基于cam的大规模并行SIMD矩阵核模乘法的实现

Kyosuke Kageyama, Hajime Hamano, Ryogo Kayama, Tetsushi Koide, T. Kumaki
{"title":"基于cam的大规模并行SIMD矩阵核模乘法的实现","authors":"Kyosuke Kageyama, Hajime Hamano, Ryogo Kayama, Tetsushi Koide, T. Kumaki","doi":"10.1109/ITC-CSCC58803.2023.10212453","DOIUrl":null,"url":null,"abstract":"Recently, semiconductor technology has been growing rapidly, and mobile devices are required to have digital convergence. In this paper, a CAM-based massive-parallel SIMD matrix core (CAMX) is proposed as a mobile device accelerator for high-performance, programmability, and versatility. The CAMX can process repeated arithmetic and table-lookup coding operations in parallel. Mobile devices need to keep secure communication on the Internet, and the cipher techniques are wide spread for secure communication like the Advanced Encryption Standard (AES), Rivest-Shamir-Adleman (RSA), and so on. In particular, RSA encryption needs to calculate modulo multiplication. In this paper, the CAMX executes modulo multiplication for secure communication. This processing involves multiplication and division processing. This processing is implemented in the CAMX, and the CAMX can execute modulo multiplication at 1,104 clock cycles in parallel. In addition, the clock cycles of the CAMX and an ARM core are compared. From the result, the CAMX can reduce clock cycles by approximately 86 % for 1,024 parallel processing data relative to the ARM core with NEON.","PeriodicalId":220939,"journal":{"name":"2023 International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC)","volume":"292 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Implementation of Modulo Multiplication with CAM-Based Massive-Parallel SIMD Matrix Core\",\"authors\":\"Kyosuke Kageyama, Hajime Hamano, Ryogo Kayama, Tetsushi Koide, T. Kumaki\",\"doi\":\"10.1109/ITC-CSCC58803.2023.10212453\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Recently, semiconductor technology has been growing rapidly, and mobile devices are required to have digital convergence. In this paper, a CAM-based massive-parallel SIMD matrix core (CAMX) is proposed as a mobile device accelerator for high-performance, programmability, and versatility. The CAMX can process repeated arithmetic and table-lookup coding operations in parallel. Mobile devices need to keep secure communication on the Internet, and the cipher techniques are wide spread for secure communication like the Advanced Encryption Standard (AES), Rivest-Shamir-Adleman (RSA), and so on. In particular, RSA encryption needs to calculate modulo multiplication. In this paper, the CAMX executes modulo multiplication for secure communication. This processing involves multiplication and division processing. This processing is implemented in the CAMX, and the CAMX can execute modulo multiplication at 1,104 clock cycles in parallel. In addition, the clock cycles of the CAMX and an ARM core are compared. From the result, the CAMX can reduce clock cycles by approximately 86 % for 1,024 parallel processing data relative to the ARM core with NEON.\",\"PeriodicalId\":220939,\"journal\":{\"name\":\"2023 International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC)\",\"volume\":\"292 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-06-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ITC-CSCC58803.2023.10212453\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ITC-CSCC58803.2023.10212453","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

最近,半导体技术发展迅速,移动设备需要具有数字融合。本文提出了一种基于cam的大规模并行SIMD矩阵核心(CAMX)作为高性能、可编程性和多功能性的移动设备加速器。CAMX可以并行处理重复的算术和表查找编码操作。移动设备需要在Internet上进行安全通信,而用于安全通信的密码技术如高级加密标准AES (Advanced Encryption Standard)、RSA (Rivest-Shamir-Adleman)等被广泛采用。特别是RSA加密需要计算模乘法。在本文中,CAMX执行模乘法以实现安全通信。这个处理包括乘法和除法处理。该处理在CAMX中实现,CAMX可以并行执行1104时钟周期的模乘法。此外,还比较了CAMX和ARM内核的时钟周期。从结果来看,相对于带有NEON的ARM内核,CAMX可以在1024个并行处理数据时减少大约86%的时钟周期。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Implementation of Modulo Multiplication with CAM-Based Massive-Parallel SIMD Matrix Core
Recently, semiconductor technology has been growing rapidly, and mobile devices are required to have digital convergence. In this paper, a CAM-based massive-parallel SIMD matrix core (CAMX) is proposed as a mobile device accelerator for high-performance, programmability, and versatility. The CAMX can process repeated arithmetic and table-lookup coding operations in parallel. Mobile devices need to keep secure communication on the Internet, and the cipher techniques are wide spread for secure communication like the Advanced Encryption Standard (AES), Rivest-Shamir-Adleman (RSA), and so on. In particular, RSA encryption needs to calculate modulo multiplication. In this paper, the CAMX executes modulo multiplication for secure communication. This processing involves multiplication and division processing. This processing is implemented in the CAMX, and the CAMX can execute modulo multiplication at 1,104 clock cycles in parallel. In addition, the clock cycles of the CAMX and an ARM core are compared. From the result, the CAMX can reduce clock cycles by approximately 86 % for 1,024 parallel processing data relative to the ARM core with NEON.
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