The article outlines the problem of long-term calculation of basin models, which remains relevant for many years of the technology existence. Information is provided on the theoretically possible increase in the performance of model calculation due to the use of parallel computing; factors that limit the achievement of a multiple increase in practice are identified. The analysis of the technical implementation of parallel computing of basin models in the PetroMod® software was carried out, its strengths and weaknesses were identified. The characteristics of the Hybrid and Combined petroleum migration modeling methods, created to reduce the total time for calculating basin models in comparison with the Darcy flow petroleum migration modeling method, are given. The key factors influencing the calculation time of basin models using different methods of petroleum migration modeling are determined. A benchmarking technique based on a set of calculations of a regional basin model consisting of more than 5,000,000 cells is described. The efficiency of parallel computing is determined depending on the number of CPU cores/threads involved in the calculation, channels and the total amount of RAM, as well as the configuration of the basin model. The increase in performance when using computers based on hardware of both the server and consumer segments is numerically estimated. Based on the benchmarking results, the main factors of high performance for all processes included in the calculation of the basin model were identified. Recommendations on the choice of hardware that allows to achieve the highest performance of basin model calculations in the PetroMod® software are given.
References
1. Baur F., Scheirer A.H., Peters K.E., Past, present, and future of basin and petroleum system modeling, AAPG Bulletin, 2018, V. 102, no. 4, pp. 549-561,
DOI: http://dx.doi.org/10.1306/08281717049
2. Hantschel T., Kauerauf A.I., Wygrala B., Finite element analysis and ray tracing modeling of petroleum migration, Marine and Petroleum Geology, 2000, V 17, pp. 815–820, DOI: https://doi.org/10.1016/S0264-8172(99)00061-6
3. Hantschel T., Kauerauf A., Fundamentals of basin and petroleum systems modeling, Berlin Heidelberg: Springer-Verlag, 2009.
4. Sutter H., The free lunch is over: A fundamental turn toward concurrency in software, Dr. Dobb’s journal, 2005, V. 30(3), pp. 202-210.
5. Amdahl G.M., Validity of a single processor approach to achieving large scale computer capabilities, Proceedings of American Federation of Information Processing Societies (AFIPS) Spring Joint Computer Conference, Atlantic City, NJ, USA, 1967, pp. 483–485, DOI: https://doi.org/10.1109/N-SSC.2007.4785615
6. Bücker H.M., Kauerauf A.I., Rasch A., A smooth transition from serial to parallel processing in the industrial petroleum system modeling package PetroMod, Computers & Geosciences, 2008, V. 34, pp. 1473-1479, DOI: http://dx.doi.org/10.1016/j.cageo.2008.01.011
7. Bogachev K., Milyutin S., Telishev A. et al., High-performance reservoir simulations on modern CPU-GPU computational platforms, ICE–2018.
8. Pegaz-Fiornet S., Carpentier B., Michel A., Wolf S., Comparison between the different approaches of secondary and tertiary hydrocarbon migration modeling in basin simulators, Basin Modeling: New Horizons in Research and Applications: AAPG Hedberg, Series, 2012, no. 4, pp. 221 – 236.
9. Patent US20140358502A1, Combined migration method for basin modeling, Inventors: Kleine A., Kauerauf A.I.
10. Baur F., Katz B., Some practical guidance for petroleum migration modeling, Marine and Petroleum Geology, 2018, V. 93, pp. 409–421,
DOI: https://doi.org/10.1016/j.marpetgeo.2018.03.003
11. Baur F., Di Benedetto M., Fuchs T. et al., Integrating structural geology and petroleum systems modeling - A pilot project from Bolivian’s fold and thrust belt, Marine and Petroleum Geology, 2009, V. 26, pp. 573–579, DOI: https://doi.org/10.1016/j.marpetgeo.2009.01.004
