Award Finalist/Winner

SC Conference - Activity Details

Scalable adaptive Mantle Convection Simulation on Petascale Supercomputers

Carsten Burstedde  (University of Texas at Austin)
Omar Ghattas  (University of Texas at Austin)
Michael Gurnis  (California Institute of Technology)
Georg Stadler  (University of Texas at Austin)
Eh Tan  (California Institute of Technology)
Tiankai Tu  (University of Texas at Austin)
Lucas Wilcox  (University of Texas at Austin)
Shijie Zhong  (University of Colorado)
ACM Gordon Bell Finalists Session
Wednesday,  04:30PM - 05:00PM
Room Ballroom G
Mantle convection is the principal control on the thermal and geological evolution of the Earth. Mantle convection modeling involves solution of the mass, momentum, and energy equations for a viscous, creeping, incompressible non-Newtonian fluid at high Rayleigh and Peclet numbers. Our goal is to conduct global mantle convection simulations that can resolve faulted plate boundaries, down to 1 km scales. Uniform resolution leads to trillion element meshes, which are intractable even on petascale supercomputers. Thus parallel mesh adaptivity is essential. We present Rhea, a new generation mantle convection code designed to scale to hundreds of thousands of cores. Rhea is built on ALPS, a parallel octree-based adaptive finite element library that supports new distributed data structures and parallel algorithms for dynamic coarsening, refinement, rebalancing, and repartitioning of the mesh. Using TACC's 580 Teraflops Ranger system, we demonstrate excellent weak and strong scalability on problems with O(10^10) unknowns.
The full paper can be found in the IEEE Xplore Digital Library and ACM Digital Library
   IEEE Computer Society  /  ACM     2 0   Y E A R S   -   U N L E A S H I N G   T H E   P O W E R   O F   H P C