CFP last date
22 April 2024
Reseach Article

Evaluating the Impact of Full Virtualized High- Performance Computing Platform on Large Scale Scientific Data using Quantum Espresso

by Michael Ametepe Kattah, Dominic Asamoah, Frimpong Twum
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Volume 177 - Number 37
Year of Publication: 2020
Authors: Michael Ametepe Kattah, Dominic Asamoah, Frimpong Twum
10.5120/ijca2020919856

Michael Ametepe Kattah, Dominic Asamoah, Frimpong Twum . Evaluating the Impact of Full Virtualized High- Performance Computing Platform on Large Scale Scientific Data using Quantum Espresso. International Journal of Computer Applications. 177, 37 ( Feb 2020), 10-14. DOI=10.5120/ijca2020919856

@article{ 10.5120/ijca2020919856,
author = { Michael Ametepe Kattah, Dominic Asamoah, Frimpong Twum },
title = { Evaluating the Impact of Full Virtualized High- Performance Computing Platform on Large Scale Scientific Data using Quantum Espresso },
journal = { International Journal of Computer Applications },
issue_date = { Feb 2020 },
volume = { 177 },
number = { 37 },
month = { Feb },
year = { 2020 },
issn = { 0975-8887 },
pages = { 10-14 },
numpages = {9},
url = { https://ijcaonline.org/archives/volume177/number37/31145-2020919856/ },
doi = { 10.5120/ijca2020919856 },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
}
%0 Journal Article
%1 2024-02-07T00:47:59.290813+05:30
%A Michael Ametepe Kattah
%A Dominic Asamoah
%A Frimpong Twum
%T Evaluating the Impact of Full Virtualized High- Performance Computing Platform on Large Scale Scientific Data using Quantum Espresso
%J International Journal of Computer Applications
%@ 0975-8887
%V 177
%N 37
%P 10-14
%D 2020
%I Foundation of Computer Science (FCS), NY, USA
Abstract

High Performance Computing (HPC) applications are becoming vital in scientific research for analyzing large scale scientific data, but there is inadequate knowledge on the impact of a fully virtualized HPC cluster on these applications when they are used to analyzed large scientific data. The main purpose of this research is to carry out a comparative experiment on a virtual HPC cluster and the traditional HPC cluster by executing a benchmarking tool called para-speedup with an input file on both clusters using Quantum Espresso (QE) as an HPC application to determine the impact on the clusters. The research focuses on Central Processing Unit (CPU) utilization, turnaround time of jobs run on the cluster, memory and input/output (I/O) operations. The virtual cluster was setup using VMWare ESXi 5.5.0 as hypervisor of choice and ROCKS was installed on the cluster as an HPC platform of choice. During the experiment, it was observed that the job was not memory and I/O intensive on both clusters, so there was little to discuss on these metrics but generally it was observed that running job using HPC applications like QE on a fully virtualized HPC cluster to analyze large scale scientific data has a negative performance impact on the completion of the job as compared to the traditional cluster.

References
  1. Ranadive, A. et al. (2008) ‘Performance implications of virtualizing multicore cluster machines’, in Proceedings of the 2nd workshop on System-level virtualization for high performance computing - HPCVirt ’08. New York, New York, USA: ACM Press, pp. 1–8. doi: 10.1145/1435452.1435453.
  2. Netto, M. A. S. et al. (2018) ‘HPC Cloud for Scientific and Business Applications: Taxonomy, Vision, and Research Challenges’, 51(1), pp. 1–29. doi: 10.1145/3150224.
  3. Gavrilovska, A. et al. (2007) ‘High-performance hypervisor architectures: Virtualization in hpc systems’, virtualization for HPC ( …. Available at: http://www.cc.gatech.edu/~adit262/docs/HPHA-HPCVirt2007.pdf (Accessed: 4 June 2017).
  4. Walters, J. P. et al. (2008) ‘A comparison of virtualization technologies for HPC’, in Proceedings - International Conference on Advanced Information Networking and Applications, AINA, pp. 861–868. Available at: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4482796 (Accessed: 25 October 2016).
  5. Regola, N. and Ducom, J.-C. J. (2010) ‘Recommendations for virtualization technologies in high performance computing’, Cloud Computing Technology and, pp. 409–416. Available at: http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=5708479 (Accessed: 25 October 2016).
  6. Giannozzi, P. and Cavazzoni, C. (2009) ‘Large-scale computing with Quantum-Espresso’, Nuovo Cimento C, Vol. 32 C,. doi: 10.1393/ncc/i2009-10368-9.
  7. Vmware.com (2015) Virtualization Technology & Virtual Machine Software | United States, VMware, Inc. Available at: https://www.vmware.com/solutions/virtualization.html (Accessed: 8 September 2017).
  8. Trangoni, M. and Cabral, M. (2012) ‘A Comparison of Provisioning Systems for Beowulf Clusters’, (3).
  9. Hassani, R. and Luksch, P. (2014) ‘Performance Analysis of HPC Applications Running in Public Cloud’, Future Generation, 00(2012). Available at: http://www.sciencedirect.com/science/article/pii/S0167739X12001458 (Accessed: 25 October 2016).
  10. Carlyle, A. G., Harrell, S. L. and Smith, P. M. (2010) ‘Cost-effective HPC: The community or the cloud?’, Proceedings - 2nd IEEE International Conference on Cloud Computing Technology and Science, CloudCom 2010, pp. 169–176. doi: 10.1109/CloudCom.2010.115.
  11. Wu, D., Terpenny, J. and Gentzsch, W. (2015) ‘Cloud-Based Design, Engineering Analysis, and Manufacturing: A Cost-Benefit Analysis’, Procedia Manufacturing. Elsevier B.V., 1, pp. 64–76. doi: 10.1016/j.promfg.2015.09.061.
  12. Marathe, A. et al. (2013) ‘A comparative study of high-performance computing on the cloud’, Proceedings of the 22nd international symposium on High-performance parallel and distributed computing SE - HPDC ’13, pp. 239–250. doi: doi: 10.1145/2462902.2462919.
  13. Pawar, S. and Singh, S. (2015) ‘Performance Comparison of VMware and Xen Hypervisor on Guest OS’, International Journal of Innovative Computer Science & Engineering Issue, 2(3), pp. 56–60. Available at: http://ijicse.in/wp-content/uploads/2015/07/v2i3-14.pdf.
Index Terms

Computer Science
Information Sciences

Keywords

High Performance Computing Cluster Virtualization VMWare Quantum Espresso ROCKS Hypervisor