2017 · 14 citations · 13 references
ComputingComputer EngineeringComputer ScienceCompilersComputer GraphicsParallel ComputingData ParallelismData-flow AnalysisGraphics Processing Unit ArchitectureGraphics Processing Unit ComputingGraphics Processing UnitsHardware AccelerationGpgpu ComputingGpu Compiler OptimizationGpu Runtime SystemsMicroarchitectural OptimizationParallel ReductionReductionsSegmented ReductionWorkload
We present and evaluate an implementation technique for regular segmented reductions on GPUs. Existing techniques tend to be either consistent in performance but relatively inefficient in absolute terms, or optimised for specific workloads and thereby exhibiting bad performance for certain input. We propose three different strategies for segmented reduction of regular arrays, each optimised for a particular workload. We demonstrate an implementation in the Futhark compiler that is able to employ all three strategies and automatically select the appropriate one at runtime. While our evaluation is in the context of the Futhark compiler, the implementation technique is applicable to any library or language that has a need for segmented reductions.
13
Rodinia: A benchmark suite for heterogeneous computing
Shuai Che, Michael Boyer, Jiayuan Meng et al. · 2009 · 3K citations
Scalable parallel programming with CUDA
John Nickolls, Ian Buck, Michael Garland et al. · 2008 · 1.6K citations
Scalable Parallel Programming with CUDA
John Nickolls, Ian Buck, Michael Garland et al. · Queue · 2008 · 1.6K citations · Full text
Programming parallel algorithms
Guy E. Blelloch · Communications of the ACM · 1996 · 454 citations · Full text
Accelerating Haskell array codes with multicore GPUs
Manuel M. T. Chakravarty, Gabriele Keller, Sean Lee et al. · 2011 · 245 citations