
==== Front
Data Brief
Data Brief
Data in Brief
2352-3409
Elsevier

S2352-3409(24)00851-5
10.1016/j.dib.2024.110888
110888
Data Article
Flow field data of three-dimensional Riemann problems
Hoppe Nils a
Fleischmann Nico a
Biller Benedikt benedikt.biller@tum.de
a⁎
Adami Stefan b
Adams Nikolaus A. a
a Chair of Aerodynamics and Fluid Mechanics, Technical University of Munich, Boltzmannstr. 15, 85748 Garching, Germany
b Munich Institute of Integrated Materials, Energy and Process Engineering, Technical, University of Munich, Lichtenbergstr. 4a, 85748 Garching, Germany
⁎ Corresponding author. benedikt.biller@tum.de
05 9 2024
12 2024
05 9 2024
57 11088810 5 2024
12 7 2024
26 8 2024
© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by-nc/4.0/ This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Common validation and verification test cases for compressible flow solvers are only one- or two-dimensional. Such flows, however, are inherently three-dimensional. The provided data contains simulation results of genuine three-dimensional Riemann problems computed with the open-source compressible flow solver ALPACA. The problems are designed so that each octant's constant initial state connects two neighboring states by only one elementary wave each. Thereby, initial conditions are chosen to induce three-dimensional effects. Furthermore, the cases are designed to trigger common shortcomings of compressible flow solvers, such as spurious pressure oscillations, unphysical symmetry breaking, or the onset of shock disturbances. The cases were simulated using a finite-volume scheme with HLLC and Roe Riemann solvers and fifth-order WENO reconstruction. The simulations were conducted on over 300 cores of a compute cluster. Besides the raw binary flow field data, input files are provided next to post-processing scripts and the visualizations obtained by them. The provided files ease setting up and simulating the respective cases with different solvers and allow quantitative comparisons of the obtained results.

Keywords

Gas dynamics
Compressible flow
High-order methods
High-resolution
==== Body
pmcSpecifications TableSubject	Computational Mechanics	
Specific subject area	Computational fluid dynamics (CFD) of compressible gases	
Type of data	binary, input file, result file, processing file, image
raw, processed	
Data collection	The data were collected via direct numerical simulations using the open-source compressible flow solver ALPACA. The simulations were conducted on the CoolMUC-2 Linux Cluster of the Leibniz Supercomputing Centre.
The raw data was postprocessed using the open-source software Paraview running on the same Linux Cluster.	
Data source location	Leibniz-Rechenzentrum (LRZ) der Bayerischen Akademie der Wissenschaften
Boltzmannstraße 1
D-85748 Garching bei München	
Data accessibility	Repository name: Flowfield Data of Three-dimensional Riemann Problems [1]
Data identification number: 1661329
Direct URL to data: https://mediatum.ub.tum.de/1459250?sortfield0=date_publication&sortfield1=&show_id=1661329
The postprocessed data can also be found in a dedicated git repository:
https://gitlab.lrz.de/nanoshock/riemann_cubes	
Related research article	N. Hoppe, N. Fleischmann, B. Biller, S. Adami, N. A. Adams, A Systematic Analysis of Three-dimensional Riemann Problems for Verification of Compressible-flow Solvers. [2]	

1 Value of the Data

• The steady growth of available computing power leads to the development of increasingly more advanced high-performance computing (HPC) frameworks that can conduct fully three-dimensional simulations [2]. However, current code verifications in the literature rely on canonical one-dimensional (1D) and two-dimensional (2D) test cases, e.g., the Sod shock tube [3] in 1D or implosion and explosion cases in 2D [4]. This data set expands the 2D Riemann problem definition of [[5], [6], [7], [8]] to three dimensions (3D). Hence, it introduces a new class of inherently 3D problems for modern flow solver verification.

• Providing inherently 3D verification data, this data set enhances the development of modern flow solvers. By comparing results from new CFD codes with this data set, developers can verify them against 3D cases instead of extrapolating the 3D performance from 1D or 2D test cases. Furthermore, quantitative comparisons of the resulting flow fields can be conducted in contrast to the qualitative plot comparisons, which the related publication [2] allows for. For example, the data set enables the extraction of the involved wave speeds for quantitative code-to-code comparisons.

