5 Hits in 6.9 sec

C-SAW: A Framework for Graph Sampling and Random Walk on GPUs [article]

Santosh Pandey, Lingda Li, Adolfy Hoisie, Xiaoye S. Li, Hang Liu
2020 arXiv   pre-print
In this paper, we propose, to the best of our knowledge, the first GPU-based framework for graph sampling/random walk.  ...  First, our framework provides a generic API which allows users to implement a wide range of sampling and random walk algorithms with ease.  ...  C-SAW: A Bias-Centric Sampling Framework C-SAW offloads sampling and random walk on GPUs with the goal of a simple and expressive API and a high performance framework.  ... 
arXiv:2009.09103v1 fatcat:m2j6vv7twnhjpimnme42jsf5aa

ThunderRW: An In-Memory Graph Random Walk Engine (Complete Version) [article]

Shixuan Sun and Yuhang Chen and Shengliang Lu and Bingsheng He and Yuchen Li
2021 arXiv   pre-print
As random walk is a powerful tool in many graph processing, mining and learning applications, this paper proposes an efficient in-memory random walk engine named ThunderRW.  ...  Compared with existing parallel systems on improving the performance of a single graph operation, ThunderRW supports massive parallel random walks.  ...  We consider C-SAW [46] , the state-of-the-art RW framework on GPUs, as well.  ... 
arXiv:2107.11983v1 fatcat:wr5aoycf4bb4zmj34p7srw7d6u

BGL: GPU-Efficient GNN Training by Optimizing Graph Data I/O and Preprocessing [article]

Tianfeng Liu
2021 arXiv   pre-print
The main bottlenecks are the process of preparing data for GPUs - subgraph sampling and feature retrieving.  ...  By a co-design of caching policy and the order of sampling, we find a sweet spot of low overhead and high cache hit ratio.  ...  In subgraph sampling, random hashing partitioning leads to extensive cross-partition communication.Some works try to improve graph sampling performance on GPUs, such as NextDoor [28] and C-SAW [41] .  ... 
arXiv:2112.08541v1 fatcat:kzel63n3ircqdpcuie2d4jd7y4

Scalable Graph Neural Network Training: The Case for Sampling [article]

Marco Serafini, Hui Guan
2021 arXiv   pre-print
Graph Neural Networks (GNNs) are a new and increasingly popular family of deep neural network architectures to perform learning on graphs.  ...  Scalability is challenging with both approaches, but we make a case that research should focus on sample-based training since it is a more promising approach.  ...  This work was partially supported by a Facebook Systems for Machine Learning Award and an AWS Cloud Credit for Research grant.  ... 
arXiv:2105.02315v1 fatcat:jke6ekujpfdsdjmx5d4avkjeky

Is Parallel Programming Hard, And, If So, What Can You Do About It? (Release v2021.12.22a) [article]

Paul E. McKenney
2021 arXiv   pre-print
Nevertheless, you should think of the information in this book as a foundation on which to build, rather than as a completed cathedral.  ...  painstakingly reinvent old wheels, enabling them to instead focus their energy and creativity on new frontiers.  ...  The design for a bridge meant to allow people to walk across a small brook might be a simple as a single wooden plank.  ... 
arXiv:1701.00854v4 fatcat:pxiajyczebd5pm76htwnrczhm4