Computational & Technology Resources
an online resource for computational,
engineering & technology publications
PROCEEDINGS OF THE FOURTH INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, GRID AND CLOUD COMPUTING FOR ENGINEERING
Edited by: P. Iványi and B.H.V. Topping
Performance Evaluation of the XDEM Framework on the OpenStack Cloud Computing Middleware
X. Besseron1, V. Plugaru2, A.H. Mahmoudi1, S. Varrette2, B. Peters1 and P. Bouvry2
1Research Unit in Engineering Science, University of Luxembourg, Luxembourg
X. Besseron, V. Plugaru, A.H. Mahmoudi, S. Varrette, B. Peters, P. Bouvry, "Performance Evaluation of the XDEM Framework on the OpenStack Cloud Computing Middleware", in P. Iványi, B.H.V. Topping, (Editors), "Proceedings of the Fourth International Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 20, 2015. doi:10.4203/ccp.107.20
Keywords: discrete element method, high performance computing, cloud computing, OpenStack, XEN, KVM.
As Cloud Computing services become ever more prominent, it appears necessary to assess the efficiency of these solutions. This paper presents a performance evaluation of the OpenStack Could Computing middleware using our XDEM application simulating the pyrolysis of biomass as a benchmark.
A systematic study is presented, based on a fully automated benchmarking framework to evaluate three different configurations: native (i.e. no virtualization), OpenStack with KVM and XEN hypervisors. The authors' approach features the following advantages: real user application, the fair comparison using the same hardware, the large scale distributed execution, while being fully automated and reproducible.
Experiments have been run on two different clusters, using up to 432 cores. Results show a moderate overhead for sequential execution and a significant penalty for distributed execution under the Cloud middleware. The overhead on multiple nodes is between 10% and 30% for OpenStack/KVM and 30% and 60% for OpenStack/XEN.
purchase the full-text of this paper (price £20)