• During the compilation of this data set, great care was taken to foster its reusability by other researchers. The provided input files contain the initial conditions for each case so that they can be computed with different flow solvers. Supplying the raw data files enables other researchers to postprocess them individually. The provided images allow for a fast comparison with results from other CFD codes.

2 Background

Numerical simulation of compressible flow requires solving the Euler equations, which are the hyperbolic part of the Navier-Stokes equations. Despite advancements, the substantial computational resources required by solvers for such flow scenarios often confine investigations to 2D domains, thereby neglecting fundamental 3D dynamics in flow evolution. Consequently, validation and verification of compressible flow solvers predominantly rely on 2D test cases. To establish a comprehensive 3D benchmark for flow solver assessment, the related publication [2] presents a collection of three-dimensional Riemann problems. The present dataset includes source code, input parameters, and solution data for the reference problems stated in [2]. Thereby, it facilitates developers in reproducing and benchmarking against results obtained from alternative flow solvers.

3 Data Description

The dataset was obtained by numerical investigation of 3D Riemann problems as described in the related publication [2]. The cases are set up in a cubic domain which is split into octants, as illustrated in Fig. 1(a). The initial pressure, density, and velocity values in each octant are chosen such that only one elementary wave develops at each octant's face, which we denote Wij. The order of the (i,j) pairs is chosen such that a left-running wave moves towards the origin of the respective axis. Each wave becomes either a right- or left-running rarefaction fan, a left- or right-running shock wave, or a contact discontinuity. Note that we call the third wave type a contact discontinuity, although it may actually be a slip line, c.f. [2] for details.Fig. 1 Schematic of the computational domain (a) and the waves developing between the octants (b). Reproduced from [2].

Fig 1:

The individual cases are uniquely described by the type and position of each of the respective waves. The data in the repository follows this naming convention. The data for each case is stored in a directory with the respective name, where Rl, Rr, Sl, Sr, and J stand for left- and right-running rarefactions, left- and right-running shocks, and contact discontinuity, respectively.

Each 3D case comprises six 2D cases recovered at the faces of the cube. For this reason, all possible 2D cases were also computed stand-alone and are a part of the data set. Fig. 1(b) illustrates the recovery of 2D problems on the sides of the unit cube.

As described in the next section, we use the Harten-Lax-van Leer approximate Riemann solver with contact restoration (HLLC) [9] for most cases. Some of the results, however, are additionally computed with different Riemann solvers such as the Roe Riemann solver [10], the Roe-M Riemann solver [11], or the Harten-Lax-van Leer approximate Riemann solver (HLL) [12].

The raw data files contain the density fields ([ρ]=kg·m−3), the pressure fields ([p]=Pa), and the velocity fields ([u]=m·s−1) in h5 format. They can be opened with common data visualization programs such as ParaView.Folder	Content	
	
