Zhao Dongyan, Wang Yubo, Shao Jin, Chen Yanning, Guo Zhiwang, Pan Cheng, Dong Guangzhi, Zhou Min, Wu Fengxia, Wang Wenhe, Zhou Keji, Xue Xiaoyong
State Grid Key Laboratory of Power Industrial Chip Design and Analysis Technology, Beijing Smart-Chip Microelectronics Technology Co., Ltd., Beijing 100192, China.
Beijing Chip Identification Technology Co., Ltd., Beijing 100192, China.
Micromachines (Basel). 2022 May 2;13(5):731. doi: 10.3390/mi13050731.
In recent years, compute-in-memory (CIM) has been extensively studied to improve the energy efficiency of computing by reducing data movement. At present, CIM is frequently used in data-intensive computing. Data-intensive computing applications, such as all kinds of neural networks (NNs) in machine learning (ML), are regarded as 'soft' computing tasks. The 'soft' computing tasks are computations that can tolerate low computing precision with little accuracy degradation. However, 'hard' tasks aimed at numerical computations require high-precision computing and are also accompanied by energy efficiency problems. Numerical computations exist in lots of applications, including partial differential equations (PDEs) and large-scale matrix multiplication. Therefore, it is necessary to study CIM for numerical computations. This article reviews the recent developments of CIM for numerical computations. The different kinds of numerical methods solving partial differential equations and the transformation of matrixes are deduced in detail. This paper also discusses the iterative computation of a large-scale matrix, which tremendously affects the efficiency of numerical computations. The working procedure of the ReRAM-based partial differential equation solver is emphatically introduced. Moreover, other PDEs solvers, and other research about CIM for numerical computations, are also summarized. Finally, prospects and the future of CIM for numerical computations with high accuracy are discussed.
近年来，为了通过减少数据移动来提高计算的能源效率，内存计算（CIM）受到了广泛研究。目前，CIM常用于数据密集型计算。数据密集型计算应用，如机器学习（ML）中的各种神经网络（NN），被视为“软”计算任务。“软”计算任务是指那些能够容忍低计算精度且精度下降很小的计算。然而，针对数值计算的“硬”任务需要高精度计算，并且还伴随着能源效率问题。数值计算存在于许多应用中，包括偏微分方程（PDE）和大规模矩阵乘法。因此，有必要研究用于数值计算的CIM。本文综述了用于数值计算的CIM的最新进展。详细推导了求解偏微分方程的不同数值方法和矩阵变换。本文还讨论了大规模矩阵的迭代计算，这对数值计算的效率有极大影响。着重介绍了基于电阻式随机存取存储器（ReRAM）的偏微分方程求解器的工作过程。此外，还总结了其他偏微分方程求解器以及关于用于数值计算的CIM的其他研究。最后，讨论了高精度数值计算的CIM的前景和未来。
