Scheduling moldable tasks on homogeneous multi-cluster platforms with GPUs

This paper examines task scheduling in homogeneous multi-cluster platforms, equipped with Graphics Processing Units (GPUs), with the aim of minimizing the makespan. We assume that tasks can be parallelized across these platforms under the moldable model. Recognizing the NP-hard nature of the problem...

Full description

Saved in:
Bibliographic Details
Published in:Computers & operations research Vol. 179; p. 107041
Main Authors: Wu, Fangfang, Zhang, Run, Zhang, Xiandong
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.07.2025
Subjects:
Approximation algorithm
Dual approximation algorithm
Moldable task scheduling
Multi-cluster platform
Dual approximation algorithm
Multi-cluster platform
Moldable task scheduling
Approximation algorithm
ISSN:0305-0548
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper examines task scheduling in homogeneous multi-cluster platforms, equipped with Graphics Processing Units (GPUs), with the aim of minimizing the makespan. We assume that tasks can be parallelized across these platforms under the moldable model. Recognizing the NP-hard nature of the problem, our goal is to develop algorithms that provide approximation ratios. While existing research has established algorithms for single-cluster GPU environments, scaling these to multi-cluster platforms introduces new challenges, especially due to the restriction that tasks cannot use processors from different clusters. We propose an integer programming-based algorithm that achieves an approximation ratio of 32+ϵ, trading off runtime for an improved approximation ratio. Additionally, leveraging recent theoretical advancements, we have created a polynomial-time algorithm with an approximation ratio of 2+ϵ. Empirical computational experiments show that our algorithms surpass their counterparts in empirical approximation ratios.
ISSN:0305-0548
DOI:10.1016/j.cor.2025.107041