Programmable Code Optimization and Empirical Tuning For High-end Computing
高端计算的可编程代码优化和经验调整
基本信息
- 批准号:1261778
- 负责人:
- 金额:$ 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:2012-09-01 至 2014-08-31
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
The complexity of modern high-end computers has made it exceedingly difficult for scientific applications to effectively manage resources such as extreme-scale parallelism, single-chip multi-processors, and deep hierarchy of shared/distributed caches and memories. In particular, as machines and applications have both evolved to become complex and massively parallel, compilers have failed to automatically bridge the gap between complex software and diverse hardware platforms.Optimization models for parallel computing have lagged far behind those for serial applications, and conventional compilers are increasingly unable to accommodate emerging high-end architectures.This research develops a new optimization model that allows1) developers to effectively interact with advanced optimizing compilers to provide both domain-specific knowledge and high-level optimization strategies (e.g., directions to enable new or choose amongst differing parallelization strategies); 2) computational specialists to easily define arbitrary domain-specific transformations to directly control performance optimizations to their code; 3) architecture-sensitive optimizations to be easily parameterized and empirically tuned to achieve portable high performance.The optimization model is supported with an integrated environment that contains two main components: ROSE, a C/C++/Fortran2003 source-to-source optimizing compiler developed at DOE/LLNL; and POET, a transformation language together with an empirical optimization engine developed at UTSA. This framework permits different levels of automation and programmer intervention, from fully-automated tuning to semi-automated development to fully programmable control. The research targets both the optimization needs of computational kernels and the more general requirements of whole program optimizations. The framework is integrated as an external development mechanism for the widely-adopted ATLAS library and is connected with empirical tuning research under DOE SciDAC program to improve the efficiency of large-scale scientific applications.
现代高端计算机的复杂性使科学应用程序非常困难,可以有效地管理诸如极端尺度并行性,单芯片多处理器以及共享/分布式缓存和记忆的深层层次结构等资源。特别是,随着机器和应用都已经发展为复杂且大规模平行,编译器未能自动弥合复杂软件和各种硬件平台之间的差距。平行计算的优化模型已远远落后于串行应用程序的差异,并且越来越多地与传统的编译器相互作用,无法促进效果的效果。编译器提供特定于领域的知识和高级优化策略(例如,启用新的或选择不同的并行化策略的方向); 2)计算专家可以轻松定义特定于特定领域的转换,以直接控制其代码的性能优化; 3)易于参数化和经验调整以实现便携式高性能的构造敏感优化。优化模型由一个包含两个主要组件的集成环境支持:rose,c/c ++/fortran2003 source-fortran2003在doe/llnl上开发的源源源源对源优化编译器;和诗人,一种转型语言,以及在UTSA开发的经验优化引擎。该框架允许不同级别的自动化和程序员干预,从完全自动化的调整到半自动化开发到完全可编程的控制。该研究既针对计算内核的优化需求,又针对整个程序优化的更一般要求。 该框架被整合为广泛采用的地图集文库的外部开发机制,并与DOE SCIDAC计划下的经验调整研究有关,以提高大型科学应用的效率。
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)

暂无数据
数据更新时间:2024-06-01
Qing Yi其他文献
An Evaluation of Rater Agreement Indices Using Generalizability Theory
使用概括性理论评估评估者一致性指数
- DOI:10.1007/978-3-319-77249-3_710.1007/978-3-319-77249-3_7
- 发表时间:20172017
- 期刊:
- 影响因子:0
- 作者:Dongmei Li;Qing Yi;Benjamin AndrewsDongmei Li;Qing Yi;Benjamin Andrews
- 通讯作者:Benjamin AndrewsBenjamin Andrews
Automatically Optimizing Stencil Computations on Many-Core NUMA Architectures
自动优化多核 NUMA 架构上的模板计算
- DOI:
- 发表时间:20162016
- 期刊:
- 影响因子:0
- 作者:Pei;Qing Yi;D. Quinlan;C. Liao;Yongqing YanPei;Qing Yi;D. Quinlan;C. Liao;Yongqing Yan
- 通讯作者:Yongqing YanYongqing Yan
R-ISS Stage-Dependent Single-Cell Sequencing Analysis Uncovers Oncogenes and Potential Immunotherapeutic Targets in Multiple Myeloma
- DOI:10.1182/blood-2023-17965410.1182/blood-2023-179654
- 发表时间:2023-11-022023-11-02
