Performance analysis of container-based hadoop cluster: OpenVZ and LXC

Hadoop is an open-source software framework that commonly used for distributed processing and distributed storage in large datasets (big data). It perform processing and stored the data across all nodes in cluster. But, the deployment and operational costs of physical cluster are costly in some reasons. Physical cluster needs high energy and rigid in manageability. As the solution, virtualization technology used to get more flexibility and reduce costs. But, hypervisor-based virtualization gives more I/O performance overhead compared to bare-metal server. Nowadays, container-based virtualization are used to obtain lower performance overhead. Open Virtuozzo (OpenVZ) and Linux Container (LXC) are the examples of container-based virtualization. Implementation of container-based virtualization gives near native performances.