Exploring the Feasibility of Using 3-D XPoint as an In-Memory Computing Accelerator
Exploring the Feasibility of Using 3-D XPoint as an In-Memory Computing Accelerator
复制标题
DOI：
10.1109/jxcdc.2021.3112238
复制
发表时间：
2021-06
影响因子：
2.4
通讯作者：
Masoud Zabihi;Salonik Resch;Husrev Cilasun;Z. Chowdhury;Zhengyang Zhao;Ulya R. Karpuzcu;Jianping Wang;S. Sapatnekar
中科院分区：
文献类型：
--
作者：
Masoud Zabihi;Salonik Resch;Husrev Cilasun;Z. Chowdhury;Zhengyang Zhao;Ulya R. Karpuzcu;Jianping Wang;S. Sapatnekar
本文描述了3-D XPoint内存阵列如何用作内存中的计算加速器。我们首先证明了阈值矩阵向量乘法(TMVM)是包括机器学习(ML)在内的许多应用中的基本计算核心，可以在三维XPoint阵列内实现，而不需要数据离开阵列进行处理。利用TMVM的实现，我们讨论了一个二进制神经推理机的实现。我们讨论了核心概念的应用来解决诸如系统可伸缩性(我们连接多个3-D XPoint阵列)和电源完整性(其中我们分析了金属线对噪声裕度的寄生影响)等问题。为了在实现过程中确保三维XPoint阵列的电源完整性，我们仔细分析了金属线的寄生效应对实现精度的影响。我们量化了寄生对限制三维XPoint阵列的大小和配置的影响，并估计了三维XPoint子阵列的最大可接受大小。
This article describes how 3-D XPoint memory arrays can be used as in-memory computing accelerators. We first show that thresholded matrix-vector multiplication (TMVM), the fundamental computational kernel in many applications including machine learning (ML), can be implemented within a 3-D XPoint array without requiring data to leave the array for processing. Using the implementation of TMVM, we then discuss the implementation of a binary neural inference engine. We discuss the application of the core concept to address issues such as system scalability, where we connect multiple 3-D XPoint arrays, and power integrity, where we analyze the parasitic effects of metal lines on noise margins. To assure power integrity within the 3-D XPoint array during this implementation, we carefully analyze the parasitic effects of metal lines on the accuracy of the implementations. We quantify the impact of parasitics on limiting the size and configuration of a 3-D XPoint array, and estimate the maximum acceptable size of a 3-D XPoint subarray.