SI2-SSE Collaborative Research: SPIKE-An Implementation of a Recursive Divide-and-Conquer Parallel Strategy for Solving Large Systems of Linear Equations
SI2-SSE 合作研究:SPIKE——求解大型线性方程组的递归分治并行策略的实现
基本信息
- 批准号:1147337
- 负责人:
- 金额:$ 25.11万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Standard Grant
- 财政年份:2012
- 资助国家:美国
- 起止时间:2012-06-01 至 2015-05-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
Drs. Negrut, Sameh, and Knepley will investigate, produce, and maintain a methodology and its software implementation that leverage emerging heterogeneous hardware architectures to solve billion-unknowns linear systems in a robust, scalable, and efficient fashion. The two classes of problems targeted under this project are banded dense and sparse general linear systems.This project is motivated by the observation that the task of solving a linear system is one of the most ubiquitous ingredients in the numerical solution of Applied Mathematics problems. It is relied upon for the implicit integration of Ordinary Differential Equation (ODE) and Differential Algebraic Equation (DAE) problems, in the numerical solution of Partial Differential Equation (PDE) problems, in interior point optimization methods, in least squares approximations, in solving eigenvalue problems, and in data analysis. In fact, the vast majority of nonlinear problems in Scientific Computing are solved iteratively by drawing on local linearizations of nonlinear operators and the solution of linear systems. Recent advances in (a) hardware architecture; i.e., the emergence of General Purpose Graphics Processing Unit (GP-GPU) cards, and (b) scalable solution algorithms, provide an opportunity to develop a new class of parallel algorithms, called SPIKE, which can robustly and efficiently solve very large linear systems of equations.Drawing on its divide-and-conquer paradigm, SPIKE builds on several algorithmic primitives: matrix reordering strategies, dense linear algebra operations, sparse direct solvers, and Krylov subspace methods. It provides a scalable solution that can be deployed in a heterogeneous hardware ecosystem and has the potential to solve billion-unknown linear systems in the cloud or on tomorrow?s exascale supercomputers. Its high degree of scalability and improved efficiency stem from (i) optimized memory access pattern owing to an aggressive pre-processing stage that reduces a generic sparse matrix to a banded one through a novel reordering strategy; (ii) good exposure of coarse and fine grain parallelism owing to a recursive, divide-and-conquer solution strategy; (iii) efficient vectorization in evaluating the coupling terms in the divide-and-conquer stage owing to a CPU+GPU heterogeneous computing approach; and (iv) algorithmic polymorphism, given that SPIKE can serve both as a direct solver or an effective preconditioner in an iterative Krylov-type method.In Engineering, SPIKE will provide the Computer Aided Engineering (CAE) community with a key component; i.e., fast solution of linear systems, required by the analysis of complex problems through computer simulation. Examples of applications that would benefit from this technology are Structural Mechanics problems (Finite Element Analysis in car crash simulation), Computational Fluid Dynamics problems (solving Navier-Stokes equations in the simulation of turbulent flow around a wing profile), and Computational Multibody Dynamics problems (solving Newton-Euler equations in large granular dynamics problems).SPIKE will also be interfaced to the Portable, Extensible Toolkit for Scientific Computation (PETSc), a two decades old flexible and scalable framework for solving Science and Engineering problems on supercomputers. Through PETSc, SPIKE will be made available to a High Performance Computing user community with more than 20,000 members worldwide. PETSc users will be able to run SPIKE without any modifications on vastly different supercomputer architectures such as the IBM BlueGene/P and BlueGene/Q, or the Cray XT5. SPIKE will thus run scalably on the largest machines in the world and will be tuned for very different network and hardware topologies while maintaining a simple code base.The experience collected and lessons learned in this project will augment a graduate level class, ?High Performance Computing for Engineering Applications? taught at the University of Wisconsin-Madison. A SPIKE tutorial and research outcomes will be presented each year at the International Conference for High Performance Computing, Networking, Storage and Analysis. A one day High Performance Computing Boot Camp will be organized each year in conjunction with the American Society of Mechanical Engineers (ASME) conference and used to disseminate the software outcomes of this effort. Finally, this project will shape the research agendas of two graduate students working on advanced degrees in Computational Science.
