Multi-GPU Graph Analytics
release_7cr2eivgbbadzbedc6mpl7cs7y
by
Yuechao Pan, Yangzihao Wang, Yuduo Wu, Carl Yang, John D. Owens
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) |
1504.04804v1
access all versions, variants, and formats of this works (eg, pre-prints)