Multi-hetero Acceleration by GPU and FPGA for Astrophysics Simulation on oneAPI Environment
Ryuta Kashino, Ryohei Kobayashi, Norihisa Fujita, Taisuke Boku
Abstract
GPU (Graphics Processing Unit) computing is one of the most popular accelerating methods for various high-performance computing applications. For scientific computations based on multi-physical phenomena, however, a single device solution on a GPU is insufficient, where the single timescale or degree of parallelism is not simply supported by a simple GPU-only solution. We have been researching a combination of a GPU and FPGA (Field Programmable Gate Array) for such complex physical simulations. The most challenging issue is how to program these multiple devices using a single code.
Topics & Concepts
Computer scienceGraphics processing unitField-programmable gate arrayCUDAParallel computingAccelerationGeneral-purpose computing on graphics processing unitsGraphicsComputational scienceCode (set theory)ComputationSimple (philosophy)Parallelism (grammar)Embedded systemComputer graphics (images)PhysicsAlgorithmEpistemologyPhilosophyClassical mechanicsProgramming languageSet (abstract data type)Electromagnetic Scattering and AnalysisScientific Research and DiscoveriesComputational Physics and Python Applications