PhotonIcs and Electromagnetics Research Symposium,
also known as Progress In Electromagnetics Research Symposium
PIERS Proceedings
Published: 2015-07-09
An Intelligent Platform for Effective Management of Time-consuming Electromagnetic Simulation Problems
By
Proceedings of 2015 Photonics & Electromagnetics Research Symposium, Prague, July 6 - 9,Page(s)2199-2203
Abstract
Modern simulation applications that carry out large scale iterative processes, such as Monte-Carlo simulations, or manipulate large data structures, tend to be extremely time- consuming due to the shortage of computational resources or the inherent nature of the simulation itself. Typical simulations can take up to several days to complete on conventional systems, while high-end supercomputers can be cost-prohibitive. Therefore, the need for effective parallelization of software execution and resource management is more imperative. The goal of this paper is to present a fully-distributed platform that enables software simulations to be executed within user-acceptable time periods, by predicting the resource requirements of each simulation. In this context, the platform analyzes files that contain historical data about past executions of the particular simulation. Processor and memory utilization, overall execution time, level of parallelization and distributed execution are some of the information collected and used by an efficient training algorithm, in order to determine the optimal amount of resources to be allocated in a particular simulation. The training algorithm applies several machine-learning techniques such as multi-linear regression in order to efficiently predict the resource vector that will satisfy the user requirements.
Citation
Panagiotis K. Gkonis, Dimitra I. Kaklamani, Andreas P. Kapsalis, George A. Kyriacou, Iakovos S. Venieris, and Constantinos L. Zekios, "An Intelligent Platform for Effective Management of Time-consuming Electromagnetic Simulation Problems," Proceedings of 2015 Photonics & Electromagnetics Research Symposium, Prague, July 6 - 9,Page(s)2199-2203
References

1. Juve, G., M. Rynge, E. Deelman, J. S. Vockler, and G. B. Berriman, "Comparing futuregrid, Amazon EC2, and open science grid for scientific workflows," Computing in Science & Engineering, Vol. 15, No. 4, 2013.
doi:10.1109/mcse.2013.44        Google Scholar

2. Vecchiola, C., R. N. Calheiros, D. Karunamoorthy, and R. Buyya, "Deadline-driven provisioning of resources for scientific applications in hybrid clouds with Aneka," Future Generation Computer Systems, Vol. 28, No. 1, 58-65, 2012.
doi:10.1016/j.future.2011.05.008        Google Scholar

3. Open Science Grid, Accessed, Mar. 2015, http://www.opensciencegrid.org/about/.

4. Quinlan, R. J., "Learning with continuous classes," Proc. of the 5th Australian Joint Conference on Artificial Intelligence, 1992.        Google Scholar

5. Xiong, P., Y. Chi, S. Zhuy, H. J. Moon, C. Pu, and H. Hacigumus, "Intelligent management of virtualized resources for database systems in cloud environment," Proc. of 2011 IEEE 27th International Conference on Data Engineering (ICDE), 87-98, Hannover, 2011.
doi:10.1109/icde.2011.5767928        Google Scholar

6. Zhao, Y., X. Fei, I. Raicu, and S. Lu, "Opportunities and challenges in running scientific workflows on the cloud," Proc. of CyberC 2011, 455-462, Beijing, 2011.
doi:10.1109/cyberc.2011.80        Google Scholar

7. Zekios, C. L., et al. "Analytical and numerical eigenanalysis of electromagnetic structures: A review," European Microwave Week, Paris, 2015.
doi:10.1109/eumc.2015.7345684        Google Scholar

8. Gkonis, P. K., et al. "On the performance of spatial multiplexing in MIMO-WCDMA networks with principal component analysis at the reception," EuCAP 2015, Lisbon, 2015.        Google Scholar