A 1.1V 16Gb DDR5 DRAM with Probabilistic-Aggressor

📋 论文概要

该论文提出了一款1.1V 16Gb DDR5 DRAM，通过概率性攻击者跟踪、多步预充电和核心偏置调制技术，解决了DRAM中由于行锤攻击导致的安全性和可靠性问题，并实现了刷新管理功能的增强。

💡 主要创新点

• 提出概率性攻击者跟踪（Probabilistic-Aggressor Tracking）机制，有效识别并缓解行锤攻击，降低刷新开销。
• 采用多步预充电（Multi-Step Precharge）技术，减少电压波动和噪声，提升数据保持可靠性。
• 引入核心偏置调制（Core-Bias Modulation），动态调整核心电压以增强抗干扰能力和能效。

🏷 关键词

📄 原文摘要

Hammer Tracking, a Multi-Step Precharge, and Core-Bias Modulation for Security and Reliability Enhancement Woongrae Kim, Chulmoon Jung, Seongnyuh Yoo, Duckhwa Hong, Jeongjin Hwang, Jungmin Yoon, Ohyong Jung, Joonwoo Choi, Sanga Hyun, Mankeun Kang, Sangho Lee, Dohong Kim, Sanghyun Ku, Donhyun Choi, Nogeun Joo, Sangwoo Yoon, Junseok Noh, Byeongyong Go, Cheolhoe Kim, Sunil Hwang, Mihyun Hwang, Seol-Min Yi, Hyungmin Kim, Sanghyuk Heo, Yeonsu Jang, Kyoungchul Jang, Shinho Chu, Yoonna Oh, Kwidong Kim, Junghyun Kim, Soohwan Kim, Jeongtae Hwang, Sangil Park, Junphyo Lee, Inchul Jeong, Joohwan Cho, Jonghwan Kim SK hynix Semiconductor, Icheon, Korea DRAM products have been recently adopted in a wide range of high-performance computing applications: such as in cloud computing, in big data systems, and IoT devices. This demand creates larger memory capacity requirements, thereby requiring aggressive DRAM technology node scaling to reduce the cost per bit [1,2]. However,

👥 作者与机构

Tracking, Refresh-Management Functionality, Per-Row