Cloud Computing Real-time Task Scheduling Optimization Based on Genetic Algorithm and the Perception of Resources

DOI

10.2991/icmmcce-15.2015.508How to use a DOI?

Keywords

Genetic algorithm, Resource scheduling, Mult-fitness, Resources perception

Abstract

Current real-time task scheduling algorithm only focus on the user tasks’ real-time demand, and these algorithms are not flexible enough to adapt for real-time change in heterogeneous systems. In this paper, by means of the characteristics of global optimization searching of genetic algorithm, from the point of user's real-time demand and the overall throughput of system, we design the fitness function based on real-time and overall throughput. Aimed at solving the existing problems of slow convergence speed of genetic algorithm, based on the strategy of resources perception, according to the size of the workload, the virtual machine parameters and load conditions (I/O intensive or CPU intensive), we guide the process of convergence of genetic algorithm, to make it to the larger probability of the corresponding type task variation or evolve to the adapted virtual machine, therefore we can accelerate the convergence process.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jian Dong
AU  - Su-Juan Qin
PY  - 2015/12
DA  - 2015/12
TI  - Cloud Computing Real-time Task Scheduling Optimization Based on Genetic Algorithm and the Perception of Resources
BT  - Proceedings of the 4th International Conference on Mechatronics, Materials, Chemistry and Computer Engineering 2015
PB  - Atlantis Press
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmcce-15.2015.508
DO  - 10.2991/icmmcce-15.2015.508
ID  - Dong2015/12
ER  -