Paper Info
Title
Energy-efficient multigrid smoothers and grid transfer operators on multi-core and GPU clusters.
Abstract
We investigate time and energy to solution for the CPU- and GPU-based execution of the compute intensive smoother and grid transfer operators in a geometric multigrid linear solver. We use a hybrid parallel implementation for both shared and distributed memory multi-core host systems comprising CUDA-capable devices. Our numerical experiments are designed to assess the effect of combining an MPI-parallel multigrid framework with OpenMP host threads or CUDA accelerators instead of MPI-only CPU computations for various parallel setups. We present runtime and energy measurements from a quad-CPU test system equipped with two GPUs. We find that using an accelerated asynchronous smoother can yield substantial savings of time and energy to solution over using a host-only Jacobi smoother in small and medium sized host systems with one or two multi-core CPUs. The acceleration of the grid transfer operators also yields a benefit, yet smaller than the benefit from the smoother. For large host systems a hybrid MPI-OpenMP parallelization turns out to be most beneficial with respect to energy consumption, although it is not the fastest option. Time and energy to solution for a parallel geometric multigrid solver are investigated by means of accurate measurements.A hybrid implementation for heterogeneous platforms with multi-core host systems and GPUs is presented.Substantial savings in time and energy are observed when using GPUs on small host systems.Considerable savings in energy are observed for a host-only hybrid MPI-OpenMP setup on large host systems.Optimization for energy may lead to other parallel configurations than optimization for time.
Year
DOI
Venue
2017
10.1016/j.jpdc.2016.05.006
J. Parallel Distrib. Comput.
Keywords
Field
DocType
Energy-aware numerics,Geometric multigrid,Hybrid parallelization,Heterogeneous platforms,Performance and energy assessment
Computer science,CUDA,Efficient energy use,Parallel computing,Distributed memory,Solver,Multi-core processor,Energy consumption,Multigrid method,Grid
Journal
Volume
Issue
ISSN
100
C
0743-7315
Citations 
PageRank 
References 
0
0.34
5
Authors
2
Name
Order
Citations
PageRank
Martin Wlotzka100.34
Vincent Heuveline217930.51