集成gpu的计算框架方面

2020 19th RoEduNet Conference: Networking in Education and Research (RoEduNet) Pub Date : 2020-12-11 DOI:10.1109/roedunet51892.2020.9324851

Grigore Lupescu, N. Tapus

{"title":"集成gpu的计算框架方面","authors":"Grigore Lupescu, N. Tapus","doi":"10.1109/roedunet51892.2020.9324851","DOIUrl":null,"url":null,"abstract":"The integrated GPU is nowadays found in many consumer and industrial System on Chips, and poses unique challenges to program (thermal, power, shared resources etc.). The goal of this work is to identify several key aspects in a compute framework (basic design, execution models, data communication, fault tolerance and latency), which handle the integrated GPU limitations and provide a proper abstraction.","PeriodicalId":140521,"journal":{"name":"2020 19th RoEduNet Conference: Networking in Education and Research (RoEduNet)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Compute framework aspects for integrated GPUs\",\"authors\":\"Grigore Lupescu, N. Tapus\",\"doi\":\"10.1109/roedunet51892.2020.9324851\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The integrated GPU is nowadays found in many consumer and industrial System on Chips, and poses unique challenges to program (thermal, power, shared resources etc.). The goal of this work is to identify several key aspects in a compute framework (basic design, execution models, data communication, fault tolerance and latency), which handle the integrated GPU limitations and provide a proper abstraction.\",\"PeriodicalId\":140521,\"journal\":{\"name\":\"2020 19th RoEduNet Conference: Networking in Education and Research (RoEduNet)\",\"volume\":\"11 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-12-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 19th RoEduNet Conference: Networking in Education and Research (RoEduNet)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/roedunet51892.2020.9324851\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 19th RoEduNet Conference: Networking in Education and Research (RoEduNet)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/roedunet51892.2020.9324851","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

集成的GPU现在被发现在许多消费和工业系统芯片上，并提出了独特的挑战程序(热，电源，共享资源等)。这项工作的目标是确定计算框架中的几个关键方面(基本设计、执行模型、数据通信、容错和延迟)，这些方面处理集成GPU的限制并提供适当的抽象。

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

Compute framework aspects for integrated GPUs

The integrated GPU is nowadays found in many consumer and industrial System on Chips, and poses unique challenges to program (thermal, power, shared resources etc.). The goal of this work is to identify several key aspects in a compute framework (basic design, execution models, data communication, fault tolerance and latency), which handle the integrated GPU limitations and provide a proper abstraction.

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

2020 19th RoEduNet Conference: Networking in Education and Research (RoEduNet)

自引率

0.00%

发文量