Improving Uintah's Scalability Through the Use of Portable Kokkos-Based Data Parallel Tasks
- Univ. of Utah, Salt Lake City, UT (United States)
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
The University of Utah’s Carbon Capture Multidisciplinary Simulation Center (CCMSC) is using the Uintah Computational Framework to predict performance of a 1000 MWe ultra-supercritical clean coal boiler. The center aims to utilize the Intel Xeon Phi-based DOE systems, Theta and Aurora, through the Aurora Early Science Program by using the Kokkos C++ library to enable node-level performance portability. This paper describes infrastructure advancements and portability improvements made possible by the integration of Kokkos within Uintah. This integration marks a step towards consolidating Uintah’s MPI+PThreads and MPI+CUDA hybrid parallelism approaches into a single MPI+Kokkos approach. Scalability results are presented that compare serial and data parallel task execution models for a challenging radiative heat transfer calculation, central to the center’s predictive boiler simulations. Here, these results demonstrate both good strong-scaling characteristics to 256 Knights Landing (KNL) processors on the NSF Stampede system, and show the KNL-based calculation to compete with prior GPU-based results for the same calculation.
- Research Organization:
- Univ. of Utah, Salt Lake City, UT (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- DOE Contract Number:
- NA0002375
- OSTI ID:
- 1582430
- Resource Relation:
- Conference: PEARC17: Practice and Experience in Advanced Research Computing 2017 on Sustainability, Success and Impact, New Orleans, LA (United States), Jul 2017
- Country of Publication:
- United States
- Language:
- English
Dynamic task scheduling for the Uintah framework
|
conference | November 2010 |
An Evaluation of Emerging Many-Core Parallel Programming Models
|
conference | January 2016 |
Investigating applications portability with the Uintah DAG-based runtime system on PetaScale supercomputers
|
conference | January 2013 |
Extending the Uintah Framework through the Petascale Modeling of Detonation in Arrays of High Explosive Devices
|
journal | January 2016 |
Spatial Domain-Based Parallelism in Large-Scale, Participating-Media, Radiative Transport Applications
|
journal | June 1997 |
Kokkos: Enabling manycore performance portability through polymorphic memory access patterns
|
journal | December 2014 |
Parallelization of the P-1 Radiation Model
|
journal | January 2006 |
A survey of high level frameworks in block-structured adaptive mesh refinement packages
|
journal | December 2014 |
Using hybrid parallelism to improve memory use in the Uintah framework
|
conference | January 2011 |
Large Eddy Simulations of Accidental Fires Using Massively Parallel Computers
|
conference | June 2012 |
Legion: Expressing locality and independence with logical regions
|
conference | November 2012 |
Efficient Parallelization of RMCRT for Large Scale LES Combustion Simulations
|
conference | June 2012 |
Fast Parallel Algorithms for Short-Range Molecular Dynamics
|
journal | March 1995 |
CHARM++: a portable concurrent object oriented system based on C++
|
conference | January 1993 |
Parallel Computations of Radiative heat Transfer Using the Discrete Ordinates Method
|
journal | December 2004 |
Uintah: a scalable framework for hazard analysis
|
conference | January 2010 |
Preliminary experiences with the uintah framework on Intel Xeon Phi and stampede
|
conference | January 2013 |
Radiation modeling using the Uintah heterogeneous CPU/GPU runtime system
|
conference | January 2012 |
An overview of the Trilinos project
|
journal | September 2005 |
Radiative Heat Transfer Calculation on 16384 GPUs Using a Reverse Monte Carlo Ray Tracing Approach with Adaptive Mesh Refinement
|
conference | May 2016 |
An Overview of Performance Portability in the Uintah Runtime System through the Use of Kokkos
|
conference | November 2016 |
Similar Records
Optimizing the hypre solver for manycore and GPU architectures
Demonstrating GPU code portability and scalability for radiative heat transfer computations