FlashMoE: Fast Distributed MoE in a Single Kernel

• URL: http://arxiv.org/abs/2506.04667v3
• Date: Sat, 08 Nov 2025 04:20:44 GMT
• Title: FlashMoE: Fast Distributed MoE in a Single Kernel
• Authors: Osayamen Jonathan Aimuyo, Byungsoo Oh, Rachee Singh,
• Abstract summary: FlashMoE is a fully GPU-resident MoE operator that fuses expert computation and inter-GPU communication into a single persistent GPU kernel.<n>We show that FlashMoE achieves up to 9x higher GPU utilization, 6x lower latency, 5.7x higher throughput, and 4x better overlap efficiency compared to state-of-the-art baselines.
• Score: 1.866526462692252
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The computational sparsity of Mixture-of-Experts (MoE) models enables sub-linear growth in compute cost as model size increases, thus offering a scalable path to training massive neural networks. However, existing implementations suffer from low GPU utilization, significant latency overhead, and a fundamental inability to leverage task locality, primarily due to CPU-managed scheduling, host-initiated communication, and frequent kernel launches. To overcome these limitations, we develop FlashMoE, a fully GPU-resident MoE operator that fuses expert computation and inter-GPU communication into a single persistent GPU kernel. FlashMoE enables fine-grained pipelining of dispatch, compute, and combine phases, eliminating launch overheads and reducing idle gaps. Unlike existing work, FlashMoE eliminates bulk-synchronous collectives for one-sided, device-initiated, inter-GPU (R)DMA transfers, thereby unlocking payload efficiency by eliminating bloated or redundant network payloads in sparsely activated layers. When evaluated on an 8-H100 GPU node with MoE models comprising up to 128 experts and 16K token sequences, FlashMoE achieves up to 9x higher GPU utilization, 6x lower latency, 5.7x higher throughput, and 4x better overlap efficiency compared to state-of-the-art baselines, despite using FP32, whereas the baselines use FP16. FlashMoE shows that principled GPU kernel-hardware co-design is key to unlocking the performance ceiling of large-scale distributed ML. We provide code at https://github.com/osayamenja/FlashMoE.

