Fast Gunrock Subgraph Matching (GSM) on GPUs

Wang, Leyuan; Owens, John D.

doi:10.48550/arxiv.2003.01527

Cited by 2 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generalized Sub-graph Matching. Many works perform generalized sub-graph matching on the CPU [1,17,50,54,66] and the GPU [9,16,28,41,57,63,64,73]. These frameworks search for an arbitrary 𝑘-vertex sub-graph and support different values of 𝑘.…”

Section: Related Workmentioning

confidence: 99%

“…The limited GPU memory capacity can severely limit parallelism because there may not be sufficient memory for tracking the execution state of the large number of threads that traverse search trees and subtrees in parallel. While there has been work on solving related problems on GPUs, such as finding maximal cliques [30,34,39,59,67,70] and generalized sub-graph matching [9,28,63], little attention has been given to 𝑘-clique counting in particular. To the best of our knowledge, there are no performant parallel solutions specialized for 𝑘-clique counting on GPUs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Parallel K-clique counting on GPUs

Al-Masri

Hajj

Nagi

et al. 2022

Proceedings of the 36th ACM International Conference on Supercomputing

View full text Add to dashboard Cite

Counting 𝑘-cliques in a graph is an important problem in graph analysis with many applications such as community detection and graph partitioning. Counting 𝑘-cliques is typically done by traversing search trees starting at each vertex in the graph. Parallelizing 𝑘-clique counting has been well-studied on CPUs and many solutions exist. However, there are no performant solutions for 𝑘-clique counting on GPUs.Parallelizing 𝑘-clique counting on GPUs comes with numerous challenges such as the need for extracting fine-grain multi-level parallelism, sensitivity to load imbalance, and constrained physical memory capacity. While there has been work on related problems such as finding maximal cliques and generalized sub-graph matching on GPUs, 𝑘-clique counting in particular has yet to be explored in depth. In this paper, we present the first parallel GPU solution specialized for the 𝑘-clique counting problem. Our solution supports both graph orientation and pivoting for eliminating redundant clique discovery. It incorporates both vertex-centric and edge-centric parallelization schemes for distributing work across thread blocks, and further partitions work within each thread block to extract fine-grain multi-level parallelism while tolerating load imbalance. It also includes optimizations such as binary encoding of induced sub-graphs and sub-warp partitioning to limit memory consumption and improve the utilization of execution resources.Our evaluation shows that our best GPU implementation outperforms the best state-of-the-art parallel CPU implementation by a geometric mean of 12.39×, 6.21×, and 18.99× for 𝑘 = 4, 7, and 10, respectively. We also perform a detailed evaluation of the trade-offs involved in the choice of parallelization scheme, and the incremental speedup of each optimization to provide an in-depth understanding of the optimization space. The insights from our optimization flow can be useful for optimizing other clique finding and graph mining solutions on GPUs. Our code will be open-sourced to enable further research on GPU parallelization of 𝑘-clique counting and other similar graph mining algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Parallel K-clique counting on GPUs

Al-Masri

Hajj

Nagi

et al. 2022

Proceedings of the 36th ACM International Conference on Supercomputing

View full text Add to dashboard Cite

show abstract

“…Hand-optimized GPM Applications. Numerous hand optimized GPM applications have been developed for various platforms, including triangle counting [33,40,48,51,81,82,96,100,111,113,118], k-clique listing [29,31] and counting [5,19,19,53,94], motif counting [3,68,71,83,91,97], subgraph listing/matching [13,14,17,47,55,60,61,66,69,73,86,87,93,98,99,105,107,110], and FSM [1,36,59,101,102,104,108,…”

Section: Related Workmentioning

confidence: 99%

Efficient and Scalable Graph Pattern Mining on GPUs

Chen¹,

Arvind²

2021

Preprint

View full text Add to dashboard Cite

We describe G 2 Miner, the first Graph Pattern Mining (GPM) framework that runs on multiple GPUs. G 2 Miner uses pattern-aware, input-aware and architecture-aware search strategies to achieve high efficiency on GPUs. To simplify programming, it provides a code generator that automatically generates pattern-aware CUDA code. G 2 Miner flexibly supports both breadth-first search (BFS) and depth-first search (DFS) to maximize memory utilization and generate sufficient parallelism for GPUs. For the scalability of G 2 Miner, we propose a customized scheduling policy to balance work among multiple GPUs. Experiments on a V100 GPU show that G 2 Miner is 5.4× and 7.2× faster than the two state-ofthe-art single-GPU systems, Pangolin and PBE, respectively. In the multi-GPU setting, G 2 Miner achieves linear speedups from 1 to 8 GPUs, for various patterns and data graphs. We also show that G 2 Miner on a V100 GPU is 48.3× and 15.2× faster than the state-of-the-art CPU-based systems, Peregrine and GraphZero, on a 56-core CPU machine.

show abstract

Fast Gunrock Subgraph Matching (GSM) on GPUs

Cited by 2 publications

References 16 publications

Parallel K-clique counting on GPUs

Parallel K-clique counting on GPUs

Efficient and Scalable Graph Pattern Mining on GPUs

Contact Info

Product

Resources

About