Section: Software and Platforms


Participant : Julien Diaz [correspondant] .


This software, written in FORTRAN 90, simulates the propagation of acoustic waves in heterogeneous 2D and 3D media. It is based on an Interior Penalty Discontinuous Galerkin Method (IPDGM). The 2D version of the code has been implemented in the Reverse Time Migration (RTM) software of Total  in the framework of the Ph.D thesis of Caroline Baldassari. The 2D code allows for the use of meshes composed of cells of various order (p-adaptivity in space). For the time discretization, we used the local time stepping strategy described at section  3.2 , item High-Order Schemes in Space and Time which permits not only the use of different time-step, but also to adapt the order of the time-discretization to the order of each cells (hp-adaptivity in time).

The main competitors of Hou10ni are codes based on Finite Differences, Spectral Element Method or other Discontinuous Galerkin Methods (such as the ADER schemes). During her Ph.D thesis, Caroline Baldassari compared the solution obtained by Hou10ni to the solution obtained by a Finite Difference Method and by a Spectral Element Method (SPECFEM). To evaluate the accuracy of the solutions, we have compared them to analytical solutions provided by the codes Gar6more (see below). The results of these comparisons are: a) that Hou10ni outperforms the Finite Difference Methods both in terms of accuracy and of computational burden and b) that its performances are similar to Spectral Element Methods. Since Hou10ni allows for the use of meshes based on tetrahedrons, which are more appropriate to mesh complex topographies, and for the p-adaptivity, we decided to implement it in the RTM code of Total . Of course, we also used these comparisons to validate the code.


Recently, we have extended the 2D version of Hou10ni for computing the solution of the harmonic wave equation (Helmholtz), in the framework of the PhD thesis of Élodie Estécahandy. This new version is able to deal with both acoustic and elastodynamic media, but also to model elastoacoustic problems. The surfaces between the different media can be approximated by curved elements. We can use up to P15 elements when dealing with curved elements and element of arbitrary order (with of course a limitation depending on the machine precision) when dealing with non-curved elements. The construction of the global matrix is perform using OpenMP and the extension to hybrid MPI/OpenMP parallelism is on development. This code has been also implemented in a solver which determine the shape of an elastic obstacle from the knowledge of its scattered field.

The 3D version of Hou10ni-Frequency-Domain is under development. The code is now able to solve acoustic problems up to P3 elements. It has been parallelized using MPI and is able to deal with partitioned meshes. Preliminary tests have been performed up to 16.000.000 unknowns. We are now considering the following features : hybrid MPI/OpenMP parallelism; extension to arbitrary polynomial degrees; extension to elastodynamic.