Dynamic thread mapping for high-performance, power-efficient heterogeneous many-core systems

2013 IEEE 31st International Conference on Computer Design (ICCD) Pub Date : 2013-11-07 DOI:10.1109/ICCD.2013.6657025

Guangshuo Liu, Jinpyo Park, Diana Marculescu

{"title":"Dynamic thread mapping for high-performance, power-efficient heterogeneous many-core systems","authors":"Guangshuo Liu, Jinpyo Park, Diana Marculescu","doi":"10.1109/ICCD.2013.6657025","DOIUrl":null,"url":null,"abstract":"This paper addresses the problem of dynamic thread mapping in heterogeneous many-core systems via an efficient algorithm that maximizes performance under power constraints. Heterogeneous many-core systems are composed of multiple core types with different power-performance characteristics. As well documented in the literature, the generic mapping problem is an NP-complete problem which can be formulated as a 0-1 integer linear program, therefore, prohibitively expensive to solve optimally in an online scenario. However, in real applications, thread mapping decisions need to be responsive to workload phase changes. This paper proposes an iterative approach bounding the runtime as O(n2/m), for mapping multi-threaded applications on n cores comprising of m core types. Compared with an optimal solution, the proposed algorithm produces results less than 0.6% away from optimum on average, with two orders of magnitude improvement in runtime. Results show that performance improvement can reach 16% under iso-power constraints compared to a random mapping. The algorithm can be brought online for hundred-core heterogeneous systems as it scales to systems comprised of 256 cores with less than one millisecond in overhead.","PeriodicalId":398811,"journal":{"name":"2013 IEEE 31st International Conference on Computer Design (ICCD)","volume":"477 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"62","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 IEEE 31st International Conference on Computer Design (ICCD)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCD.2013.6657025","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 62

Abstract

This paper addresses the problem of dynamic thread mapping in heterogeneous many-core systems via an efficient algorithm that maximizes performance under power constraints. Heterogeneous many-core systems are composed of multiple core types with different power-performance characteristics. As well documented in the literature, the generic mapping problem is an NP-complete problem which can be formulated as a 0-1 integer linear program, therefore, prohibitively expensive to solve optimally in an online scenario. However, in real applications, thread mapping decisions need to be responsive to workload phase changes. This paper proposes an iterative approach bounding the runtime as O(n2/m), for mapping multi-threaded applications on n cores comprising of m core types. Compared with an optimal solution, the proposed algorithm produces results less than 0.6% away from optimum on average, with two orders of magnitude improvement in runtime. Results show that performance improvement can reach 16% under iso-power constraints compared to a random mapping. The algorithm can be brought online for hundred-core heterogeneous systems as it scales to systems comprised of 256 cores with less than one millisecond in overhead.

查看原文本刊更多论文

用于高性能、高能效异构多核系统的动态线程映射

本文通过一种有效的算法解决了异构多核系统中的动态线程映射问题，该算法在功率限制下最大化了性能。异构多核系统是由具有不同功率性能特征的多个核类型组成的系统。正如文献记载的那样，一般映射问题是一个np完全问题，可以被表述为0-1整数线性规划，因此，在在线场景中最优解决的代价非常昂贵。然而，在实际应用程序中，线程映射决策需要响应工作负载阶段的变化。本文提出了一种迭代方法，将运行时限定为O(n2/m)，用于在包含m个核心类型的n个核心上映射多线程应用程序。与最优解相比，本文算法产生的结果与最优解的平均误差小于0.6%，运行时间提高了两个数量级。结果表明，与随机映射相比，在等功率约束下，性能提高可达16%。该算法可以在线用于百核异构系统，因为它可以扩展到由256核组成的系统，开销不到1毫秒。

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

2013 IEEE 31st International Conference on Computer Design (ICCD)

自引率

0.00%

发文量