Multi-GPU Graph Analytics release_7cr2eivgbbadzbedc6mpl7cs7y

by Yuechao Pan, Yangzihao Wang, Yuduo Wu, Carl Yang, John D. Owens

Released as a article .

2015  

Abstract

We present a single-node, multi-GPU programmable graph processing library that allows programmers to easily extend single-GPU graph algorithms to achieve scalable performance on large graphs with billions of edges. Directly using the single-GPU implementations, our design only requires programmers to specify a few algorithm-dependent concerns, hiding most multi-GPU related implementation details. We analyze the theoretical and practical limits to scalability in the context of varying graph primitives and datasets. We describe several optimizations, such as direction optimizing traversal, and a just-enough memory allocation scheme, for better performance and smaller memory consumption. Compared to previous work, we achieve best-of-class performance across operations and datasets, including excellent strong and weak scalability on most primitives as we increase the number of GPUs in the system.
In text/plain format

Archived Files and Locations

application/pdf  525.1 kB
file_hvndpebyh5fmldnn5axnxx5fve
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2015-04-19
Version   v1
Language   en ?
arXiv  1504.04804v1
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: d75cb73f-52e6-4aee-b869-e2858371db21
API URL: JSON