ThreeDimensionalRiemannProblems	Data of the 3D Riemann problems computed using the HLLC Riemann solver.	
ThreeDimensionalRiemannProblems/Data	Contains the raw simulation data of the 3D cases. The structure for an exemplary 3D case is shown below.	
ThreeDimensionalRiemannProblems/Data
/six_allleft_s	Case folder.	
ThreeDimensionalRiemannProblems/Data /six_allleft_s/six_allelft_s.log	Solver output file.	
ThreeDimensionalRiemannProblems/Data /six_alleft_s/domain	Directory containing the raw simulation data of this case.	
ThreeDimensionalRiemannProblems/Data
/Executables	Executables used to compute the 3D data.	
ThreeDimensionalRiemannProblems/Data
/Executables/BuildInstructions	Compiler and linker information to build the executables.	
ThreeDimensionalRiemannProblems/Images
/FieldPlots	Compressed visualizations of the temporal evolution of the flow.
Depicted are the density field and density contours.	
ThreeDimensionalRiemannProblems/Images
/SelfSimilarMachNumberPlots	Compressed visualizations of the temporal evolution of self similar Mach number contours.	
ThreeDimensionalRiemannProblems/Inputfiles	Input files.	
ThreeDimensionalRiemannProblems/Jobfiles	Job files.	
ThreeDimensionalRiemannProblems
/PressureDensityRange	Pressure and density ranges used for the images.	
ThreeDimensionalRiemannProblems/Timesteps	List of time steps for all cases.	
TwoDimensionalRiemannProblems	Data of the 2D Riemann problems computed using the HLLC Riemann solver.
The structure is identical to the ThreeDimensionalRiemannProblems directory.	
DifferentRiemannSolverCases	Data of 3D Riemann problems recomputed with different Riemann solvers.	
DifferentRiemannSolverCases
/FourOpposingSTwoAllJCarbuncleRoe	Case recomputed with the Roe Riemann solver.	
DifferentRiemannSolverCases
/FourOpposingSTwoAllJCarbuncleRoeM	Case recomputed with the Roe-M Riemann solver.	
DifferentRiemannSolverCases/SixAllJAllClockwiseHll	Case recomputed with the HLL Riemann solver.	
PostProcessing	All files necessary to reproduce the post-processed data.	
PostProcessing/Jobfiles	Job files.	
PostProcessing/Scripts	Python scripts to create the images.	
riemann_cubes-master.zip	Source code of the dedicated git repository.	

4 Experimental Design, Materials and Methods

The compressible Euler equations were solved on a Cartesian grid to create this data set. The equations are closed with an isentropic equation of state (EOS) or the stiffened gas EOS, c.f. [2] for more details. For an appropriate treatment of discontinuities in the flow field, such as shock waves and slip lines, we use a weighted essentially non-oscillatory (WENO) reconstruction stencil of fifth order [13,14] together with the HLLC Riemann solver [9]. To highlight effects such as grid-aligned shock instabilities [15], we additionally apply the Roe [10], Roe-M [11], or HLL Riemann solver [12] for some cases. For the temporal integration of the equations, we employ a third-order total variation diminishing (TVD) Runge-Kutta (RK) scheme [16]. We apply zero-gradient conditions on all domain boundaries.

We use the open-source compressible flow solver ALPACA [17] for our computations, which implements the above-mentioned high-resolution methods. Its inherent multiresolution (MR) compression allows for an effective resolution of 1024×1024×1024 finite volume cells in all our 3D cases. Furthermore, an adaptive local time stepping (ALTS) scheme adapts the time step to the local cell size during each RK sub-stage [18].

The simulations have been conducted on up to 12 nodes of the Cool-MUC2 Cluster at the LRZ. Each node holds a 28-way Intel Xeon E5-2697 v3 codename “Haswell” processor with FDR14 Infiniband interconnect1,2.

A dedicated git repository3 contains step-by-step instructions for the reproduction of the data set. In short, these are the installation of the ALPACA software, applying the correct settings, running the simulations, and postprocessing the results, see Fig. 2. The repository provides guidelines for the installation of the ALPACA software and patch files to apply the correct settings. Then, ALPACA can be run with the input files from the data set. Also, the git repository includes instructions and scripts for postprocessing the raw data files.Fig. 2 Flow chart illustrating the process to reproduce the data set.

Fig. 2:

Limitations

Not applicable.

Ethics Statement

The authors have read and follow the ethical requirements for publication in Data in Brief. The current work does nor involve human subjects, animal experiments, or any data collected from social media platforms.

CRediT authorship contribution statement

Nils Hoppe: Conceptualization, Methodology, Software, Investigation, Data curation, Writing – original draft. Nico Fleischmann: Conceptualization, Writing – review & editing. Benedikt Biller: Software, Investigation, Data curation, Writing – review & editing. Stefan Adami: Writing – review & editing, Supervision, Project administration. Nikolaus A. Adams: Writing – review & editing, Supervision, Funding acquisition.

Data Availability

Flowfield Data of Three-dimensional Riemann Problems (Original data) (mediaTUM).

