Integrating Odeint Time Stepping into OpenFPM for Distributed and GPU Accelerated Numerical Solvers

Abhinav Singh; Landfried Kraatz; Serhii Yaskovets; Pietro Incardona; Ivo F. Sbalzarini

doi:10.5334/jors.598

Abstract

We present a scientific numerical software for multi-stage, multi-step, and adaptive explicit time integration on distributed-memory parallel computers and on Graphics Processing Units (GPUs). Our implementation integrates the Odeint library from Boost with the OpenFPM framework for scientific computing, enabling compact and scalable numerical simulation codes. Specifically, we extend the Odeint data types to OpenFPM’s metaprogramming system. This makes the time-integration methods from Odeint available in a concise template-expression language for numerical codes distributed and parallelized using OpenFPM. We benchmark the software for exponential and sigmoidal dynamics and present application examples to the 3D Gray-Scott reaction-diffusion problem and the “dam break” problem from fluid mechanics. We find a strong-scaling efficiency of 80% on up to 512 CPU cores and a five-fold speedup on a single GPU.

References

Ahnert K, Mulansky M, Simos TE, Psihoyios G, Tsitouras C, Anastassi Z. Odeint – solving ordinary differential equations in C++. In AIP Conference Proceedings. AIP; 2011. DOI: 10.1063/1.3637934
Open DOI Search in Google Scholar Back to article
Ahnert M. ODEINT state types, algebras and operations; 2015. URL https://www.boost.org/doc/libs/1_66_0/libs/numeric/odeint/doc/html/index.html. [Online; accessed 20-April-2021].
Search in Google Scholar Back to article
Alam S, Najma B, Singh A, Laprade J, Gajeshwar G, Yevick HG, Baskaran A, Foster PJ, Duclos G. Active Fréedericksz transition in active nematic droplets. Phys. Rev. X. 2024;14:041002. URL https://link.aps.org/doi/10.1103/PhysRevX.14.041002
Search in Google Scholar Back to article
Crespo A, Domínguez J, Rogers B, Gómez-Gesteira M, Longshaw S, Canelas R, Vacondio R, Barreiro A, García-Feal O. Dualsphysics: Open-source parallel cfd solver based on smoothed particle hydrodynamics (sph). Computer Physics Communications. 2015;187:204–216. URL https://www.sciencedirect.com/science/article/pii/S0010465514003397
Search in Google Scholar Back to article
Gardner DJ, Reynolds DR, Woodward CS, Balos CJ. Enabling new flexibility in the SUNDIALS suite of nonlinear and differential/algebraic equation solvers. ACM Transactions on Mathematical Software; 2022. DOI: 10.1145/3539801
Open DOI Search in Google Scholar Back to article
Harris CR, Millman KJ, van der Walt SJ, Gommers R, Virtanen P, Cournapeau D, Wieser E, Taylor J, Berg S, Smith NJ, Kern R, Picus M, Hoyer S, van Kerkwijk MH, Brett M, Haldane A, del Río JF, Wiebe M, Peterson P, Gérard-Marchant P, Sheppard K, Reddy T, Weckesser W, Abbasi H, Gohlke C, Oliphant TE. Array programming with NumPy. Nature. 2020;585(7825):357–362. DOI: 10.1038/s41586-020-2649-2
Open DOI Search in Google Scholar Back to article
Incardona P, Gupta A, Yaskovets S, Sbalzarini IF. A portable C++ library for memory and compute abstraction on multi-core CPUs and GPUs. Concurrency Computat. Pract. Exper. 2023;e7870. DOI: 10.1002/cpe.7870
Open DOI Search in Google Scholar Back to article
Incardona P, Leo A, Zaluzhnyi Y, Ramaswamy R, Sbalzarini IF. OpenFPM: A scalable open framework for particle and particle-mesh codes on parallel computers. Computer Physics Communications. 2019;241:155–177. URL https://www.sciencedirect.com/science/article/pii/S0010465519300852
Search in Google Scholar Back to article
Karol S, Nett T, Castrillon J, Sbalzarini IF. A domain-specific language and editor for parallel particle methods. ACM Trans. Math. Softw. 2018;44(3):34:1–34:32. DOI: 10.1145/3175659
Open DOI Search in Google Scholar Back to article
Khouzami N, Michel F, Incardona P, Castrillon J, Sbalzarini IF. Model-based autotuning of discretization methods in numerical simulations of partial differential equations. Journal of Computational Science. 2022;57:101489. URL https://www.sciencedirect.com/science/article/pii/S1877750321001563
Search in Google Scholar Back to article
Khouzami N, Schütze L, Incardona P, Kraatz L, Subic T, Castrillon J, Sbalzarini IF. The OpenPME problem solving environment for numerical simulations. In Paszynski M, Kranzlmüller D, Krzhizhanovskaya VV, Dongarra JJ, Sloot PMA, editors. Computational Science – ICCS 2021. Lecture Notes in Computer Science, Springer International Publishing, Cham. 2021:614–627. DOI: 10.1007/978-3-030-77961-0_49
Open DOI Search in Google Scholar Back to article
MATLAB. Version 7.10.0 (R2010a). The MathWorks Inc., Natick, Massachusetts; 2010.
Search in Google Scholar Back to article
Monaghan JJ. Smoothed particle hydrodynamics. Annu. Rev. Astron. Astrophys. 1992;30:543–574. DOI: 10.1146/annurev.aa.30.090192.002551
Open DOI Search in Google Scholar Back to article
Press WH, Teukolsky SA. Adaptive stepsize Runge-Kutta integration. Computers in Physics. 1992;6(2):188–191. DOI: 10.1063/1.4823060
Open DOI Search in Google Scholar Back to article
Schrader B, Reboux S, Sbalzarini IF. Discretization correction of general integral pse operators for particle methods. Journal of Computational Physics. 2010;229(11):4159–4182. DOI: 10.1016/j.jcp.2010.02.004
Open DOI Search in Google Scholar Back to article
Singh A, Incardona P, Sbalzarini IF. A C++ expression system for partial differential equations enables generic simulations of biological hydrodynamics. Eur. Phys. J. E. 2021;44(9):117. DOI: 10.1140/epje/s10189-021-00121-x
Open DOI Search in Google Scholar Back to article
Singh A, Suhrcke PH, Incardona P, Sbalzarini IF. A numerical solver for active hydrodynamics in three dimensions and its application to active turbulence. Physics of Fluids. 2023;35(10). DOI: 10.1063/5.0169546
Open DOI Search in Google Scholar Back to article
Singh A, Vagne Q, Jülicher F, Sbalzarini IF. Spontaneous flow instabilities of active polar fluids in three dimensions. Phys. Rev. Res. 2023;5:L022061. URL https://link.aps.org/doi/10.1103/PhysRevResearch.5.L022061
Search in Google Scholar Back to article

Integrating Odeint Time Stepping into OpenFPM for Distributed and GPU Accelerated Numerical Solvers

Abstract

Paradigm

My account