Parallel graph algorithms by blocks Article Swipe

View

Related Concepts

Computer science Graph algorithms Implementation Graph Parallel computing Theoretical computer science Analytics Wait-for graph Out-of-core algorithm Big data Graph database Algorithm Programming language Database Data mining

Abdurrahman Yaşar , Kasimir Gabert , Ümit V. Çatalyürek ·

YOU? · · 2021 · Open Access · · DOI: https://doi.org/10.1145/3457388.3459987 · OA: W3157939047

In this poster presentation we briefly explain our generalized algorithmic framework for parallel block-based graph algorithms, called PGAbB. PGAbB proposes that block-based graph algorithms offer a sweet spot between efficient parallelism and architecture agnostic algorithm design for a wide class of graph problems that performs close to hand-optimized HPC implementations. Picking triangle counting problem as the use-case we explain different components of our algorithmic framework for shared-memory, multi-core, multi-GPU machines and show how that approach outperform state-of-the art implementations.