reduction-object-based interface for specifying a data mining algorithm. We show how our runtime system can apply any of the techniques we have developed starting from a common specification of the algorithm. We have carried out a detailed evaluation of the parallelization techniques and the programming interface. We have experimented with apriori and fp-tree-based association mining, k-means clustering, k-nearest neighbor classifier, and decision tree construction. The main results from our experiments are as follows: 1) Among full replication, optimized full locking, and cache-sensitive locking, there is no clear winner. Each of these three techniques can outperform others depending upon machine and dataset parameters. These three techniques perform significantly better than the other two techniques. 2) Good parallel efficiency is achieved for each of the four algorithms we experimented with, using our techniques and runtime system. 3) The overhead of the interface is within 10 percent in almost all cases. 4) In the case of decision tree construction, combining different techniques turned out to be crucial for achieving high performance.

关键词： shared memory parallelization programming interfaces association mining clustering decision tree construction

来源：评论

学校读者我要写书评

暂无评论

LibreGrowth: A tumor growth code based on reaction-diffusion equations using shared memory

引用

COMPUTER PHYSICS COMMUNICATIONS 2019年 243卷 97-105页

作者： Lujan, E. Rosito, M. S. Soba, A. Suarez, C. Consejo Nacl Invest Cient & Tecn Ctr Simulac Computac Aplicac Tecnol Buenos Aires DF Argentina Univ Buenos Aires Fac Ciencias Exactas & Nat Dept Comp Buenos Aires DF Argentina CONICET UBA Inst Fis Plasma Lab Sistemas Complejos Buenos Aires DF Argentina CONICET UBA Inst Astron & Fis Espacio Buenos Aires DF Argentina Consejo Nacl Invest Cient & Tecn Comis Nacl Energia Atom Buenos Aires DF Argentina

In recent years, in-silico experimentation within the field of oncological medicine has been intensively investigated with the aim of better understanding tumor dynamics and dose-response relationships in cancer treatments. In a series of previous works, Lujan et al. (2018, 2017, 2016)we described the micro-environmental influence on micro-tumor infiltration patterns through in-silico/in-vitro experimentation. Here we present the latest version of the software utilized for, but not limited to, those studies: LibreGrowth, a libre tumor growth code able to simulate the core growth and peripheral tumor cell infiltration, considering a benign and a malignant stages. We implemented a reaction-diffusion based model, with spatially variable diffusion coefficient, into a three-dimensional domain, using C++ and OpenMP over a GNU/Linux system. LibreGrowth aims to provide a flexible implementation for depicting heterogeneous tissues and infiltration processes, and to shed light in current therapy optimization strategies. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Tumor growth models Reaction-diffusion equations shared memory parallelization

来源：评论

学校读者我要写书评

暂无评论

Nested parallelization with OpenMP

引用

INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING 2007年第5期35卷 459-476页

作者： Mey, Dieter an Sarholz, Samuel Terboven, Christian Rhein Westfal TH Aachen Ctr Comp & Commun Aachen Germany

OpenMP is widely accepted as a de facto standard for shared memory parallel programming in Fortran, C and C++. Nested parallelization has been included in the first OpenMP specification, but it took a few years until the first commercially available compilers supported this optional part of the specification. We employed nested parallelization using OpenMP in three production codes: a C++ code for content-based image retrieval, a C++ code for the computation of critical points in multi-block CFD datasets, and a multi-block Navier-Stokes solver written in Fortran90. In this paper we discuss the opportunities as well as the deficiencies of the nested parallelization support in OpenMP.

关键词： OpenMP nested parallelization ccNUMA shared memory parallelization

来源：评论

学校读者我要写书评

暂无评论

A programming interface for NUMA shared-memory clusters

A programming interface for NUMA shared-memory clusters

引用

High Performance Computing and Networking Europe 1997 Conference

作者： Dormanns, M Sprangers, W Ertl, H Bemmerl, T RWTH Aachen Lehrstuhl Betriebssyst D-52056 Aachen Germany

ISBN: (纸本)3540628983

We describe a programming interface for parallel computing on NUMA (Non-Uniform memory Access) shared memory machines. Although the interest in this architecture is rapidly growing and more and more hardware manufacturers offer products of this type, there is still a lack in parallelization support. We developed SMI, the shared memory Interface and implemented it as a library on an SCI-coupled cluster of workstations. It aims at providing sophisticated support to account for the NUMA performance characteristic and to allow a step-by-step parallelization. We show it's application to the parallelization of a sparse matrix computation.

