当前位置:首页 > 报告详情

DC25_SLIDES_Track1_StatisticalModelingOfSystemPowerIntegrity_Sharma_V2.pdf

上传人: S** 编号:1240864 2026-05-16 26页 2.40MB

1、Information Classification:GeneralPublicWelcome to ConferenceJanuary 2830,2025Santa Clara Convention Center1ExpoJanuary 2930,2025 Information Classification:GeneralPublicStatistical Modeling of System Power Integrity in Adaptive Embedded SoC for Artificial Intelligence(AI)Computing Ajay Kumar Sharma

2、,(AMD)Thomas To,(AMD)Ajay Kumar Sharma(AMD),Thomas To(AMD)Susmita Mutsuddy(AMD)2Information Classification:GeneralPublicImageImage SPEAKERSAjay Kumar SharmaPMTS,AMDAjay.Kumar.Sis a Principal Engineer at AMD.He is involved in Signal integrity characterization including SSN,PDN&Memory system character

3、ization.He joined Xilinx in July 2011.Ajay completed his M.Tech.from Thapar Institute of Engineering&Technology in 2005.Before joining Xilinx,Ajay was leading a Signal Integrity&Package Design team at Freescale Semiconductor India Pvt.Ltd.Thomas ToSenior Fellow,AMDThomas.T,AMDis a Senior Fellow in A

4、MD focusing on System IO Platform Architecture Development.Prior to AMDs acquisition of Xilinx,Thomas was a Distinguished Engineer in System Memory Signal Integrity group.Before joining Xilinx,Thomas was with NVIDIA Advanced Technology Group focused on highspeed circuits&system channel designs.Befor

5、e NVIDIA,Thomas worked for Intel for more than 16 years covered and led many different types of system memory IO development.Thomas received his PhD degree in Electrical Engineering from the Ohio State University&he is the inventor of over 40 patents in the fields of mixed signal IO circuits and sys

6、tem memory configurations as well as high speed clocking for highspeed memory designs.3Information Classification:GeneralPublic Overview of Embedded SoC for AI Computing Growth Trend and Challenges for Power Integrity Design Overview of Adaptive Embedded SoC Statistical Power Noise Analysis Set Up S

word格式文档无特别注明外均可编辑修改,预览文件经过压缩,下载原文更清晰!
三个皮匠报告文库所有资源均是客户上传分享,仅供网友学习交流,未经上传用户书面授权,请勿作商用。
1. **会议与主题**:2025年1月28-30日圣克拉拉会议中心,主题为“自适应嵌入式SoC的AI计算系统电源完整性统计建模”。 2. **核心方法**:通过活动检测器(Activity Detector)收集计算单元(CE)的电流激励(α_Iij)和时间间隙(Δt_ij)数据,采用多元统计分析优化电源分配网络(PDN)设计。 3. **关键数据**:高利用率场景下电压跌落比低利用率场景高42mV,活动因子与噪声实测结果强相关。 4. **验证方式**:结合片上环形振荡器频率变化和直接C4探针测量,验证活动签名与电压噪声的关联性。 5. **结论**:基于应用使用统计的PDN设计优化方法可有效提升电源完整性,适用于多样化AI计算平台。
**AI芯片功耗优化?** **统计模型如何提升PDN设计?** **活动检测器如何量化噪声?**
客服
商务合作
小程序
服务号
折叠