Acknowledgments

The authors have received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No. 667483 ). The first author would like to thank the Federal Government and the Heads of Government of the Länder, as well as the Joint Science Conference (GWK), for their funding and support within the framework of the NFDI4Ing consortium. Funded by the German Research Foundation (DFG) - project number 442146713 . The authors gratefully acknowledge the Gauss Centre for Supercomputing e.V. for funding this project by providing computing time on the CoolMUC-2 Linux-Cluster at Leibniz Supercomputing Centre.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

1 https://doku.lrz.de/display/PUBLIC/CoolMUC-2.

2 https://www.lrz.de/services/compute/linux-cluster/overview.

3 https://gitlab.lrz.de/nanoshock/riemann_cubes.
==== Refs
References

1 N. Hoppe, N. Fleischmann, B. Biller, S. Adami, and N. A. Adams, “Flowfield data of three-dimensional riemann problems.” 10.14459/2022mp1661329.
2 N. Hoppe, N. Fleischmann, B. Biller, S. Adami, and N. A. Adams, “A systematic analysis of three-dimensional riemann problems for verification of compressible-flow solvers”.
3 Sod G.A. A survey of several finite difference methods for systems of nonlinear hyperbolic conservation laws J Comput Phys 27 1 1978 1 31 10.1016/0021-9991(78)90023-2
4 Hui W.H. Li P.Y. Li Z.W. A unified coordinate system for solving the two-dimensional euler equations J Comput Phys 153 2 1999 596 637 10.1006/jcph.1999.6295
5 Zhang T. Zheng Y.X. Conjecture on the structure of solutions of the riemann problem for two-dimensional gas dynamics systems SIAM J Math Anal 21 3 May 1990 593 630 10.1137/0521032
6 Schulz-Rinne C.W. Classification of the Riemann problem for two-dimensional gas dynamics SIAM J Math Anal 24 1 1993 76 88 10.1137/0524006
7 Chang T. Chen G.-Q. Yang S. On the 2-D Riemann problem for the compressible Euler equations I. Interaction of shocks and rarefaction waves Discr Conti Dyn Syst - A 1 4 1995 555 584 10.3934/dcds.1995.1.555
8 Lax P.D. Liu X.-D. Solution of two-dimensional riemann problems of gas dynamics by positive schemes SIAM J Sci Comput 19 2 1998 319 340 10.1137/S1064827595291819
9 Toro E.F. Spruce M. Speares W. Restoration of the contact surface in the HLL-Riemann solver Shock Waves 4 1 1994 25 34 10.1007/BF01414629
10 Roe P.L. Approximate Riemann solvers, parameter vectors, and difference schemes J Comput Phys 43 2 1981 357 372 10.1016/0021-9991(81)90128-5
11 Fleischmann N. Adami S. Hu X.Y. Adams N.A. A low dissipation method to cure the grid-aligned shock instability J Comput Phys 401 2020 109004 10.1016/j.jcp.2019.109004
12 Harten A. Lax P.D. van Leer B. On Upstream Differencing and godunov-type schemes for hyperbolic conservation laws SIAM Rev 25 1 1983 35 61 10.1137/1025002
13 Jiang G.-S. Shu C.-W. Efficient implementation of weighted ENO schemes J Comput Phys 126 1 1996 202 228 10.1006/jcph.1996.0130
14 Liu X.-D. Osher S. Chan T. Weighted essentially non-oscillatory schemes J Comput Phys 115 1 1994 200 212 10.1006/jcph.1994.1187
15 Quirk J.J. A contribution to the great riemann solver debate Upwind and High-Resolution Schemes 1997 Springer Berlin Heidelberg Berlin, Heidelberg 550 569 10.1007/978-3-642-60543-7_22
16 Harten A. High resolution schemes for hyperbolic conservation laws J Comput Phys 49 3 1983 357 393 10.1016/0021-9991(83)90136-5
17 Hoppe N. Winter J.M. Adami S. Adams N.A. ALPACA - a level-set based sharp-interface multiresolution solver for conservation laws Comput Phys Commun 272 2022 108246 10.1016/j.cpc.2021.108246
18 Kaiser J.W.J. Hoppe N. Adami S. Adams N.A. An adaptive local time-stepping scheme for multiresolution simulations of hyperbolic conservation laws J Comput Phy: X 4 2019 100038 10.1016/j.jcpx.2019.100038
