Mixed-precision in-memory computing

Abstract

As CMOS scaling reaches its technological limits, a radical departure from traditional von Neumann systems, which involve separate processing and memory units, is needed in order to significantly extend the performance of today's computers. In-memory computing is a promising approach in which nanoscale resistive memory devices, organized in a computational memory unit, are used for both processing and memory. However, to reach the numerical accuracy typically required for data analytics and scientific computing, limitations arising from device variability and non-ideal device characteristics need to be addressed. Here we introduce the concept of mixed-precision in-memory computing, which combines a von Neumann machine with a computational memory unit. In this hybrid system, the computational memory unit performs the bulk of a computational task, while the von Neumann machine implements a backward method to iteratively improve the accuracy of the solution. The system therefore benefits from both the high precision of digital computing and the energy/areal efficiency of in-memory computing. We experimentally demonstrate the efficacy of the approach by accurately solving systems of linear equations, in particular, a system of 5, 000 equations using 998, 752 phase-change memory devices.

Plasmonic nanogap enhanced phase-change devices with dual electrical-optical functionality

“…Importantly, the non-volatile nature our platform, provides an exciting outlook in the development of switchable and reconfigurable metadevices by means of optical or electrical stimuli; enabling novel approaches to switchable metamaterial-based optical components (37)(38)(39).We anticipate that a plethora of novel devices and platforms should arise in the coming years, which will capitalize on the bridge between the electrical and photonic domains that are demonstrated herein. These devices potentially herald true device-level integration of hybrid optoelectronic computing platforms with in-memory computing and multilevel data storage which is readily applicable to this work (40,41).…”

Memristor‐Based Biologically Plausible Memory Based on Discrete and Continuous Attractor Networks for Neuromorphic Systems

“…Under this principle, the VMM computing process directly happened in situ, thus avoiding the memory wall (von Neumann Bottleneck) caused by fetching data from memory. Especially in supervised learning, it can reduce the computational complexity of feed forward process and backpropagation from NP to P . Therefore, current studies mostly focus on classification and regression tasks to make use of this new computing mechanism as a complement to complementary metal oxide semiconductor (CMOS) circuits.…”

Resistive Memory‐Based In‐Memory Computing: From Device and Large‐Scale Integration System Perspectives

“…It is commonly believed that the following types of random or sequential access memories have great potentials: static random access memory (SRAM), [8][9][10] flash memory, [11,12] magnetic random access memory (MRAM), [13,14] racetrack memory, [15,16] phase change memory (PCM), [3,[17][18][19] and resistive random access memory (RRAM). It is commonly believed that the following types of random or sequential access memories have great potentials: static random access memory (SRAM), [8][9][10] flash memory, [11,12] magnetic random access memory (MRAM), [13,14] racetrack memory, [15,16] phase change memory (PCM), [3,[17][18][19] and resistive random access memory (RRAM).…”

“…[8][9][10] Besides persistent data storage, nonvolatile memory technologies generally have a higher density, i.e., <100 F 2 cell size, where F represents the technology feature size. [18] RRAM offers versatility, including high resistivity (MΩ order of magnitude), the support of 3D integration, stochastic programming, and multilevel cell (up to 6 bits). MRAM highlights fast write speed (in the same order as SRAM) and stochastic programming.…”