arXiv:2609.13585v1 Announce Type: cross
Abstract: Communication has become a bottleneck in distributed training and inference of large models. Overlapping communication with computation at the granul...
By Ziming Mao, Yihan Zhang, Shawn Wei Chew, Shuang Ma, Costin Raiciu, Yang Zhou, Scott Shenker, Ion Stoica
arXiv:2607. 28633v1 Announce Type: cross Abstract: Disaggregated LLM inference creates a datacenter networking problem that no existing system solves correctly.
By Sanjeev Rao Ganjihal
arXiv:2610.07098v1 Announce Type: cross
Abstract: Large transformer-based models increasingly depend on multi-GPU execution, which requires frequent collective communication among GPUs. Existing comm...
By Keyvan Dadashzadeh, Yuehong Zhou, Minyu Cui, Miquel Pericas
The paper introduces NVLinkOneSided, a one‑sided expert‑parallel all‑to‑all communication scheme for Mixture‑of‑Experts inference on GPUs connected via NVLink. By pushing tokens directly into target GPU buffers without metadata exchange and using grouped GEMMs to read them in place, the method reduces fixed‑cost latency, especially at small batch sizes. Two synchronization protocols—Fence for high throughput and Sentinel for low latency—enable the approach, which achieves 1.2–3.0× faster dispatch and combine than DeepEP and up to 2.4× faster MoE layers on Blackwell GPUs, outperforming DeepGEMM MegaMoE up to 128 tokens per GPU. The implementation is modular, agnostic to activation quantization, and easily integrable into inference frameworks.
By Bo Li, Chulian Zhang, Or Biri, Anthony Chang, Anton Korzh, Noam Duvdevani, XingFei Xi, Daniel Stokes, Zhenhuan Chen, Ludwig Schneider, Artem Y. Polyakov, Julien Demouth, Chandler Zhou
Scepsy is a serving system designed to efficiently schedule arbitrary multi‑LLM agentic workflows on GPU clusters. It leverages the observation that each LLM’s share of execution time remains relatively stable across requests, profiling LLMs under various parallelism levels to build an Aggregate LLM Pipeline that predicts throughput and latency. Using this predictor, Scepsy searches for optimal GPU allocations—balancing fractional GPU shares, tensor parallelism, and replica counts—and then heuristically places them on the cluster to reduce fragmentation and honor network topology, achieving up to 2.5× higher throughput and 1.0–3.3× lower latency compared to baseline approaches.
By Otto White, Marcel Wagenl\"ander, Britannio Jarrett, Xijin Zhao, Yanda Tao, Pedro Silvestre, Guo Li, Huanzhou Zhu, Llu\'is Vilanova, Peter Pietzuch
Expert parallelism (EP) shards the experts of a Mixture-of-Experts (MoE) layer across GPUs, so every MoE layer requires two all-to-all communications, namely dispatch and combine. Compared with traini...
arXiv:2504. 08791v3 Announce Type: replace-cross Abstract: On-device inference offers privacy, offline use, and instant response, but consumer hardware restricts large language models (LLMs) to low throughput and capability.
By Zonghang Li, Tao Li, Wenjiao Feng, Rongxing Xiao, Jianshu She, Hong Huang, Mohsen Guizani, Hongfang Yu, Qirong Ho, Wei Xiang, Xue Liu
arXiv:2512. 10236v2 Announce Type: replace-cross Abstract: Modern ML workloads demand distributing training and inference across multiple GPUs.
By Shagnik Pal, Shaizeen Aga, Suchita Pati, Mahzabeen Islam, Lizy K. John
arXiv:2607. 23264v1 Announce Type: cross Abstract: Fine-grained, device-initiated communication lets persistent GPU kernels in distributed diffusion transformer (DiT) inference issue remote stores and overlap data movement with Tensor Core computation.
By Jianwen Xian, Zhiyuan Xu, Yuchen Li, Ziliang Lai, Kang He, Zhen Huang, Aichen Feng, Jinyan Chen, Yilin Zhang, Qinqin Chen, Chengru Song
arXiv:2511. 04791v2 Announce Type: replace Abstract: Modern LLM serving systems must sustain high throughput while meeting strict latency SLOs across two distinct inference phases: compute-intensive prefill and memory-bound decode phases.
By Lei Gao, Chaoyi Jiang, Hossein Entezari Zarch, Daniel Wong, Mark Hill, Murali Annavaram
The paper investigates how to optimally partition streaming inference models into megakernels and conventional graph components. It introduces a framework that predicts the best megakernel composition without full end‑to‑end runs, showing that multiple megakernels or a fully conventional graph often outperform a single megakernel across various models and batch sizes. The study defines four model categories based on megakernel stability and demonstrates significant latency and throughput gains on an H100 NVL GPU using a three‑megakernel design.
By Aditya Chatterjee, Anand Joseph X M, Sanjay Singh Chauhan
arXiv:2607. 19539v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures increase model capacity without proportionally increasing computation cost and have become a key building block for scaling large language models (LLMs) to trillion-parameter regimes.
By Minyu Cui, Anna Wingkvist, Morgan Ericsson