English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Open

Presents the pdf analysis as premium feature

PDF Analysis

Insights

Presents the summarisation as premium feature

Summarisation

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Finite element schemes based on discontinuous Galerkin methods possessfeatures amenable to massively parallel computing accelerated with generalpurpose graphics processing units (GPUs). However, the computationalperformance of such schemes strongly depends on their implementation. In thepast, several implementation strategies have been proposed. They are basedexclusively on specialized compute kernels tuned for each operation, or theycan leverage BLAS libraries that provide optimized routines for basic linearalgebra operations. In this paper, we present and analyze up-to-dateperformance results for different implementations, tested in a unifiedframework on a single NVIDIA GTX980 GPU. We show that specialized kernelswritten with a one-node-per-thread strategy are competitive for polynomialbases up to the fifth and seventh degrees for acoustic and elastic models,respectively. For higher degrees, a strategy that makes use of the NVIDIAcuBLAS library provides better results, able to reach a net arithmeticthroughput 35.7% of the theoretical peak value.

GPU performance analysis of a nodal discontinuous Galerkin method for acoustic and elastic models