Evaluating Open-Source Cloud Computing Solutions for Geosciences
Qunying Huang,Chaowei Yang,Kai Liu,Jizhe Xia,Chen Xu,Jing Li,Zhipeng Gui,Min Sun,Zhenglong Li
DOI: https://doi.org/10.1016/j.cageo.2013.05.001
IF: 5.168
2013-01-01
Computers & Geosciences
Abstract:Many organizations start to adopt cloud computing for better utilizing computing resources by taking advantage of its scalability, cost reduction, and easy to access characteristics. Many private or community cloud computing platforms are being built using open-source cloud solutions. However, little has been done to systematically compare and evaluate the features and performance of open-source solutions in supporting Geosciences. This paper provides a comprehensive study of three open-source cloud solutions, including OpenNebula, Eucalyptus, and CloudStack. We compared a variety of features, capabilities, technologies and performances including: (1) general features and supported services for cloud resource creation and management, (2) advanced capabilities for networking and security, and (3) the performance of the cloud solutions in provisioning and operating the cloud resources as well as the performance of virtual machines initiated and managed by the cloud solutions in supporting selected geoscience applications. Our study found that: (1) no significant performance differences in central processing unit (CPU), memory and I/O of virtual machines created and managed by different solutions, (2) OpenNebula has the fastest internal network while both Eucalyptus and CloudStack have better virtual machine isolation and security strategies, (3) Cloudstack has the fastest operations in handling virtual machines, images, snapshots, volumes and networking, followed by OpenNebula, and (4) the selected cloud computing solutions are capable for supporting concurrent intensive web applications, computing intensive applications, and small-scale model simulations without intensive data communication.