Title: Review of in-memory stateful logic gates from the perspective of materials to circuit

Abstract

Combining Boolean logic and nonvolatile memory functions, memristor-based stateful logic gates can alleviate the data movement during the computing process to achieve futuristic in-memory computing. This may solve the problem of the von Neumann bottleneck in the current computing architecture. In this review, the recent developments in memristor-based stateful logic gates are discussed from the perspectives of materials to circuit. The stateful logic gates correspond to the in-memory logic gates, with the resistance as the physical entities of inputs and outputs. The review firstly summarizes the developments in stateful logic primitive gates reported in the past decade including the material classification, the circuit structure, the logic mapping and the trigger conditions. Then, the methods for allocating the logic gates onto the crossbar array are explained, which are for implementing the complex computing instances in the crossbar array. Finally, the merits and evaluations of using different stateful logic gates to construct the in-memory computing paradigm are discussed.