Research on the performance of virtualization-based remote sensing data processing platform

With the development of virtualization, a significant technology used in cloud computing, a new kind of remote sensing data processing platform based on virtualization has been formulating. In order to find out how virtual machines will influence the performance of remote sensing data processing platforms, KVM, VMware and Xen are tested as the instances of virtual machines by the drought model algorithm as a test application on remote sensing platforms. Besides, the test application is carefully studied and then divided into three procedures to find out which procedure is the main factor. In this paper, the results show that VMware and Xen deliver the superior to build the remote sensing data processing platforms, and that I/O performance, especially reading, is the obstacle to hinder the remote sensing application to migrate from Host Operating Systems to Guest Operating Systems.