Drs。Negrut、Sameh和Knepley将研究、生产和维护一种方法及其软件实现,该方法利用新兴的异构硬件架构,以健壮、可扩展和高效的方式解决数十亿未知的线性系统。本课题研究的两类问题是带状密集和稀疏一般线性系统。这个项目的动机是观察到解决线性系统的任务是应用数学问题的数值解决中最普遍的成分之一。它被用于常微分方程(ODE)和微分代数方程(DAE)问题的隐式积分,偏微分方程(PDE)问题的数值解,内点优化方法,最小二乘近似,求解特征值问题以及数据分析。事实上,科学计算中的绝大多数非线性问题都是通过利用非线性算子的局部线性化和线性系统的解来迭代解决的。(a)硬件架构的最新进展;也就是说,通用图形处理单元(GP-GPU)卡的出现,以及(b)可扩展的解决算法,为开发一类新的并行算法提供了机会,称为SPIKE,它可以鲁棒且有效地求解非常大的线性方程组。利用其分而治之的范例,SPIKE建立在几个算法基元上:矩阵重新排序策略、密集线性代数操作、稀疏直接求解器和Krylov子空间方法。它提供了一个可扩展的解决方案,可以部署在异构硬件生态系统中,并且有潜力解决云中或未来数十亿未知的线性系统。百亿亿次超级计算机。它的高度可扩展性和提高的效率源于:(1)通过一种新颖的重排序策略,通过积极的预处理阶段将一般稀疏矩阵减少到带状矩阵,从而优化了内存访问模式;(ii)由于递归的分而治之的解决策略,可以很好地暴露粗颗粒和细颗粒的并行性;(iii)由于CPU+GPU异构计算方法,在分而治之阶段对耦合项进行有效的矢量化评估;(iv)算法多态性,因为在迭代krylov型方法中,SPIKE既可以作为直接求解器,也可以作为有效的前置条件。在工程方面,SPIKE将为计算机辅助工程(CAE)社区提供关键组件;即,线性系统的快速解,需要通过计算机模拟分析复杂问题。从该技术中受益的应用实例包括结构力学问题(汽车碰撞模拟中的有限元分析)、计算流体动力学问题(在模拟机翼轮廓周围的湍流中求解Navier-Stokes方程)和计算多体动力学问题(在大颗粒动力学问题中求解牛顿-欧拉方程)。SPIKE还将与便携式可扩展科学计算工具包(PETSc)连接,PETSc是一个已有20年历史的灵活可扩展框架,用于解决超级计算机上的科学和工程问题。通过PETSc, SPIKE将提供给全球拥有超过20,000名成员的高性能计算用户社区。PETSc用户将能够在完全不同的超级计算机架构(如IBM BlueGene/P和BlueGene/Q或Cray XT5)上运行SPIKE而无需任何修改。因此,SPIKE将在世界上最大的机器上可扩展地运行,并将针对非常不同的网络和硬件拓扑进行调优,同时保持简单的代码库。在这个项目中收集到的经验和教训将增加研究生水平的课程。工程应用的高性能计算?曾在威斯康星大学麦迪逊分校任教。每年将在国际高性能计算、网络、存储和分析会议上发表一篇SPIKE教程和研究成果。每年将与美国机械工程师协会(ASME)会议一起组织为期一天的高性能计算新兵训练营,并用于传播这项工作的软件成果。最后,这个项目将塑造两名攻读计算科学高级学位的研究生的研究议程。
项目成果
期刊论文数量(0)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
数据更新时间:{{ journalArticles.updateTime }}
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Dan Negrut其他文献
Linear Algebra Considerations for the Multi-Threaded Simulation of Mechanical Systems
- DOI:
10.1023/a:1024515521451 - 发表时间:
2003-08-01 - 期刊:
- 影响因子:2.400
- 作者:
Dan Negrut - 通讯作者:
Dan Negrut
Human-automated vehicle interactions: Voluntary driver intervention in car-following
人机交互车辆:在跟车过程中驾驶员的自愿干预
- DOI:
10.1016/j.trc.2024.104969 - 发表时间:
2025-02-01 - 期刊:
- 影响因子:7.900
- 作者:
Xinzhi Zhong;Yang Zhou;Amudha Varshini Kamaraj;Zhenhao Zhou;Wissam Kontar;Dan Negrut;John D. Lee;Soyoung Ahn - 通讯作者:
Soyoung Ahn
Using high fidelity discrete element simulation to calibrate an expeditious terramechanics model in a multibody dynamics framework
使用高保真离散元模拟在多体动力学框架中校准一个快速的岩土力学模型
- DOI:
10.1007/s11044-024-10051-z - 发表时间:
2025-01-22 - 期刊:
- 影响因子:2.400
- 作者:
Yuemin Zhang;Junpeng Dai;Wei Hu;Dan Negrut - 通讯作者:
Dan Negrut
Dan Negrut的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Dan Negrut', 18)}}的其他基金
Collaborative Research: Frameworks: Simulating Autonomous Agents and the Human-Autonomous Agent Interaction
协作研究:框架:模拟自主代理和人机交互
- 批准号:
2209791 - 财政年份:2022
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Research: Differentiable and Expressive Simulators for Designing AI-enabled Robots
协作研究:用于设计人工智能机器人的可微分和富有表现力的模拟器
- 批准号:
2153855 - 财政年份:2022
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Research: Elements:Software:NSCI: Chrono - An Open-Source Simulation Platform for Computational Dynamics Problems
合作研究:Elements:Software:NSCI: Chrono - 计算动力学问题的开源仿真平台
- 批准号:
1835674 - 财政年份:2019
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Towards Modeling & Simulation-Enabled Design of Intelligent Robots A Meeting Dedicated to Identifying Opportunities, Summarizing Challenges, and Brainstorming for Impactful Di
迈向建模
- 批准号:
1830129 - 财政年份:2018
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Using Mixed Discrete-Continuum Representations to Characterize the Dynamics of Large Many-Body Dynamics Problems