- 期刊:
- 影响因子:
- 作者:Ling Zhong;Ji Luo;Lan Luo;Qing Yi;Tao JiangLing Zhong;Ji Luo;Lan Luo;Qing Yi;Tao Jiang
- 通讯作者:Tao JiangTao Jiang
Collective Specification and Verification of Behavior Models and Object-oriented Implementations
行为模型和面向对象实现的集体规范和验证
- DOI:10.5220/000343930015002410.5220/0003439300150024
- 发表时间:20112011
- 期刊:
- 影响因子:0
- 作者:Qing Yi;Jianwei Niu;Anitha R. MarneniQing Yi;Jianwei Niu;Anitha R. Marneni
- 通讯作者:Anitha R. MarneniAnitha R. Marneni
Evaluation of the Technical Performance of Football Players in the UEFA Champions League
欧洲冠军联赛足球运动员技术表现评价
- DOI:10.3390/ijerph1702060410.3390/ijerph17020604
- 发表时间:20202020
- 期刊:
- 影响因子:0
- 作者:Qing Yi;M. Gómez;Hongyou Liu;Shaoliang Zhang;B. Gao;Fabian Wunderlich;D. MemmertQing Yi;M. Gómez;Hongyou Liu;Shaoliang Zhang;B. Gao;Fabian Wunderlich;D. Memmert
- 通讯作者:D. MemmertD. Memmert
共 41 条
- 1
- 2
- 3
- 4
- 5
- 6
- 9
Qing Yi的其他基金
SHF: Small: Whole-application Coordiated Parallelization Through The Optimization Of Abstraction Hierarchies
SHF:小型:通过抽象层次结构的优化实现全应用程序协调并行化
- 批准号:19104881910488
- 财政年份:2019
- 资助金额:$ 11万$ 11万
- 项目类别:Standard GrantStandard Grant
I-Corps: Optimized Compiler Applications
I-Corps:优化的编译器应用程序
- 批准号:17237121723712
- 财政年份:2017
- 资助金额:$ 11万$ 11万
- 项目类别:Standard GrantStandard Grant
SHF: Small: Specializing Compilers For High Performance Computing Through Coordinated Data and Algorithm Optimizations
SHF:小型:通过协调数据和算法优化实现高性能计算的专用编译器
- 批准号:14214431421443
- 财政年份:2014
- 资助金额:$ 11万$ 11万
- 项目类别:Standard GrantStandard Grant
CAREER: Multilayer Code Synthesis For Correctness and Performance
职业:多层代码合成以确保正确性和性能
- 批准号:12618111261811
- 财政年份:2012
- 资助金额:$ 11万$ 11万
- 项目类别:Continuing GrantContinuing Grant
SHF: Small: Collaborative Research: Programming Interface And Runtime For Self-Tuning Scalable C/C++ Data Structures
SHF:小型:协作研究:自调整可扩展 C/C 数据结构的编程接口和运行时
- 批准号:12615841261584
- 财政年份:2012
- 资助金额:$ 11万$ 11万
- 项目类别:Standard GrantStandard Grant
SHF: Small: Collaborative Research: Programming Interface And Runtime For Self-Tuning Scalable C/C++ Data Structures
SHF:小型:协作研究:自调整可扩展 C/C 数据结构的编程接口和运行时
- 批准号:12181791218179
- 财政年份:2012
- 资助金额:$ 11万$ 11万
- 项目类别:Standard GrantStandard Grant
CAREER: Multilayer Code Synthesis For Correctness and Performance
职业:多层代码合成以确保正确性和性能
- 批准号:07473570747357
- 财政年份:2008
- 资助金额:$ 11万$ 11万
- 项目类别:Continuing GrantContinuing Grant
Programmable Code Optimization and Empirical Tuning For High-end Computing
高端计算的可编程代码优化和经验调整
- 批准号:08332030833203
- 财政年份:2008
- 资助金额:$ 11万$ 11万
- 项目类别:Standard GrantStandard Grant
相似国自然基金
代码混淆和编译优化鲁棒的函数签名重构
- 批准号:62302193
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:青年科学基金项目
战略与管理研究类:海洋工程学科代码优化方案研究
- 批准号:52342103
- 批准年份:2023
- 资助金额:30 万元
- 项目类别:专项基金项目
矿业、石油与安全学科代码优化战略研究
- 批准号:52242403
- 批准年份:2022
- 资助金额:15.00 万元
- 项目类别:专项项目
冶金与材料工程学科代码优化战略研究
- 批准号:52242404
- 批准年份:2022
- 资助金额:15.00 万元
- 项目类别:专项项目
新概念材料与材料共性科学学科申请代码与研究方向的调整优化研究
- 批准号:52242310
- 批准年份:2022
- 资助金额:10.00 万元
- 项目类别:专项项目
相似海外基金
Opioid-Sparing Non-Surgical, Bioresorbable Nerve Stimulator for Pain Relief
节省阿片类药物的非手术生物可吸收神经刺激器,用于缓解疼痛
- 批准号:1075964210759642
- 财政年份:2023
- 资助金额:$ 11万$ 11万
- 项目类别:
Increasing the Value of Genomic Medicine through Private Pharmacogenomic Reporting
通过私人药物基因组报告增加基因组医学的价值
- 批准号:1076011910760119
- 财政年份:2023
- 资助金额:$ 11万$ 11万
- 项目类别:
Harmony AI: Natural Language Processing Enabling Advanced Biomanufacturing
Harmony AI:自然语言处理实现先进生物制造
- 批准号:1076108210761082
- 财政年份:2023
- 资助金额:$ 11万$ 11万
- 项目类别:
Computational Infrastructure for Automated Force Field Development and Optimization
用于自动力场开发和优化的计算基础设施
- 批准号:1069920010699200
- 财政年份:2023
- 资助金额:$ 11万$ 11万
- 项目类别:
Harmony AI: State of the Art Natural Language Processing for Genetic Engineering
Harmony AI:用于基因工程的最先进的自然语言处理
- 批准号:1069880510698805
- 财政年份:2023
- 资助金额:$ 11万$ 11万
- 项目类别: