Stochastic optimization of GeantV code by use of genetic algorithms

Journal Article · 01 October 2017 · Journal of Physics. Conference Series

DOI: https://doi.org/10.1088/1742-6596/898/4/042026 · OSTI ID:1421538

Amadio, G. ^[1]; Apostolakis, J. ^[2]; Bandieramonte, M. ^[2]; Behera, S. P. ^[3]; Brun, R. ^[2]; Canal, P. ^[4]; Carminati, F. ^[2]; Cosmo, G. ^[2]; Duhem, L. ^[5]; Elvira, D. ^[4]; Folger, G. ^[2]; Gheata, A. ^[6]; Gheata, M. ^[7]; Goulas, I. ^[2]; Hariri, F. ^[2]; Jun, S. Y. ^[4]; Konstantinov, D. ^[2]; Kumawat, H. ^[3]; Ivantchenko, V. ^[2]; Lima, G. ^[4] ; Nikitina, T. ^[2]; Novak, M. ^[2]; Pokorski, W. ^[2]; Ribon, A. ^[2]; Seghal, R. ^[3]; Shadura, O. ^[2]; Vallecorsa, S. ^[6]; Wenzel, S. ^[2]

---

1. Sao Paulo State Univ. (UNESP), Sao Paulo (Brazil). Parallel Computing Center
2. European Organization for Nuclear Research (CERN), Meyrin (Switzerland)
3. Bhabha Atomic Research Centre (BARC), Mumbai (India)
4. Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
5. Intel Corporation, Santa Clara, CA (United States)
6. European Organization for Nuclear Research (CERN), Meyrin (Switzerland); Intel Corporation, Santa Clara, CA (United States)
7. European Organization for Nuclear Research (CERN), Meyrin (Switzerland); Inst. of Space Sciences, Bucharest-Magurele (Romania)

GeantV is a complex system based on the interaction of different modules needed for detector simulation, which include transport of particles in fields, physics models simulating their interactions with matter and a geometrical modeler library for describing the detector and locating the particles and computing the path length to the current volume boundary. The GeantV project is recasting the classical simulation approach to get maximum benefit from SIMD/MIMD computational architectures and highly massive parallel systems. This involves finding the appropriate balance between several aspects influencing computational performance (floating-point performance, usage of off-chip memory bandwidth, specification of cache hierarchy, etc.) and handling a large number of program parameters that have to be optimized to achieve the best simulation throughput. This optimization task can be treated as a black-box optimization problem, which requires searching the optimum set of parameters using only point-wise function evaluations. Here, the goal of this study is to provide a mechanism for optimizing complex systems (high energy physics particle transport simulations) with the help of genetic algorithms and evolution strategies as tuning procedures for massive parallel simulations. One of the described approaches is based on introducing a specific multivariate analysis operator that could be used in case of resource expensive or time consuming evaluations of fitness functions, in order to speed-up the convergence of the black-box optimization problem.

• Research Organization: Fermi National Accelerator Laboratory (FNAL), Batavia, IL (United States)
• Sponsoring Organization: USDOE Office of Science (SC), High Energy Physics (HEP)
• Grant/Contract Number: AC02-07CH11359
• OSTI ID: 1421538
• Report Number(s): FERMILAB-CONF-16-766-CD; 1638148; TRN: US1801524
• Journal Information: Journal of Physics. Conference Series, Vol. 898, Issue 4; ISSN 1742-6588
• Publisher: IOP PublishingCopyright Statement
• Country of Publication: United States
• Language: English