Resource Self-adaptive Allocation Method Based on Mixed Prediction Cloud Platform

• DOI: 10.14257/ijdta.2015.8.3.23
• Publication Date: 2016-02-24T21:22:13Z
• Authors: Hong Qi; Honge Ren; Guanglei Zhang
• Keywords: 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

There are some problems in the existing cloud platform resource allocation methods, such as low rates of utilization and lack accurate prediction trend changes resources, etc.To solve these problems, MPRA(Mixed Prediction Based Resource Allocation ) was proposed.According to periodic non-periodic characteristics service resources demand, MPRA first adopts FFT(Fast Fourier Transform) judge characteristics.For without characteristics, it uses Markov process predict, obtains higher accuracy, thus, ensure user experience.The experimental results show that can accurately predict change requirements, then allocate virtual machine self-adaptively according results.Obviously, has improved utilization, reduced occupation number physical machines effectively violation times SLA (Service-level Agreement).