Related papers

• ZipMoE: Efficient On-Device MoE Serving via Lossless Compression and Cache-Affinity Scheduling [56.88966608455977]
  ZipMoE exploits the synergy between the hardware properties of edge devices and the statistical redundancy inherent to MoE parameters.<n>ZipMoE achieves up to $72.77%$ inference latency reduction and up to $6.76times$ higher throughput than the state-of-the-art systems.
  arXiv  Detail & Related papers  (2026-01-29T02:51:59Z)
• FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge Devices [0.0]
  Mixture-of-Experts (MoE) models have gained attention for efficiently scaling large language models.<n>MoE models are extremely large, their sparse activation enables inference to be performed by accessing only a fraction of the model at a time.<n>We propose FlashMoE, a system that offloads inactive experts to SSD, enabling efficient MoE inference under limited RAM.
  arXiv  Detail & Related papers  (2026-01-22T17:07:33Z)
• SonicMoE: Accelerating MoE with IO and Tile-aware Optimizations [54.303301888915406]
  Mixture of Experts (MoE) models have emerged as the de facto architecture for scaling up language models without significantly increasing the computational cost.<n>We propose a memory-efficient algorithm to compute the forward and backward passes of MoEs with minimal activation caching.<n>We also propose a novel "token rounding" method that minimizes the wasted compute due to padding in Grouped GEMM kernels.
  arXiv  Detail & Related papers  (2025-12-16T04:39:10Z)
• Eliminating Multi-GPU Performance Taxes: A Systems Approach to Efficient Distributed LLMs [61.953548065938385]
  We introduce the ''Three Taxes'' (Bulk Synchronous, Inter- Kernel Data Locality, and Kernel Launch Overhead) as an analytical framework.<n>We propose moving beyond the rigid BSP model to address key inefficiencies in distributed GPU execution.<n>We observe a 10-20% speedup in end-to-end latency over BSP-based approaches.
  arXiv  Detail & Related papers  (2025-11-04T01:15:44Z)
• BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity [66.94629945519125]
  We introduce a novel MoE architecture, BlockFFN, as well as its efficient training and deployment techniques.<n>Specifically, we use a router integrating ReLU activation and RMSNorm for differentiable and flexible routing.<n>Next, to promote both token-level sparsity (TLS) and chunk-level sparsity ( CLS), CLS-aware training objectives are designed, making BlockFFN more acceleration-friendly.
  arXiv  Detail & Related papers  (2025-07-11T17:28:56Z)
• Nexus:Proactive Intra-GPU Disaggregation of Prefill and Decode in LLM Serving [4.309392302169281]
  Engine-level prefill-decode (PD) disaggregation avoids interference but incurs higher hardware and coordination overhead.<n>PD achieves up to 2.2x higher throughput, 20x lower TTFT, and 2.5x lower TBT than vLLM; outperforms SG by up to 2x; and matches or exceeds disaggregated vLLM.
  arXiv  Detail & Related papers  (2025-07-09T07:27:18Z)
• FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs [2.9406946721643092]
  This paper introduces FlashMLA-ETAP, a novel framework that enhances MLA inference for the single-instance deployment scenario.<n>ETAP reconfigures attention computation through transposition to align the KV context length with the (M)-dimension in WGMMA operations.<n>FlashMLA-ETAP achieves a 2.78x speedup over FlashMLA at 64K sequence length (batch size 16), with 5.24x and 4.94x improvements over FlashAttention-3 and FlashInfer, respectively.
  arXiv  Detail & Related papers  (2025-05-13T17:45:34Z)
• FloE: On-the-Fly MoE Inference on Memory-constrained GPU [22.2581000412208]
  FloE is built on the insight that there exists substantial untapped redundancy within sparsely activated experts.<n>FloE achieves a 9.3x compression of parameters per expert in Mixtral-8x7B.<n>It enables deployment on a GPU with only 11GB VRAM, reducing the memory footprint by up to 8.5x.
  arXiv  Detail & Related papers  (2025-05-09T10:53:47Z)
• MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism [26.923312725688735]
  Mixture-of-Experts (MoE) showcases tremendous potential to scale large language models (LLMs) with enhanced performance and reduced computational complexity.<n>We present MegaScale-Infer, an efficient and cost-effective system for serving large-scale MoE models.
  arXiv  Detail & Related papers  (2025-04-03T04:20:44Z)
• MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUs [55.95879347182669]
  MoE architecture is renowned for its ability to increase model capacity without a proportional increase in inference cost. MoE-Lightning introduces a novel CPU-GPU-I/O pipelining schedule, CGOPipe, with paged weights to achieve high resource utilization. MoE-Lightning can achieve up to 10.3x higher throughput than state-of-the-art offloading-enabled LLM inference systems for Mixtral 8x7B on a single T4 GPU (16GB)
  arXiv  Detail & Related papers  (2024-11-18T01:06:12Z)
• FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision [14.426543629408984]
  Attention is the bottleneck for large language models and long-context applications. We develop three main techniques to speed up attention on GPUs. We demonstrate that our method, FlashAttention-3, achieves speedup on H100 GPU by 1.5-2.0$times$ with FP16 reaching up to 740 TFLOPs/s (75% utilization) and with FP8 reaching close to 1.2 PFLOPs/s.
  arXiv  Detail & Related papers  (2024-07-11T15:44:48Z)
• FLUX: Fast Software-based Communication Overlap On GPUs Through Kernel Fusion [9.5114389643299]
  This paper proposes a novel method, Flux, to significantly hide communication latencies with dependent computations for GPUs. Flux can potentially overlap up to 96% of communication given a fused kernel. Overall, it can achieve up to 1.24x speedups for training over Megatron-LM on a cluster of 128 GPU with various GPU generations and interconnects.
  arXiv  Detail & Related papers  (2024-06-11T00:17:39Z)
• Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity [12.663030430488922]
  We propose Flash-LLM for enabling low-cost and highly-efficient large generative model inference on high-performance Cores. At SpMM kernel level, Flash-LLM significantly outperforms the state-of-the-art library, i.e., Sputnik and SparTA by an average of 2.9x and 1.5x, respectively.
  arXiv  Detail & Related papers  (2023-09-19T03:20:02Z)
• FusionAI: Decentralized Training and Deploying LLMs with Massive Consumer-Level GPUs [57.12856172329322]
  We envision a decentralized system unlocking the potential vast untapped consumer-level GPU. This system faces critical challenges, including limited CPU and GPU memory, low network bandwidth, the variability of peer and device heterogeneity.
  arXiv  Detail & Related papers  (2023-09-03T13:27:56Z)
• Edge-MoE: Memory-Efficient Multi-Task Vision Transformer Architecture with Task-level Sparsity via Mixture-of-Experts [60.1586169973792]
  M$3$ViT is the latest multi-task ViT model that introduces mixture-of-experts (MoE) MoE achieves better accuracy and over 80% reduction computation but leaves challenges for efficient deployment on FPGA. Our work, dubbed Edge-MoE, solves the challenges to introduce the first end-to-end FPGA accelerator for multi-task ViT with a collection of architectural innovations.
  arXiv  Detail & Related papers  (2023-05-30T02:24:03Z)
• FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU [89.2451963569343]
  FlexGen is a generation engine for running large language model (LLM) inference on a single commodity GPU. When running OPT-175B on a single 16GB GPU, FlexGen achieves significantly higher throughput compared to state-of-the-art offloading systems. On the HELM benchmark, FlexGen can benchmark a 30B model with a 16GB GPU on 7 representative sub-scenarios in 21 hours.
  arXiv  Detail & Related papers  (2023-03-13T05:19:28Z)
• VersaGNN: a Versatile accelerator for Graph neural networks [81.1667080640009]
  We propose textitVersaGNN, an ultra-efficient, systolic-array-based versatile hardware accelerator. textitVersaGNN achieves on average 3712$times$ speedup with 1301.25$times$ energy reduction on CPU, and 35.4$times$ speedup with 17.66$times$ energy reduction on GPU.
  arXiv  Detail & Related papers  (2021-05-04T04:10:48Z)
• DistGNN: Scalable Distributed Training for Large-Scale Graph Neural Networks [58.48833325238537]
  Full-batch training on Graph Neural Networks (GNN) to learn the structure of large graphs is a critical problem that needs to scale to hundreds of compute nodes to be feasible. In this paper, we presentGNN that optimize the well-known Deep Graph Library (DGL) for full-batch training on CPU clusters. Our results on four common GNN benchmark datasets show up to 3.7x speed-up using a single CPU socket and up to 97x speed-up using 128 CPU sockets.
  arXiv  Detail & Related papers  (2021-04-14T08:46:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.