High-Performance GPU-to-CPU Transpilation and Optimization via High-Level Parallel Constructs
William S. Moses, Ivan R. Ivanov, Jens Domke, Toshio Endo, Johannes Doerfert, Oleksandr Zinenko
Abstract
While parallelism remains the main source of performance, architectural implementations and programming models change with each new hardware generation, often leading to costly application re-engineering. Most tools for performance portability require manual and costly application porting to yet another programming model.
Topics & Concepts
PortingSoftware portabilityComputer scienceParallelism (grammar)Parallel computingImplementationProgramming paradigmComputer architectureData parallelismParallel programming modelCUDAOperating systemSoftwareProgramming languageParallel Computing and Optimization TechniquesDistributed and Parallel Computing SystemsAdvanced Data Storage Technologies