Photonic Fabric Platform for AI Accelerators
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 12 of its 29 references that are in the map. First posted on arXiv 18 Jul 2025.
Builds on
12 of 29 references.
References · 12
- 2017Attention Is All You Needcited by 60,361
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,673
- 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,868
- 2021Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LMcited by 632
- 2022Reducing Activation Recomputation in Large Transformer Modelscited by 302
- 2024Vidur: A Large-Scale Simulation Framework For LLM Inferencecited by 117
- 2021Software-Hardware Co-design for Fast and Scalable Training of Deep Learning Recommendation Modelscited by 78
- 2024LLMServingSim: A HW/SW Co-Simulation Infrastructure for LLM Inference Serving at Scalecited by 31
- 2023vTrain: A Simulation Framework for Evaluating Cost-effective and Compute-optimal Large Language Model Trainingcited by 23
- 2022DeepFlow: A Cross-Stack Pathfinding Framework for Distributed AI Systemscited by 6
- 2024ISO: Overlap of Computation and Communication within Seqenence For LLM Inferencecited by 3
Built on by
Not cited yet.