MP-Queue:嵌入式流媒体平台的高效通信库

A. D. Torre, M. Ruggiero, L. Benini
{"title":"MP-Queue:嵌入式流媒体平台的高效通信库","authors":"A. D. Torre, M. Ruggiero, L. Benini","doi":"10.1109/ESTMED.2007.4375813","DOIUrl":null,"url":null,"abstract":"In this paper we present MP-queue, a flexible and efficient queue-based communication library for MPSoCs. Our library is suitable for a wide range of hardware platforms and its configuration space is explored across a wide number of dimensions. We introduce an upper-bound evaluation metric to compare the efficiency of the library against an ideal point-to-point data transfer. We can thus quantitatively assess the overhead introduced by the synchronization protocol and by shared bus contention. We discuss source-level optimizations introduced in the library that enable aggressive compiler optimizations, without compromising code portability. A significant speedup is achieved w.r.t a non-optimized library (15% for small-size messages), while communication efficiency rises up to 90% for large messages.","PeriodicalId":428196,"journal":{"name":"2007 IEEE/ACM/IFIP Workshop on Embedded Systems for Real-Time Multimedia","volume":"17 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2007-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":"{\"title\":\"MP-Queue: an Efficient Communication Library for Embedded Streaming Multimedia Platforms\",\"authors\":\"A. D. Torre, M. Ruggiero, L. Benini\",\"doi\":\"10.1109/ESTMED.2007.4375813\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper we present MP-queue, a flexible and efficient queue-based communication library for MPSoCs. Our library is suitable for a wide range of hardware platforms and its configuration space is explored across a wide number of dimensions. We introduce an upper-bound evaluation metric to compare the efficiency of the library against an ideal point-to-point data transfer. We can thus quantitatively assess the overhead introduced by the synchronization protocol and by shared bus contention. We discuss source-level optimizations introduced in the library that enable aggressive compiler optimizations, without compromising code portability. A significant speedup is achieved w.r.t a non-optimized library (15% for small-size messages), while communication efficiency rises up to 90% for large messages.\",\"PeriodicalId\":428196,\"journal\":{\"name\":\"2007 IEEE/ACM/IFIP Workshop on Embedded Systems for Real-Time Multimedia\",\"volume\":\"17 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2007-11-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"8\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2007 IEEE/ACM/IFIP Workshop on Embedded Systems for Real-Time Multimedia\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ESTMED.2007.4375813\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2007 IEEE/ACM/IFIP Workshop on Embedded Systems for Real-Time Multimedia","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ESTMED.2007.4375813","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8

摘要

本文提出了一种灵活高效的基于队列的mpsoc通信库MP-queue。我们的库适用于各种硬件平台,其配置空间在许多维度上进行了探索。我们引入了一个上界评估指标来比较库与理想点对点数据传输的效率。因此,我们可以定量地评估由同步协议和共享总线争用引入的开销。我们将讨论库中引入的源代码级优化,这些优化支持积极的编译器优化,而不会影响代码的可移植性。在使用非优化库的情况下实现了显著的加速(对于小型消息为15%),而对于大型消息,通信效率可提高到90%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
MP-Queue: an Efficient Communication Library for Embedded Streaming Multimedia Platforms
In this paper we present MP-queue, a flexible and efficient queue-based communication library for MPSoCs. Our library is suitable for a wide range of hardware platforms and its configuration space is explored across a wide number of dimensions. We introduce an upper-bound evaluation metric to compare the efficiency of the library against an ideal point-to-point data transfer. We can thus quantitatively assess the overhead introduced by the synchronization protocol and by shared bus contention. We discuss source-level optimizations introduced in the library that enable aggressive compiler optimizations, without compromising code portability. A significant speedup is achieved w.r.t a non-optimized library (15% for small-size messages), while communication efficiency rises up to 90% for large messages.
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:481959085
Book学术官方微信