Open AccessMulti-objective Task Scheduling Algorithm Based on Harmony Search for Grid Microservice Optimization
Author(s) -
Yuqing Xie,
Yuan Wang,
Ying Jiang,
Zirong Peng,
Yongli Wang
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1746/1/012040
Subject(s) - harmony search , computer science , mathematical optimization , scheduling (production processes) , multi objective optimization , test functions for optimization , grid , algorithm , pareto principle , optimization problem , multi swarm optimization , artificial intelligence , mathematics , machine learning , geometry
In order to solve the multi-objective task scheduling optimization problem of data platform microservice combination, this paper proposes an improved harmony search algorithm MODRHS, which fuses the multi-objective optimization NSGA-II algorithm, and improve its elite selection strategy by a circular elimination method, and strengthens the uniformity of the non-dominated solutions at first layer and improves the algorithm’s Pareto optimization ability. At the same time, it describes the mathematical expression of the Dockers model and service optimization goals of the data platform microservice scheduling. The improved harmony search algorithm is applied to practical problems. The performance test of MODRHS is carried out through standard test functions and real data sets. Compared with other classic multi-objective optimization algorithms, the convergence and diversity of the MODRHS algorithm have been effectively improved.