S. Kumar, M. Santhanalakshmi, R. Navaneethakrishnan
{"title":"用于节能计算应用的内容可寻址存储器","authors":"S. Kumar, M. Santhanalakshmi, R. Navaneethakrishnan","doi":"10.58414/scientifictemper.2023.14.2.30","DOIUrl":null,"url":null,"abstract":"Content Addressable Memory (CAM) also known as associate memory isa special kind of semiconductor memory device that works differently from conventional Random Access Memory (RAM). A Content Addressable Memory is a memory unit that matches content over a single clock rather than using addresses. Its inherent parallel search mechanism makes it more advantageous than RAM in terms of speed of search operation. Designers aim to reduce two design characteristics: increasing silicon size and power consumption. As the need for CAM increases, the problem of power consumption also increases. Recent research on CAM is concentrated around diminishing power utilization without forfeiting speed or area. The main reason for the high-power consumption in conventional CAM architecture is devoid of control over the voltage on the Match Line recharge and Search Line precharge. A novel CAM architecture is proposed by removing the necessity of the search line recharge and also by introducing a transistor with gate connected to ML_Eval input that act as a control over the search operation. An Extra transistor with gate connected to Mask_Bar decides whether the circuit can be operated as Ternary Content Addressable Memory (TCAM) or Binary Content Addressable Memory (Bi-CAM). This CAM Architecture is found to be power efficient up to 50% due to the control over recharged voltage on ML. It is also inferred that the delay associated with the search operation can be reduced to a certain extent. The proposed CAM architecture is simulated using Cadence Virtuoso IC 6.1.6 in General Process Design Kit (GPDK) with90nm technology.","PeriodicalId":443629,"journal":{"name":"THE SCIENTIFIC TEMPER","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Content addressable memory for energy efficient computing applications\",\"authors\":\"S. Kumar, M. Santhanalakshmi, R. Navaneethakrishnan\",\"doi\":\"10.58414/scientifictemper.2023.14.2.30\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Content Addressable Memory (CAM) also known as associate memory isa special kind of semiconductor memory device that works differently from conventional Random Access Memory (RAM). A Content Addressable Memory is a memory unit that matches content over a single clock rather than using addresses. Its inherent parallel search mechanism makes it more advantageous than RAM in terms of speed of search operation. Designers aim to reduce two design characteristics: increasing silicon size and power consumption. As the need for CAM increases, the problem of power consumption also increases. Recent research on CAM is concentrated around diminishing power utilization without forfeiting speed or area. The main reason for the high-power consumption in conventional CAM architecture is devoid of control over the voltage on the Match Line recharge and Search Line precharge. A novel CAM architecture is proposed by removing the necessity of the search line recharge and also by introducing a transistor with gate connected to ML_Eval input that act as a control over the search operation. An Extra transistor with gate connected to Mask_Bar decides whether the circuit can be operated as Ternary Content Addressable Memory (TCAM) or Binary Content Addressable Memory (Bi-CAM). This CAM Architecture is found to be power efficient up to 50% due to the control over recharged voltage on ML. It is also inferred that the delay associated with the search operation can be reduced to a certain extent. The proposed CAM architecture is simulated using Cadence Virtuoso IC 6.1.6 in General Process Design Kit (GPDK) with90nm technology.\",\"PeriodicalId\":443629,\"journal\":{\"name\":\"THE SCIENTIFIC TEMPER\",\"volume\":\"8 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-06-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"THE SCIENTIFIC TEMPER\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.58414/scientifictemper.2023.14.2.30\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"THE SCIENTIFIC TEMPER","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.58414/scientifictemper.2023.14.2.30","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
内容可寻址存储器(CAM)也称为关联存储器,是一种特殊的半导体存储设备,其工作原理与传统的随机存取存储器(RAM)不同。内容可寻址存储器是一种通过单个时钟而不是使用地址来匹配内容的存储器单元。其固有的并行搜索机制使其在搜索操作速度上比RAM更具优势。设计人员的目标是减少两个设计特征:增加硅尺寸和功耗。随着对CAM需求的增加,功耗问题也在增加。最近对CAM的研究集中在降低功率利用率而不丧失速度或面积。传统CAM结构中功耗高的主要原因是缺乏对匹配线充电和搜索线预充电电压的控制。通过消除搜索线充电的必要性,并通过引入一个带门的晶体管连接到ML_Eval输入,作为对搜索操作的控制,提出了一种新的CAM架构。一个额外的晶体管与栅极连接到Mask_Bar决定电路是否可以作为三元内容可寻址存储器(TCAM)或二进制内容可寻址存储器(Bi-CAM)操作。由于对ML上充电电压的控制,该CAM架构的功耗效率高达50%,并且推断与搜索操作相关的延迟可以在一定程度上减少。采用通用工艺设计工具包(GPDK)中的Cadence Virtuoso IC 6.1.6,采用90纳米技术对所提出的CAM架构进行了仿真。
Content addressable memory for energy efficient computing applications
Content Addressable Memory (CAM) also known as associate memory isa special kind of semiconductor memory device that works differently from conventional Random Access Memory (RAM). A Content Addressable Memory is a memory unit that matches content over a single clock rather than using addresses. Its inherent parallel search mechanism makes it more advantageous than RAM in terms of speed of search operation. Designers aim to reduce two design characteristics: increasing silicon size and power consumption. As the need for CAM increases, the problem of power consumption also increases. Recent research on CAM is concentrated around diminishing power utilization without forfeiting speed or area. The main reason for the high-power consumption in conventional CAM architecture is devoid of control over the voltage on the Match Line recharge and Search Line precharge. A novel CAM architecture is proposed by removing the necessity of the search line recharge and also by introducing a transistor with gate connected to ML_Eval input that act as a control over the search operation. An Extra transistor with gate connected to Mask_Bar decides whether the circuit can be operated as Ternary Content Addressable Memory (TCAM) or Binary Content Addressable Memory (Bi-CAM). This CAM Architecture is found to be power efficient up to 50% due to the control over recharged voltage on ML. It is also inferred that the delay associated with the search operation can be reduced to a certain extent. The proposed CAM architecture is simulated using Cadence Virtuoso IC 6.1.6 in General Process Design Kit (GPDK) with90nm technology.