流驱动GPGPU编程结合文本和图形编程

Thomas Hoegg, G. Fiedler, C. Koehler, A. Kolb
{"title":"流驱动GPGPU编程结合文本和图形编程","authors":"Thomas Hoegg, G. Fiedler, C. Koehler, A. Kolb","doi":"10.1145/2883404.2883412","DOIUrl":null,"url":null,"abstract":"GPGPUs (General Purpose Computation on Graphics Processing Unit) have become the most important invention in the last years in computer graphics and the vision domain. Despite improvement of the two main programming platforms, CUDA (Compute Unified Device Architecture) and OpenCL (Open Computing Language), GPGPU programming and development is still a complex, time consuming and error-prone task. To overcome these problems for general software engineering, the graphical modeling language UML (Unified Modeling Language) was introduced and became the first choice for designing software systems. However, its generic design causes representations of algorithmic problem descriptions to be either limited or too complicated. We present GU-DSL, a novel domain-specific language (DSL), including novel modeling concepts (new activity-diagram node types and special language constructs), based on Eclipse Xtext and GMF, adopting and extending class- and activity-diagrams in a textual and graphical form. Furthermore, we present a C++ and OpenCL code generation framework in combination with a heterogeneous C++ GPGPU computing framework allowing for a smooth connection with our DSL and graphical editors.","PeriodicalId":185841,"journal":{"name":"Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores","volume":"191 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"Flow Driven GPGPU Programming combining Textual and Graphical Programming\",\"authors\":\"Thomas Hoegg, G. Fiedler, C. Koehler, A. Kolb\",\"doi\":\"10.1145/2883404.2883412\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"GPGPUs (General Purpose Computation on Graphics Processing Unit) have become the most important invention in the last years in computer graphics and the vision domain. Despite improvement of the two main programming platforms, CUDA (Compute Unified Device Architecture) and OpenCL (Open Computing Language), GPGPU programming and development is still a complex, time consuming and error-prone task. To overcome these problems for general software engineering, the graphical modeling language UML (Unified Modeling Language) was introduced and became the first choice for designing software systems. However, its generic design causes representations of algorithmic problem descriptions to be either limited or too complicated. We present GU-DSL, a novel domain-specific language (DSL), including novel modeling concepts (new activity-diagram node types and special language constructs), based on Eclipse Xtext and GMF, adopting and extending class- and activity-diagrams in a textual and graphical form. Furthermore, we present a C++ and OpenCL code generation framework in combination with a heterogeneous C++ GPGPU computing framework allowing for a smooth connection with our DSL and graphical editors.\",\"PeriodicalId\":185841,\"journal\":{\"name\":\"Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores\",\"volume\":\"191 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-03-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/2883404.2883412\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2883404.2883412","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

摘要

gpgpu(通用计算图形处理单元)已成为近年来计算机图形学和视觉领域最重要的发明。尽管两个主要的编程平台CUDA(计算统一设备架构)和OpenCL(开放计算语言)得到了改进,但GPGPU的编程和开发仍然是一个复杂、耗时和易出错的任务。为了克服这些问题,通用软件工程引入了图形化建模语言UML(统一建模语言),并成为设计软件系统的首选语言。然而,它的通用设计导致算法问题描述的表示要么受限,要么过于复杂。我们提出了一种新的领域特定语言(DSL),包括基于Eclipse Xtext和GMF的新的建模概念(新的活动图节点类型和特殊的语言结构),以文本和图形形式采用和扩展类图和活动图。此外,我们提出了一个c++和OpenCL代码生成框架,结合了一个异构的c++ GPGPU计算框架,允许与我们的DSL和图形编辑器平滑连接。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Flow Driven GPGPU Programming combining Textual and Graphical Programming
GPGPUs (General Purpose Computation on Graphics Processing Unit) have become the most important invention in the last years in computer graphics and the vision domain. Despite improvement of the two main programming platforms, CUDA (Compute Unified Device Architecture) and OpenCL (Open Computing Language), GPGPU programming and development is still a complex, time consuming and error-prone task. To overcome these problems for general software engineering, the graphical modeling language UML (Unified Modeling Language) was introduced and became the first choice for designing software systems. However, its generic design causes representations of algorithmic problem descriptions to be either limited or too complicated. We present GU-DSL, a novel domain-specific language (DSL), including novel modeling concepts (new activity-diagram node types and special language constructs), based on Eclipse Xtext and GMF, adopting and extending class- and activity-diagrams in a textual and graphical form. Furthermore, we present a C++ and OpenCL code generation framework in combination with a heterogeneous C++ GPGPU computing framework allowing for a smooth connection with our DSL and graphical editors.
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
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信