Speakers
Description
Realistic ultrasound simulations are becoming integral part of many novel medical procedures such as photoacoustic screening and non-invasive treatment planning. The common denominator of all these applications is the need for cheap and relatively large-scale ultrasound simulations with sufficient accuracy. Typical medical applications require full-wave simulations which take frequency-dependent absorption and non-linearity into account.
Our poster investigates performance of k-Wave acoustic simulation toolbox across evolving IT4Innovations supercomputer infrastructure. Our primary focus is on new clusters equipped with multi-GPU nodes – Barbora and Karolina. The behavior of k-Wave code should be useful in performance predictions for broader class of local Fourier basis pseudo-spectral codes. Our results from these new clusters are compared to other accelerated machines such as Salomon and Piz Daint and dense many-GPU node DGX-2. Finally, these results are put into the context of medical applications utilizing k-Wave toolbox.
