2019 · 27 citations · 4 references
ComputingComputer ScienceHigh-performance ComputingComputational Fluid DynamicsFluid DynamicsScientific ComputingComputational AerodynamicsData ParallelismGraphics Processing Unit ComputingGraphics Processing UnitsParallel Numerical Linear AlgebraParallel Numerical MethodsCfd SimulationGpu ArchitectureGpu Parallel AlgorithmsKernel OptimizationMemory Traffic ReductionScientific Computing On GpusSteady FlowUnsteady Flow
This paper presents a new mixed-precision implementation of a linear-solver kernel used in practical large-scale CFD simulations to improve GPU performance. The new implementation reduces memory traffic by using the half-precision format for some critical computations while maintaining double-precision solution accuracy. As the linear-solver kernel is memory bound on GPUs, a reduction in memory traffic directly translates to improved performance. The performance of the new implementation is assessed for a benchmark steady flow simulation and a large-scale unsteady turbulent flow application. Both studies were conducted using NVIDIAÆ Tesla V100 GPUs on the Summit system at the Oak Ridge Leadership Computing Facility.
4