Pilot specification
Collaborating institutions
CHEESE partners: LMU, INGV
Others: Stanford University (Prof. Eric Dunham)
Associated Natural hazard
Earthquakes, tsunamis
PD4: Physics-based tsunami-earthquake interaction
| Main objective / mission | We develop a fully coupled framework for the simulation of multi-physics earthquake-tsunami scenarios, which covers physical effects from frictional fault failure, to elastic and acoustic wave propagation to the simulation of tsunami waves. In our fully coupled model (realised in the open-source software SeisSol), the Earth and the ocean are modeled as elastic and acoustic media respectively. Furthermore, a linearized gravitational free surface boundary condition (Lotto et al., 2015) enables the simulation of tsunami propagation. The resulting model captures the full 3D complexity of earthquake-induced tsunami generation. It can help to improve the general understanding of how earthquakes cause tsunamis and to increase the accuracy of initial conditions for simulating tsunami propagation. |
| Workflow description | SeisSol offers a workflow to integrate topography data, seismic velocity profiles and fault geometries into the simulation setup (e.g., via the tool easi). Scripts are provided to pipe results of the simulated tsunami generation into tsunami propagation software. |
| PD configuration | The pilot demonstrator will tackle a high-resolution model for the Hellenic Arc subduction zone in the Mediterranean. The model is currently evaluated for benchmark scenarios and for the 2018 Palu-Sulawesi earthquake-tsunami event. |
| Tested architectures | SuperMUC-NG, Shaheen-II, Mahti |
| Target TRL | 4 |
| Relevant stakeholders | Researchers in the field of tsunami generation |
| Achievements up to M41 | High-order ADER-DG discretisation of 3D elastic-acoustic models with gravity boundary condition to accurately simulate elastic (in solid Earth), acoustic and gravity waves (in the ocean) during earthquake-tsunami events. Scalability of the implementation in SeisSol evaluated up to 3072 nodes on SuperMUC-NG for a model with approx. 200 bio degrees of freedom. |
| Related work and further information | L. Krenz et al.: 3D Acoustic-Elastic Coupling with Gravity: The Dynamics of the 2018 Palu, Sulawesi Earthquake and Tsunami, SC21, http://dx.doi.org/10.1145/3458817.3476173 |
Exascale outlook
Exascale target (capability/capacity)
The currently largest setups already require petascale performance, despite their limited size of the simulated domain (local tsunami in the Palu Bay).
Larger simulated domains and higher resolution will require a further increase in the degrees of freedom.
Higher resolution is particularly required for resolving the acoustic layer in the ocean, which promises new avenues for tsunami early warning.
Extension towards parameter studies and uncertainty quantification will lead to further setups that require exascale performance.
Main benefits expected from large increase in compute resources
Higher resolution to capture a wider range of frequencies for seismic and acoustic waves.
Possibility to tackle domains with larger extent.
Parameter studies and uncertainty quantification are only possible if time to solution for individual runs is reduced by 2-3 orders of magnitude.
Main challenges in harnessing large compute resources
Full support of the elastic-acoustic model (esp. incl. dynamic rupture and boundary conditions) also in the GPU version of SeisSol, ensuring performance portability on upcoming exascale architectures.
Specific actions towards exascale at present time
Porting of SeisSol (incl. elastic-acoustic coupling with necessary boundary conditions and dynamic rupture) to GPU supercomputers.
Extension of the underlying code generator YATeTo towards various heterogeneous architectures.
Optimisation of SeisSol’s local time stepping algorithm.
Computational details
| Number of cores / GPUs: | Memory (GB): | Storage (GB) both temporal and permanent: | #files written both temporal and permanent |
I/O data traffic per hour during job |
|
| Minimum: | 6144 | 1000 | 100 | 3000 | 10 GB/h |
| Average: | 48000 | 5000 | 500 | 3000 | 30 GB/h |
| Maximum: | 147456 | 25000 | 3000 | 3000 | 120 GB/h |