RBHK_LNCS_2009

Wavelet based adaptive simulation of complex systems on multicore architectures


Rossinelli D., Bergdorf M., Hejazialhosseini B., Koumoutsakos P., Lecture Notes In Computer Science, 5704, 10, 721-734

ABSTRACT

We build wavelet-based adaptive numerical methods for the simulation of  advection dominated flows  that develop multiple spatial scales, with an emphasis on fluid mechanics problems.
Wavelet based adaptivity is inherently sequential and in this work we demonstrate that these  numerical methods can be implemented in software that is  capable of harnessing
the capabilities of multi-core architectures while maintaining their computational efficiency. Recent designs in frameworks for multi-core software development allow us  to rethink parallelism as task-based, where parallel tasks are specified and automatically mapped into physical threads. This  way of exposing parallelism enables the parallelization of algorithms that were considered inherently sequential, such as wavelet-based adaptive simulations. In this paper we present  a framework  that combines wavelet-based adaptivity with the task-based parallelism. We demonstrate good scaling performance obtained by simulating diverse physical systems on different multi-core and SMP architectures using up to 16 cores.

PDF