A new method based on PSR and EA-GMDH for host load prediction in cloud computing system

Qiangpeng Yang, Chenglei Peng, He Zhao, Yao Yu, Yu Zhou, Ziqiang Wang, Sidan Du

    Research output: Contribution to journalArticleResearchpeer-review

    32 Citations (Scopus)

    Abstract

    Host load prediction is one of the most effective measures for improving resource utilization in cloud computing systems. Due to the drastic fluctuation of the host load in the Cloud, accurately predicting the host load remains a challenge. In this paper, we propose a new prediction method that combines the Phase Space Reconstruction method and the Group Method of Data Handling based on an Evolutionary Algorithm. The performance of our proposed method is evaluated using two real-world load traces. The first is the load trace in a traditional distributed system, whereas the second is in a Google data center. The results show that the proposed method achieves a better prediction performance than some state-of-the-art methods.

    Original languageEnglish
    Pages (from-to)1402-1417
    Number of pages16
    JournalJournal of Supercomputing
    Volume68
    Issue number3
    DOIs
    Publication statusPublished - Jun 2014

    Keywords

    • Evolutionary Algorithm
    • Group Method of Data Handling
    • Host load prediction
    • Phase Space Reconstruction

    Cite this