Copertina

Large-Scale LLM Inference with Heterogeneous Workloads: Prefill-Decode Contention and Asymptotically Optimal Control

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Lin, Ruihan, Ding, Zezhen, Han, Zean, Zhang, Jiheng
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Distributed, Parallel, and Cluster Computing Optimization and Control
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Green-LLM: Optimal Workload Allocation for Environmentally-Aware Distributed Inference
di: Cheng, Jiaming, et al.
Pubblicazione: (2025)

From Sequential to Parallel: Reformulating Dynamic Programming as GPU Kernels for Large-Scale Stochastic Combinatorial Optimization
di: Zhao, Jingyi, et al.
Pubblicazione: (2026)

PrefillOnly: An Inference Engine for Prefill-only Workloads in Large Language Model Applications
di: Du, Kuntai, et al.
Pubblicazione: (2025)

Modeling Task Mapping for Data-intensive Applications in Heterogeneous Systems
di: Wilhelm, Martin, et al.
Pubblicazione: (2022)

First Provably Optimal Asynchronous SGD for Homogeneous and Heterogeneous Data
di: Maranjyan, Artavazd
Pubblicazione: (2026)

A GPU-Accelerated Distributed Algorithm for Optimal Power Flow in Distribution Systems
di: Ryu, Minseok, et al.
Pubblicazione: (2025)

Decentralized Nonconvex Optimization under Heavy-Tailed Noise: Normalization and Optimal Convergence
di: Yu, Shuhua, et al.
Pubblicazione: (2025)

Distributed Stochastic Momentum Tracking with Local Updates: Achieving Optimal Communication and Iteration Complexities
di: Huang, Kun, et al.
Pubblicazione: (2025)

D-PDLP: Scaling PDLP to Distributed Multi-GPU Systems
di: Li, Hongpei, et al.
Pubblicazione: (2026)

Accelerating Optimal Power Flow with GPUs: SIMD Abstraction of Nonlinear Programs and Condensed-Space Interior-Point Methods
di: Shin, Sungho, et al.
Pubblicazione: (2023)

Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity
di: Mahran, Ammar, et al.
Pubblicazione: (2026)

Ringleader ASGD: The First Asynchronous SGD with Optimal Time Complexity under Data Heterogeneity
di: Maranjyan, Artavazd, et al.
Pubblicazione: (2025)

Temporal Parallelisation of the HJB Equation and Continuous-Time Linear Quadratic Control
di: Särkkä, Simo, et al.
Pubblicazione: (2022)

Decentralized Gradient-Free Methods for Stochastic Non-Smooth Non-Convex Optimization
di: Lin, Zhenwei, et al.
Pubblicazione: (2023)

Optimality in Decentralized Optimization under Bandwidth Constraints
di: Tyurin, Alexander
Pubblicazione: (2026)

OptPipe: Memory- and Scheduling-Optimized Pipeline Parallelism for LLM Training
di: Li, Hongpei, et al.
Pubblicazione: (2025)

Distributed Constrained Combinatorial Optimization leveraging Hypergraph Neural Networks
di: Heydaribeni, Nasimeh, et al.
Pubblicazione: (2023)

Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systems
di: Wilkins, Grant, et al.
Pubblicazione: (2024)

Load Balancing with Network Latencies via Distributed Gradient Descent
di: Balseiro, Santiago R., et al.
Pubblicazione: (2025)

An Optimistic Gradient Tracking Method for Distributed Minimax Optimization
di: Huang, Yan, et al.
Pubblicazione: (2025)

Tailoring Gradient Methods for Differentially-Private Distributed Optimization
di: Wang, Yongqiang, et al.
Pubblicazione: (2022)

A First-Order Algorithm for Decentralised Min-Max Problems
di: Malitsky, Yura, et al.
Pubblicazione: (2023)

Problem-Parameter-Free Decentralized Nonconvex Stochastic Optimization
di: Li, Jiaxiang, et al.
Pubblicazione: (2024)

Decentralized Distributed Optimization for Saddle Point Problems
di: Rogozin, Alexander, et al.
Pubblicazione: (2021)

Decentralized Nonsmooth Nonconvex Optimization with Client Sampling
di: Chen, Xinyan, et al.
Pubblicazione: (2026)

Non-ergodic linear convergence property of the delayed gradient descent under the strongly convexity and the Polyak-Łojasiewicz condition
di: Choi, Hyung Jun, et al.
Pubblicazione: (2023)

Krylov Solvers for Interior Point Methods with Applications in Radiation Therapy and Support Vector Machines
di: Liu, Felix, et al.
Pubblicazione: (2023)

Smoothed Gradient Clipping and Error Feedback for Decentralized Optimization under Symmetric Heavy-Tailed Noise
di: Yu, Shuhua, et al.
Pubblicazione: (2023)

Distributed Constraint-Coupled Optimization: Harnessing ADMM-consensus for robustness
di: Messilem, Mohamed Abdelmouamin, et al.
Pubblicazione: (2025)

An efficient implementation of parallel simulated annealing algorithm in GPUs
di: Ferreiro, A. M., et al.
Pubblicazione: (2024)

One-Point Feedback for Composite Optimization with Applications to Distributed and Federated Learning
di: Beznosikov, Aleksandr, et al.
Pubblicazione: (2021)

Average-case optimization analysis for distributed consensus algorithms on regular graphs
di: Nguyen, Nhat Trung, et al.
Pubblicazione: (2024)

GPU-Accelerated Primal Heuristics for Mixed Integer Programming
di: Çördük, Akif, et al.
Pubblicazione: (2025)

Disaggregated Prefill and Decoding Inference System for Large Language Model Serving on Multi-Vendor GPUs
di: Chen, Xing, et al.
Pubblicazione: (2025)

Intelligent Router for LLM Workloads: Improving Performance Through Workload-Aware Load Balancing
di: Jain, Kunal, et al.
Pubblicazione: (2024)

The Limits and Potentials of Local SGD for Distributed Heterogeneous Learning with Intermittent Communication
di: Patel, Kumar Kshitij, et al.
Pubblicazione: (2024)

Efficient Federated Learning against Heterogeneous and Non-stationary Client Unavailability
di: Xiang, Ming, et al.
Pubblicazione: (2024)

Uncoded Storage Coded Transmission Elastic Computing with Straggler Tolerance in Heterogeneous Systems
di: Zhong, Xi, et al.
Pubblicazione: (2024)

FIARSE: Model-Heterogeneous Federated Learning via Importance-Aware Submodel Extraction
di: Wu, Feijie, et al.
Pubblicazione: (2024)

FedCanon: Non-Convex Composite Federated Learning with Efficient Proximal Operation on Heterogeneous Data
di: Zhou, Yuan, et al.
Pubblicazione: (2025)