在具有语言扩展和运行时支持的任务数据流中引入推测优化

2012 Data-Flow Execution Models for Extreme Scale Computing Pub Date : 2012-09-19 DOI:10.1109/DFM.2012.15

Nathaniel Azuelos, Yoav Etsion, I. Keidar, A. Zaks, E. Ayguadé

{"title":"在具有语言扩展和运行时支持的任务数据流中引入推测优化","authors":"Nathaniel Azuelos, Yoav Etsion, I. Keidar, A. Zaks, E. Ayguadé","doi":"10.1109/DFM.2012.15","DOIUrl":null,"url":null,"abstract":"We argue that speculation leads to increased parallelism in the coarse-grain dataflow paradigm. To do so, we present a framework for adding speculation in a popular and well-established framework. We specify a limited set of additions to the OmpSs language and changes required in its supporting runtime environment. These modifications enable speculation across the system in a flexible way. We evaluate our implementation using a simple benchmark leading to a promising 10% speedup.","PeriodicalId":130397,"journal":{"name":"2012 Data-Flow Execution Models for Extreme Scale Computing","volume":"111 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Introducing Speculative Optimizations in Task Dataflow with Language Extensions and Runtime Support\",\"authors\":\"Nathaniel Azuelos, Yoav Etsion, I. Keidar, A. Zaks, E. Ayguadé\",\"doi\":\"10.1109/DFM.2012.15\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We argue that speculation leads to increased parallelism in the coarse-grain dataflow paradigm. To do so, we present a framework for adding speculation in a popular and well-established framework. We specify a limited set of additions to the OmpSs language and changes required in its supporting runtime environment. These modifications enable speculation across the system in a flexible way. We evaluate our implementation using a simple benchmark leading to a promising 10% speedup.\",\"PeriodicalId\":130397,\"journal\":{\"name\":\"2012 Data-Flow Execution Models for Extreme Scale Computing\",\"volume\":\"111 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2012-09-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2012 Data-Flow Execution Models for Extreme Scale Computing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/DFM.2012.15\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 Data-Flow Execution Models for Extreme Scale Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DFM.2012.15","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 2

摘要

我们认为，推测导致了粗粒度数据流范式中并行性的增加。为此，我们提出了一个框架，在一个流行和完善的框架中添加投机。我们指定了一组有限的对OmpSs语言的添加和对其支持的运行时环境所需的更改。这些修改使整个系统能够以灵活的方式进行推测。我们使用一个简单的基准测试来评估我们的实现，结果是有希望的10%的加速。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Introducing Speculative Optimizations in Task Dataflow with Language Extensions and Runtime Support

We argue that speculation leads to increased parallelism in the coarse-grain dataflow paradigm. To do so, we present a framework for adding speculation in a popular and well-established framework. We specify a limited set of additions to the OmpSs language and changes required in its supporting runtime environment. These modifications enable speculation across the system in a flexible way. We evaluate our implementation using a simple benchmark leading to a promising 10% speedup.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2012 Data-Flow Execution Models for Extreme Scale Computing

自引率

0.00%

发文量