Related works such as (13; 16; 19) primarily aim to improve system-level performance through optimizing scheduling, energy efficiency, and model parallelism.
scheduling in llm inference using machine learning
29 verified PDFs
Machine learning has shifted scheduling research from designing better solvers to designing learning-enabled decision systems. This paradigm shift moves ...
by B Pang · 2025 · Cited by 6 — The numerical experiments demonstrate throughput gains of 8% to 28% and capacity improvements of 22% compared to traditional static batching ...
In this paper we describe the implementation of a simu- lated datacenter which we used to develop a machine learning- based scheduling algorithm, ML-SJF.
ABSTRACT. To improve the efficiency of distributed large language model (LLM) inference, various parallelization strategies, such as tensor and pipeline ...
by AR Bambhaniya · Cited by 1 — HERMES empowers system designers to navigate the evolving landscape of LLM inference, providing actionable insights into optimizing hardware ...
This survey aims to provide researchers and practitioners with a comprehensive understanding of DLT job scheduling in GPU clusters and to point out directions ...
In Section VI we present an overview of various machine learning approaches used in scheduling. Finally, in Section VI1 we suggest areas for future research.
by P Vellaisamy · 2025 · Cited by 6 — This work focuses on LLM inference, specifically encoder-only and decoder- only workloads, to derive insights for optimizing latency in evolving CPU-GPU coupled ...
