单周期异或(SCXOR)和有状态n位并行加法器的二维RRAM Crossbar实现

Bhanprakash Goswami, M. Suri
{"title":"单周期异或(SCXOR)和有状态n位并行加法器的二维RRAM Crossbar实现","authors":"Bhanprakash Goswami, M. Suri","doi":"10.1145/3565478.3572329","DOIUrl":null,"url":null,"abstract":"The motivation to find a solution to the Memory Wall problem led the research community to explore non-von-Neumann architectures. Compute In-Memory (CIM) architectures with emerging memory technologies are promising for minimizing data movement. In line with the CIM direction, several logical and arithmetic operations were demonstrated in the literature for maximizing operations per second per watt using the RRAM crossbar. In this work, we propose a novel way of realizing stateful XOR logic using RRAM crossbar memory. The proposed XOR design is free from the operand switching issue, and since it needs cells within a single column of the 2D crossbar, logic cascading with other logic gates in the same column is straightforward. Secondly, we offer a novel data shifting technique between two consecutive RRAM cell columns/rows of the crossbar. Leveraging the proposed methods, we realize a stateful n-bit parallel adder that takes n+3 computation cycles and 5n RRAM cells within the crossbar. With the proposed n-bit parallel adder design for n>3, we obtain a minimum 1.4X speedup compared to the literature without using an increased number of RRAM cells.","PeriodicalId":125590,"journal":{"name":"Proceedings of the 17th ACM International Symposium on Nanoscale Architectures","volume":"27 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Single Cycle XOR (SCXOR) and Stateful n-bit Parallel Adder Implementation Using 2D RRAM Crossbar\",\"authors\":\"Bhanprakash Goswami, M. Suri\",\"doi\":\"10.1145/3565478.3572329\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The motivation to find a solution to the Memory Wall problem led the research community to explore non-von-Neumann architectures. Compute In-Memory (CIM) architectures with emerging memory technologies are promising for minimizing data movement. In line with the CIM direction, several logical and arithmetic operations were demonstrated in the literature for maximizing operations per second per watt using the RRAM crossbar. In this work, we propose a novel way of realizing stateful XOR logic using RRAM crossbar memory. The proposed XOR design is free from the operand switching issue, and since it needs cells within a single column of the 2D crossbar, logic cascading with other logic gates in the same column is straightforward. Secondly, we offer a novel data shifting technique between two consecutive RRAM cell columns/rows of the crossbar. Leveraging the proposed methods, we realize a stateful n-bit parallel adder that takes n+3 computation cycles and 5n RRAM cells within the crossbar. With the proposed n-bit parallel adder design for n>3, we obtain a minimum 1.4X speedup compared to the literature without using an increased number of RRAM cells.\",\"PeriodicalId\":125590,\"journal\":{\"name\":\"Proceedings of the 17th ACM International Symposium on Nanoscale Architectures\",\"volume\":\"27 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-12-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 17th ACM International Symposium on Nanoscale Architectures\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3565478.3572329\",\"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 17th ACM International Symposium on Nanoscale Architectures","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3565478.3572329","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

为了找到内存墙问题的解决方案,研究团体开始探索非冯-诺伊曼架构。采用新兴内存技术的内存计算(CIM)体系结构有望最大限度地减少数据移动。与CIM方向一致,在文献中演示了几种逻辑和算术运算,以使用RRAM横杆最大化每秒每瓦的操作次数。在这项工作中,我们提出了一种使用RRAM交叉存储器实现有状态异或逻辑的新方法。所提出的异或设计不存在操作数切换问题,并且由于它需要2D交叉条的单列内的单元,因此与同一列中的其他逻辑门的逻辑级联是直接的。其次,在横杆的两个连续的RRAM单元列/行之间提供了一种新的数据转移技术。利用所提出的方法,我们实现了一个有状态的n位并行加法器,该加法器在交叉栏内需要n+3个计算周期和5n个RRAM单元。对于n>3的n位并行加法器设计,与文献相比,我们在不使用增加RRAM单元数量的情况下获得了至少1.4倍的加速。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Single Cycle XOR (SCXOR) and Stateful n-bit Parallel Adder Implementation Using 2D RRAM Crossbar
The motivation to find a solution to the Memory Wall problem led the research community to explore non-von-Neumann architectures. Compute In-Memory (CIM) architectures with emerging memory technologies are promising for minimizing data movement. In line with the CIM direction, several logical and arithmetic operations were demonstrated in the literature for maximizing operations per second per watt using the RRAM crossbar. In this work, we propose a novel way of realizing stateful XOR logic using RRAM crossbar memory. The proposed XOR design is free from the operand switching issue, and since it needs cells within a single column of the 2D crossbar, logic cascading with other logic gates in the same column is straightforward. Secondly, we offer a novel data shifting technique between two consecutive RRAM cell columns/rows of the crossbar. Leveraging the proposed methods, we realize a stateful n-bit parallel adder that takes n+3 computation cycles and 5n RRAM cells within the crossbar. With the proposed n-bit parallel adder design for n>3, we obtain a minimum 1.4X speedup compared to the literature without using an increased number of RRAM cells.
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