Photonic Fabric Platform for AI Accelerators

TL;DR

This paper introduces the Photonic FabricTM and PFA, a photonic-enabled switch and memory system that significantly enhances AI accelerator performance and energy efficiency by enabling scalable, high-bandwidth shared memory for distributed training and inference.

Contribution

The paper presents a novel photonic fabric platform and appliance that overcome fixed memory-to-compute ratios, enabling scalable, high-bandwidth shared memory for AI accelerators.

Findings

Up to 3.66x throughput improvement in LLM inference

Up to 7.04x throughput improvement in large models

60-90% energy savings in data movement

Abstract

This paper presents the Photonic FabricTM and the Photonic Fabric ApplianceTM (PFA), a photonic-enabled switch and memory subsystem that delivers low latency, high bandwidth, and low per-bit energy. By integrating high-bandwidth HBM3E memory, an on-module photonic switch, and external DDR5 in a 2.5D electro-optical system-in-package, the PFA offers up to 32 TB of shared memory alongside 115 Tbps of all-to-all digital switching. The Photonic FabricTM enables distributed AI training and inference to execute parallelism strategies more efficiently. The Photonic Fabric removes the silicon beachfront constraint that limits the fixed memory-to-compute ratio observed in virtually all current XPU accelerator designs. Replacing a local HBM stack on an XPU with a chiplet that connects to the Photonic Fabric increases its memory capacity and correspondingly its memory bandwidth by offering a flexible path to scaling well beyond the limitations of on-package HBM alone. We introduce CelestiSim, a lightweight analytical simulator validated on NVIDIA H100 and H200 systems. It is used to evaluate the performance of LLM reference and energy savings on PFA, without any significant change to the GPU core design. With the PFA, the simulation results show that up to 3.66x throughput and 1.40x latency improvements in LLM inference at 405B parameters, up to 7.04x throughput and 1.41x latency improvements at 1T parameters, and 60-90% energy savings in data movement for heavy collective operations in all LLM training scenarios. While these results are shown for NVIDIA GPUs, they can be applied similarly to other AI accelerator designs (XPUs) that share the same fundamental limitation of fixed memory to compute.