Mixed Precision Photonic Computing with 3D Electronic-Photonic Integrated Circuits

TL;DR

This paper introduces a novel 3D integrated photonic-electronic computing system using phase-change materials and AlGaAs-CMOS technology, achieving high precision, scalability, and parallelism for efficient scientific computations.

Contribution

It presents a new architecture for mixed-precision photonic in-memory computing with hierarchical scaling, high Q-factor resonators, and general matrix multiplication capabilities.

Findings

Achieves over 12-bit precision in in-memory computing.

Supports scalable matrix multiplication across multiple wavelengths.

Enables high-dimensional PDE problem solving in constant time.

Abstract

We propose advancing photonic in-memory computing through three-dimensional photonic-electronic integrated circuits using phase-change materials (PCM) and AlGaAs-CMOS technology. These circuits offer high precision (greater than 12 bits), scalability (greater than 1024 by 1024), and massive parallelism (greater than 1 million operations) across the wavelength, spatial, and temporal domains at ultra-low power (less than 1 watt per PetaOPS). Monolithically integrated hybrid PCM-AlGaAs memory resonators handle coarse-precision iterations (greater than 5-bit most significant bit precision) through reversible PCM phase transitions. Electro-optic memristive tuning enables fine-precision updates (greater than 8-bit least significant bit precision), resulting in over 12-bit precision for in-memory computing. The use of low-loss PCM (less than 0.01 dB per cm) and electro-optical tuning yields memristive optical resonators with high Q-factors (greater than 1 million), low insertion loss, and low tuning power. A W by W photonic tensor core composed of PCM-AlGaAs memresonators performs general matrix multiplication (GEMM) across W wavelengths from optical frequency combs, with minimal crosstalk and loss. Hierarchical scaling in the wavelength domain (K) and spatial domain (L) enables this system to address high-dimensional (N) scientific partial differential equation (PDE) problems in a single constant-time operation, compared to the conventional quadratic-time (N squared) computational complexity.