The RACECAR heuristic for automatic function specialization on multi-core heterogeneous systems

International Conference on Compilers, Architecture, and Synthesis for Embedded Systems Pub Date : 2012-10-07 DOI:10.1145/2380403.2380423

J. Wernsing, G. Stitt

{"title":"The RACECAR heuristic for automatic function specialization on multi-core heterogeneous systems","authors":"J. Wernsing, G. Stitt","doi":"10.1145/2380403.2380423","DOIUrl":null,"url":null,"abstract":"Embedded systems increasingly combine multi-core processors and heterogeneous resources such as graphics-processing units and field-programmable gate arrays. However, significant application design complexity for such systems caused by parallel programming and device-specific challenges has often led to untapped performance potential. Application developers targeting such systems currently must determine how to parallelize computation, create different device-specialized implementations for each heterogeneous resource, and then determine how to apportion work to each resource. In this paper, we present the RACECAR heuristic to automate the optimization of applications for multi-core heterogeneous systems by automatically exploring implementation alternatives that include different algorithms, parallelization strategies, and work distributions. Experimental results show RACECAR-specialized implementations can effectively incorporate provided implementations and parallelize computation across multiple cores, graphics-processing units, and field-programmable gate arrays, improving performance by an average of 47x compared to a CPU, while the fastest provided implementations are only able to average 33x.","PeriodicalId":136293,"journal":{"name":"International Conference on Compilers, Architecture, and Synthesis for Embedded Systems","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Compilers, Architecture, and Synthesis for Embedded Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2380403.2380423","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 7

Abstract

Embedded systems increasingly combine multi-core processors and heterogeneous resources such as graphics-processing units and field-programmable gate arrays. However, significant application design complexity for such systems caused by parallel programming and device-specific challenges has often led to untapped performance potential. Application developers targeting such systems currently must determine how to parallelize computation, create different device-specialized implementations for each heterogeneous resource, and then determine how to apportion work to each resource. In this paper, we present the RACECAR heuristic to automate the optimization of applications for multi-core heterogeneous systems by automatically exploring implementation alternatives that include different algorithms, parallelization strategies, and work distributions. Experimental results show RACECAR-specialized implementations can effectively incorporate provided implementations and parallelize computation across multiple cores, graphics-processing units, and field-programmable gate arrays, improving performance by an average of 47x compared to a CPU, while the fastest provided implementations are only able to average 33x.

查看原文本刊更多论文

多核异构系统中自动功能专门化的RACECAR启发式算法

嵌入式系统越来越多地结合多核处理器和异构资源，如图形处理单元和现场可编程门阵列。然而，由于并行编程和特定于设备的挑战，这类系统的应用程序设计非常复杂，常常导致未开发的性能潜力。针对此类系统的应用程序开发人员当前必须确定如何并行化计算，为每个异构资源创建不同的设备专用实现，然后确定如何将工作分配给每个资源。在本文中，我们提出了RACECAR启发式算法，通过自动探索包括不同算法、并行化策略和工作分布在内的实现替代方案，自动优化多核异构系统的应用程序。实验结果表明，racecar专用实现可以有效地整合提供的实现，并跨多核、图形处理单元和现场可编程门阵列并行计算，与CPU相比，性能平均提高47倍，而提供的最快实现平均只能提高33倍。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Conference on Compilers, Architecture, and Synthesis for Embedded Systems

自引率

0.00%

发文量