arXiv Open Access 2010

A Multi-Stage CUDA Kernel for Floyd-Warshall

Ben Lund Justin W Smith

Lihat Sumber

Abstrak

We present a new implementation of the Floyd-Warshall All-Pairs Shortest Paths algorithm on CUDA. Our algorithm runs approximately 5 times faster than the previously best reported algorithm. In order to achieve this speedup, we applied a new technique to reduce usage of on-chip shared memory and allow the CUDA scheduler to more effectively hide instruction latency.

Topik & Kata Kunci

cs.DC cs.PF

Penulis (2)

Ben Lund

Justin W Smith

Format Sitasi

APA MLA BibTeX

Lund, B., Smith, J.W. (2010). A Multi-Stage CUDA Kernel for Floyd-Warshall. https://arxiv.org/abs/1001.4108

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2010
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