Grid resource prediction based on support vector regression and genetic algorithms

Authors:
Liang Hu;Guosheng Hu;Kuo Tang;Xilong Che
Affiliations:
College of Computer Science and Technology, Jilin University, ChangChun, China;College of Computer Science and Technology, Jilin University, ChangChun, China;College of Computer Science and Technology, Jilin University, ChangChun, China;College of Computer Science and Technology, Jilin University, ChangChun, China
Venue:
ICNC'09 Proceedings of the 5th international conference on Natural computation
Year:
2009

Citing 7
Cited 0

Multilayer feedforward networks are universal approximators

Neural Networks
The grid: blueprint for a new computing infrastructure

The grid: blueprint for a new computing infrastructure
The network weather service: a distributed resource performance forecasting service for metacomputing

Future Generation Computer Systems - Special issue on metacomputing
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control and Artificial Intelligence

Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control and Artificial Intelligence
An Evaluation of Linear Models for Host Load Prediction

HPDC '99 Proceedings of the 8th IEEE International Symposium on High Performance Distributed Computing
Support Vector Machines: Theory and Applications (Studies in Fuzziness and Soft Computing)

Support Vector Machines: Theory and Applications (Studies in Fuzziness and Soft Computing)
Predict task running time in grid environments based on CPU load predictions

Future Generation Computer Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to manage the grid resources more effectively and provide a more suitable task scheduling strategy, the prediction information of grid resources is necessary in the grid system. In this study, support vector regression (SVR), which is a novel and effective regression algorithm, is applied to grid resource prediction. In order to build an effective SVR model, SVR's parameters must be selected carefully. Therefore, we develop a genetic algorithm-based SVR (GA-SVR) model that can automatically determine the optimal parameters of SVR with higher predictive accuracy and generalization ability simultaneously. This study pioneered on employing genetic algorithm to optimize the parameters of SVR for grid resource prediction. The performance of the hybrid model (GASVR), the back-propagation neural network (BPNN) and traditional SVR model whose parameters are obtained by trial-and-error procedure (T-SVR) have been compared with benchmark data set. Experimental results demonstrate that GA-SVR model works better than the other two models.