关键词： parallel programming interface shared memory parallelization NUMA multiprocessor

来源：评论

学校读者我要写书评

暂无评论

The Parallel Processing Approach to the Dynamic Programming Algorithm of Knapsack Problem

The Parallel Processing Approach to the Dynamic Programming ...

引用

IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (ElConRus)

作者： Sin, Si Thu Thant Natl Res Univ Elect Technol MIET Inst Microdevices & Control Syst Moscow Russia

ISBN: (纸本)9781665404761

This paper aims at comparing the serial, shared memory parallelization, and distributed memory parallelization of the dynamic programming algorithm for the Knapsack Problem. Knapsack Problem is one of the most popular optimization problems. This is the decision-making problem and uses for real-world situations such as business projects, airline cargo business, cryptography, and decision-making industry processes, etc. The algorithm under consideration is the table-based dynamic programming algorithm based on Bellman's optimality principle. We used the C-HF programming language. To solve this problem on shared memory systems, we used the OpenMP. For the distributed memory parallelization, we employed the MPL The structure of the algorithm, the data distribution, synchronization, and communication schemes are explained in detail. Extensive experiments for the developed algorithms were carried out. The obtained results helped to make a comparative analysis of the developed algorithms.

关键词： parallel computing OpenMP MPI dynamic programming discrete optimization Knapsack Problem shared memory parallelization distributed memory parallelization

来源：评论

学校读者我要写书评

暂无评论

parallelization of scientific applications on the SX-4 - The CSCS/SCSC-NEC joint program in application porting and development

引用

NEC RESEARCH & DEVELOPMENT 1998年第4期39卷 482-490页

作者： Ballabio, M Boverat, M Gasser, L Maric, D Haberhauer, S Henriet, C Hausammann, R Swiss Ctr. for Scientific Computing Switzerland NEC European Supercomputer Systems Germany

With the installation af the NEC SX-4/16 in 1996 at the Swiss Center for Scientific Computing ABSTRACT (CSCS/SCSC), CSCS/SCSC and NEC embarked on a joint program for the porting and development of applications of strategic importance to the Swiss user community, also known as the 'SX-4 Task Force.' The primary objective of this collaborative program was to contribute to the progress of the users' R&D programs by ensuring optimum use of the installed SX-4 supercomputer. The results presented demonstrate the great benefit to the user community from the Swiss Federal Institutes of Technology and the Swiss universities. Significant contributions to computational science in Switzerland could be made. Examples are given where the outstanding performance obtained for key application codes opened the door, in the sense of true feasibility breakthroughs, to novel types of simulations and modeling. Notable examples are the simulation of molecules of unprecedented size and the direct simulation of turbulence at resolutions unattainable thus far.

关键词： supercomputer HPC (High Performance Computing) parallel-vector processor shared memory parallelization MPI (Message Passing Interface) computational fluid dynamics direct simulation of turbulence computational chemistry materials science

来源：评论

学校读者我要写书评

暂无评论

A massively-parallel multicore acceleration of a point contact solid mechanics simulation

引用

Civil-Comp Proceedings 2017年 111卷

作者： Kolman, M. Kosec, G. Parallel and Distributed Systems Laboratory Joef Stefan Institute Ljubljana Slovenia

This paper deals with the numerical determination of the stress and displacement distribution in a solid body subjected to the applied external force. The tackled solid mechanics problem is governed by the Navier-Cauchy equation that describes the deformation within the solid body through the displacement vector field. To obtain the solution, a coupled system of non-linear Partial Differential Equations (PDE) of second order has to be solved. In this paper, the problem is approached by a strong form Moving Least Squares (MLS) based numerical discretization also referred to as a Meshless Local Strong Form Method (MLSM). A generic C++ implementation of a MLSM is used for demonstration of parallel solution of a Point Contact problem on Intel® Xeon Phi™ multicore accelerator. All tests are executed on either the host machine with two Intel® Xeon® E5-2620 v3 6 core processors or offloaded to its 60 core Intel® Xeon Phi™ SE10/7120 series. The shared memory parallelization is implemented through an OpenMP API. © Civil-Comp Press, 2017.

关键词： Point contacts Application programming interfaces (API) Least squares approximations Numerical methods Partial differential equations Displacement vector fields Meshless MLSM Nonlinear partial differential equations OpenMP Parallel implementations shared memory parallelization Stress and displacement distribution

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：