Performance comparison for Resource Management and OAR SLURM in Scientific Application Simulation Ionospheric

• Gabriela Luisa Eckel UFSM
• Fernando Emilio Puntel UFSM
• Adriano Petry INPE

DOI: https://doi.org/10.5753/eradrs.2020.10755

Abstract

With the constant evolution of technology, high-performance computing is one of the main areas of research in computational means. high performance environments are often used by applications that need to handle a lot of data in a short period of time. For this they can be used distributed systems aimed at increasing efficiency and execution speed. With the large amount of computational resources in high-performance environments, management systems features such as OAR and SLURM, are employed to optimize their use. We carried out a comparative analysis of the managers mentioned systems in a real scientific application, daily simulates the behavior of the Earth's ionosphere in South America by generating total electron content maps.

Keywords: Evaluation, Performance Measurement and Prediction, Cloud Computing, grid, cluster (cluster) and peer-to-peer, Scheduling and load balancing, Languages, Compilers and Tools Parallel and Distributed Computing, Distributed systems, Techniques Modeling and Simulation, Fault Tolerance

References

Nicolas, C., Joseph, E., and ZIRST, M. (2003-2016). Oar documentation-user guide. LIG laboratory, Laboratoire d’Informatique de Grenoble Bat. ENSIMAG-antenne de Montbonnot ZIRST.

Petry, A., de Souza, J. R., de Campos Velho, H. F., Pereira, A. G., and Bailey, G. J. (2014). First results of operational ionospheric dynamics prediction for the brazilian space weather program. Advances in Space Research, 54(1):22–36.

Prabhu, C. (2008). Grid and cluster computing. PHI Learning Pvt. Ltd.

Yoo, A. B., Jette, M. A., and Grondona, M. (2003). Slurm: Simple linux utility for resource management. In Feitelson, D., Rudolph, L., and Schwiegelshohn, U., editors, Job Scheduling Strategies for Parallel Processing, pages 44–60, Berlin, Heidelberg. Springer Berlin Heidelberg.

Zhou, X., Chen, H., Wang, K., Lang, M., and Raicu, I. (2013). Exploring distributed resource allocation techniques in the slurm job management system. Illinois Institute of Technology, Department of Computer Science, Technical Report.