Toward a Multi-level Parallel Framework on GPU Cluster with PetSC-CUDA for PDE-based Optical Flow Computation
Copyright policy )Toward a Multi-level Parallel Framework on GPU Cluster with PetSC-CUDA for PDE-based Optical Flow Computation
AbstractIn this work we present a multi-level parallel framework for the Optical Flow computation on a GPUs cluster, equipped with a scientific computing middleware (the PetSc library). Starting from a flow-driven isotropic method, which models the optical flow problem through a parabolic partial differential equation (PDE), we have designed a parallel algorithm and its software implementation that is suitable for heterogeneous computing environments (multiprocessor, single GPU and cluster of GPUs). The proposed software has been tested on real SAR images sequences. Numerical experiments highlight the performance of the proposed software framework, which can reach a gain of about 95% with respect to the sequential implementation.
Differential equations, Middleware, Numerical solution of PDEs, Numerical solution, Optical flow, Mulitilevel parallel computing, Sequential implementation, Synthetic aperture radar, Software implementation, Computer programming; Differential equations; Middleware; Optical flows; Program processors; Software testing; Synthetic aperture radar, GP-GPU; Heterogeneous computing; Numerical experiments; Numerical solution; Optical flow computation; Parabolic partial differential equations; Sequential implementation; Software implementation, Cluster computing; GP-GPU; Mulitilevel parallel computing; Numerical solution of PDEs; Optical Flow; Scientific computing libraries, Computer programming, Program processors, GP-GPU, Software testing, Optical flows, Parabolic partial differential equations, Optical Flow, Optical flow computation, Scientific computing libraries, Heterogeneous computing, Cluster computing, Numerical experiments
Differential equations, Middleware, Numerical solution of PDEs, Numerical solution, Optical flow, Mulitilevel parallel computing, Sequential implementation, Synthetic aperture radar, Software implementation, Computer programming; Differential equations; Middleware; Optical flows; Program processors; Software testing; Synthetic aperture radar, GP-GPU; Heterogeneous computing; Numerical experiments; Numerical solution; Optical flow computation; Parabolic partial differential equations; Sequential implementation; Software implementation, Cluster computing; GP-GPU; Mulitilevel parallel computing; Numerical solution of PDEs; Optical Flow; Scientific computing libraries, Computer programming, Program processors, GP-GPU, Software testing, Optical flows, Parabolic partial differential equations, Optical Flow, Optical flow computation, Scientific computing libraries, Heterogeneous computing, Cluster computing, Numerical experiments
selected citations These citations are derived from selected sources.This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).14 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Average influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Top 10% impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Top 10%
Citations provided by BIP!Popularity provided by BIP!