Research progress in architecture and application of RRAM with computing-in-memory

Abstract

The development of new technologies has led to an explosion of data, while the computation ability of traditional computers is approaching its upper limit.

Resistive Memory Based on Graphene Oxide Prepared by Multi-Step Spin-Coating

“…There is an urgent need for non-volatile memory with a larger memory capacity, lower power consumption, and long-life span with the arrival of the Big Data age [1] . Many people take an interest in Resistive Random Access Memories (RRAM), which have benefits such as a simple structure, a rapid switching speed, and compatibility with conventional integrated circuit processes [2] .…”

Understanding the Reversible Transition of Unipolar and Bipolar Resistive Switching Characteristics in Solution-Derived Nanocrystalline Au–Co3O4 Thin-Film Memristors

“…Wang et al fabricated a flexible HfAlO x -based memristor, suggesting that their memristor is capable of simulating brain-inspired neuromorphic computing behaviors, such as long-term plasticity and long-term depression . The memristors for memory applications are called resistive random access memory (RRAM) . RRAM devices with a sandwich-based structure are considered as a future promising competitor due to their unique advantages such as low-power consumption, fast speed of reading/writing process, and compatibility with the semiconductor fabrication process. , …”

“…6 The memristors for memory applications are called resistive random access memory (RRAM). 7 RRAM devices with a sandwich-based structure are considered as a future promising competitor due to their unique advantages such as low-power consumption, fast speed of reading/writing process, and compatibility with the semiconductor fabrication process. 8,9 RS effect can be classified into two types, unipolar and bipolar switching behaviors, based on the characteristics of voltage polarity.…”

Perovskite material-based memristors for applications in information processing and artificial intelligence

“…In addition, the RRAM is automatically updated by the calculation results without additional delay to transmit and write the information. 7 The integration of memory and processing makes it possible for a new generation of computers to overcome limitations of the traditional von Neumann architecture. This integration has been represented simply in Fig.…”