三维片上网络的低开销路由算法

Akram Ben Ahmed, B. Abderazek
{"title":"三维片上网络的低开销路由算法","authors":"Akram Ben Ahmed, B. Abderazek","doi":"10.1109/ICNC.2012.14","DOIUrl":null,"url":null,"abstract":"Despite the higher scalability and parallelism integration offered by Network-on-Chip (NoC) over the traditional shared-bus based systems, it is still not an ideal solution for future large scale Systems-on-Chip (SoCs), due to some limitations such as high power consumption and high cost communication. Recently, extending 2D-NoC to the third dimension (3D-NoC) has been proposed to deal with these problems. One of the most important design choices for 3D-NoC designs is the implementation of a fast and high throughput system with a reasonable hardware complexity. Moreover, adopting an efficient routing algorithm has become an important task since the flits' path selection has a direct impact on the overall system performance. In this paper, we propose a low overhead, and high throughput 3D-NoC routing algorithm named Look-ahead-XYZ (LA-XYZ). We implemented the proposed algorithm in our 3D-NoC deign, named 3D-OASIS-NoC, and we prototyped it on FPGA. We evaluated its performance using Transpose traffic pattern and two selected real applications (JPEG-encoder and Matrix-multiplication). Evaluation results show that the proposed algorithm reduces the communication latency by 29.22% and 35.12% and enhances the throughput with an average of 26% and 38.95% when compared to XYZ and Randomized Partially Minimal (RPM) routing algorithms respectively.","PeriodicalId":442973,"journal":{"name":"2012 Third International Conference on Networking and Computing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"29","resultStr":"{\"title\":\"Low-overhead Routing Algorithm for 3D Network-on-Chip\",\"authors\":\"Akram Ben Ahmed, B. Abderazek\",\"doi\":\"10.1109/ICNC.2012.14\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Despite the higher scalability and parallelism integration offered by Network-on-Chip (NoC) over the traditional shared-bus based systems, it is still not an ideal solution for future large scale Systems-on-Chip (SoCs), due to some limitations such as high power consumption and high cost communication. Recently, extending 2D-NoC to the third dimension (3D-NoC) has been proposed to deal with these problems. One of the most important design choices for 3D-NoC designs is the implementation of a fast and high throughput system with a reasonable hardware complexity. Moreover, adopting an efficient routing algorithm has become an important task since the flits' path selection has a direct impact on the overall system performance. In this paper, we propose a low overhead, and high throughput 3D-NoC routing algorithm named Look-ahead-XYZ (LA-XYZ). We implemented the proposed algorithm in our 3D-NoC deign, named 3D-OASIS-NoC, and we prototyped it on FPGA. We evaluated its performance using Transpose traffic pattern and two selected real applications (JPEG-encoder and Matrix-multiplication). Evaluation results show that the proposed algorithm reduces the communication latency by 29.22% and 35.12% and enhances the throughput with an average of 26% and 38.95% when compared to XYZ and Randomized Partially Minimal (RPM) routing algorithms respectively.\",\"PeriodicalId\":442973,\"journal\":{\"name\":\"2012 Third International Conference on Networking and Computing\",\"volume\":\"11 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2012-12-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"29\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2012 Third International Conference on Networking and Computing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICNC.2012.14\",\"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 Third International Conference on Networking and Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICNC.2012.14","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 29

摘要

尽管与传统的基于共享总线的系统相比,片上网络(NoC)提供了更高的可扩展性和并行集成度,但由于其高功耗和高通信成本等局限性,它仍然不是未来大规模片上系统(soc)的理想解决方案。最近,有人提出将2D-NoC扩展到三维(3D-NoC)来解决这些问题。3D-NoC设计最重要的设计选择之一是实现具有合理硬件复杂性的快速高吞吐量系统。此外,采用高效的路由算法已成为一项重要任务,因为flits的路径选择直接影响到系统的整体性能。在本文中,我们提出了一种低开销,高吞吐量的3D-NoC路由算法,称为前瞻性xyz (LA-XYZ)。我们在3D-NoC设计中实现了所提出的算法,命名为3D-OASIS-NoC,并在FPGA上进行了原型设计。我们使用转置流量模式和两个选定的实际应用(jpeg编码器和矩阵乘法)来评估其性能。评估结果表明,与XYZ和随机部分最小路由算法相比,该算法的通信延迟分别降低了29.22%和35.12%,吞吐量平均提高了26%和38.95%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Low-overhead Routing Algorithm for 3D Network-on-Chip
Despite the higher scalability and parallelism integration offered by Network-on-Chip (NoC) over the traditional shared-bus based systems, it is still not an ideal solution for future large scale Systems-on-Chip (SoCs), due to some limitations such as high power consumption and high cost communication. Recently, extending 2D-NoC to the third dimension (3D-NoC) has been proposed to deal with these problems. One of the most important design choices for 3D-NoC designs is the implementation of a fast and high throughput system with a reasonable hardware complexity. Moreover, adopting an efficient routing algorithm has become an important task since the flits' path selection has a direct impact on the overall system performance. In this paper, we propose a low overhead, and high throughput 3D-NoC routing algorithm named Look-ahead-XYZ (LA-XYZ). We implemented the proposed algorithm in our 3D-NoC deign, named 3D-OASIS-NoC, and we prototyped it on FPGA. We evaluated its performance using Transpose traffic pattern and two selected real applications (JPEG-encoder and Matrix-multiplication). Evaluation results show that the proposed algorithm reduces the communication latency by 29.22% and 35.12% and enhances the throughput with an average of 26% and 38.95% when compared to XYZ and Randomized Partially Minimal (RPM) routing algorithms respectively.
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