Abstract. This work deals with the construction of a low-order absorbing boundary condition (ABC) for 2D elliptic TTI media, preserving the system stability. The construction is based on comparing and then connecting the slowness curves for isotropic and elliptic TTI waves. Numerical experiments illustrate the performance of the new ABC. They are performed by integrating the ABC in a DG formulation of Elastodynamics. When applied in a TTI medium, the new ABC performs well with the same level of accuracy than the standard isotropic ABC set in an isotropic medium. The condition demonstrates also a good robustness when applied for large times of simulation.Résumé. Ce travail porte sur la construction d'une condition aux limites absorbante (CLA) pour les milieux TTI elliptiques, préservant la stabilité du système. La mise en oeuvre repose sur la comparaison puis la mise en relation des courbes de lenteurs pour les ondes isotropes et TTI elliptiques. Des expériences numériques illustrent la performance de ces nouvelles CLA. Elles sont réalisées en intégrant la CLA dans une formulation de type Galerkin discontinue pour l'Élastodynamique. Pour un milieu TTI, la nouvelle CLA est efficace avec le même niveau de précision que la CLA isotrope standard pour un milieu isotrope. On observe aussi que la condition est robuste en temps long.

The level of hardware complexity of current supercomputers is forcing the High Performance Computing (HPC) community to reconsider parallel programming paradigms and standards. The high-level of hardware abstraction provided by task-based paradigms make them excellent candidates for writing portable codes that can consistently deliver high performance across a wide range of platforms. While this paradigm has proved efficient for achieving such goals for dense and sparse linear solvers, it is yet to be demonstrated that industrial parallel codes-relying on the classical Message Passing Interface (MPI) standard and that accumulate dozens of years of expertise (and countless lines of code)-may be revisited to turn them into efficient task-based programs. In this paper, we study the applicability of task-based programming in the case of a Reverse Time Migration (RTM) application for Seismic Imaging. The initial MPI-based application is turned into a task-based code executed on top of the PaRSEC runtime system. Preliminary results show that the approach is competitive with (and even potentially superior to) the original MPI code on a homogeneous multicore node, and can more efficiently exploit complex hardware such as a cache coherent Non Uniform Memory Access (ccNUMA) node or an Intel Xeon Phi accelerator.

No abstract