使用混合离散连续体表示来表征大型多体动力学问题的动力学
- 批准号:
1635004 - 财政年份:2016
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
GOALI: Computational Multibody Dynamics: Addressing Modeling and Simulation Limitations in Problems with Friction and Contact
GOALI:计算多体动力学:解决摩擦和接触问题中的建模和仿真限制
- 批准号:
1362583 - 财政年份:2014
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
CAREER: Advanced Computational Multi-Body Dynamics for Next Generation Simulation-Based Engineering
职业:下一代基于仿真的工程的高级计算多体动力学
- 批准号:
0840442 - 财政年份:2009
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Research: Simulation of Multibody Dynamics. Leveraging New Numerical Methods and Multiprocessor Capabilities
合作研究:多体动力学模拟。
- 批准号:
0700191 - 财政年份:2007
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
相似国自然基金
化脓性链球菌分泌性酯酶Sse抑制LC3相关吞噬促其侵袭的机制研究
- 批准号:
- 批准年份:2022
- 资助金额:30 万元
- 项目类别:青年科学基金项目
太阳能电池Cu2ZnSn(SSe)4/CdS界面过渡层结构模拟及缺陷态消除研究
- 批准号:
- 批准年份:2022
- 资助金额:55 万元
- 项目类别:面上项目
掺杂实现Cu2ZnSn(SSe)4吸收层表层稳定弱n型特性的第一性原理研究
- 批准号:12004100
- 批准年份:2020
- 资助金额:24.0 万元
- 项目类别:青年科学基金项目
基于SSE的航空信息系统信息安全保障评价指标体系的研究
- 批准号:60776808
- 批准年份:2007
- 资助金额:19.0 万元
- 项目类别:联合基金项目
相似海外基金
Collaborative Research: SI2-SSE: WRENCH: A Simulation Workbench for Scientific Worflow Users, Developers, and Researchers
协作研究:SI2-SSE:WRENCH:面向科学 Worflow 用户、开发人员和研究人员的模拟工作台
- 批准号:
1642369 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
SI2-SSE: Collaborative Research: Integrated Tools for DNA Nanostructure Design and Simulation
SI2-SSE:合作研究:DNA 纳米结构设计和模拟的集成工具
- 批准号:
1740212 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Research: NSCI: SI2-SSE: Time Stepping and Exchange-Correlation Modules for Massively Parallel Real-Time Time-Dependent DFT
合作研究:NSCI:SI2-SSE:大规模并行实时瞬态 DFT 的时间步进和交换相关模块
- 批准号:
1740219 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
SI2-SSE: Collaborative Research: Integrated Tools for DNA Nanostructure Design and Simulation
SI2-SSE:合作研究:DNA 纳米结构设计和模拟的集成工具
- 批准号:
1740282 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Research: SI2-SSE: An open source multi-physics platform to advance fundamental understanding of plasma physics and enable impactful application of plasma systems
合作研究:SI2-SSE:一个开源多物理平台,可促进对等离子体物理学的基本理解并实现等离子体系统的有效应用
- 批准号:
1740300 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
SI2-SSE: Collaborative Research: Software Framework for Strongly Correlated Materials: from DFT to DMFT
SI2-SSE:协作研究:强相关材料的软件框架:从 DFT 到 DMFT
- 批准号:
1740112 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
SI2-SSE: Collaborative Research: A Sustainable Future for the Glue Multi-Dimensional Linked Data Visualization Package
SI2-SSE:协作研究:Glue 多维关联数据可视化包的可持续未来
- 批准号:
1740229 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
SI2-SSE: Collaborative Research: Software Framework for Strongly Correlated Materials: from DFT to DMFT
SI2-SSE:协作研究:强相关材料的软件框架:从 DFT 到 DMFT
- 批准号:
1740111 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Proposal: SI2-SSE: An open source multi-physics platform to advance fundamental understanding of plasma physics and enable impactful application of plasma systems
合作提案:SI2-SSE:一个开源多物理平台,可促进对等离子体物理学的基本理解并实现等离子体系统的有效应用
- 批准号:
1740310 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant
Collaborative Research: SI2-SSE: WRENCH: A Simulation Workbench for Scientific Workflow Users, Developers, and Researchers
协作研究:SI2-SSE:WRENCH:面向科学工作流程用户、开发人员和研究人员的模拟工作台
- 批准号:
1642335 - 财政年份:2017
- 资助金额:
$ 25.11万 - 项目类别:
Standard Grant