Parallel Algorithms for Finding Large Cliques in Sparse Graphs

@article{Gianinazzi2021ParallelAF,
  title={Parallel Algorithms for Finding Large Cliques in Sparse Graphs},
  author={Lukas Gianinazzi and Maciej Besta and Yannick Schaffner and Torsten Hoefler},
  journal={Proceedings of the 33rd ACM Symposium on Parallelism in Algorithms and Architectures},
  year={2021}
}
We present a parallel k-clique listing algorithm with improved work bounds (for the same depth) in sparse graphs with low degeneracy or arboricity. We achieve this by introducing and analyzing a new pruning criterion for a backtracking search. Our algorithm has better asymptotic performance, especially for larger cliques (when k is not constant), where we avoid the straightforwardly exponential runtime growth with respect to the clique size. In particular, for cliques that are a constant factor… Expand
1 Citations

Figures and Tables from this paper

SISA: Set-Centric Instruction Set Architecture for Graph Mining on Processing-in-Memory Systems
TLDR
The key design idea is to alleviate the bandwidth needs of SISA instructions by mapping set operations to two types of PIM: in-DRAM bulk bitwise computing for bitvectors representing high- degree vertices, and near-memory logic layers for integer arrays representing low-degree vertices. Expand

References

SHOWING 1-10 OF 71 REFERENCES
Listing k-cliques in Sparse Real-World Graphs*
TLDR
This work revisits the iconic algorithm of Chiba and Nishizeki and develops the most efficient parallel algorithm for list all k-cliques in graphs containing up to tens of millions of edges, which is faster than state-of-the-art algorithms, while boasting an excellent degree of parallelism. Expand
Synthesis of Parallel Algorithms
TLDR
These thorough but introductory presentations demonstrate the most important algorithmic techniques and their use in exposing the hidden parallelism within problems. Expand
  • 2021
  • 2021
GraphMineSuite: Enabling High-Performance and Programmable Graph Mining Algorithms with Set Algebra
TLDR
GraphMineSuite (GMS) is proposed, the first benchmarking suite for graph mining that facilitates evaluating and constructing highperformance graph mining algorithms and is supported with a broad concurrency analysis for portability in performance insights, and a novel performance metric to assess the throughput of graphs mining algorithms. Expand
Parallel Clique Counting and Peeling Algorithms
TLDR
A new parallel algorithm is presented that has polylogarithmic span and is work-efficient with respect to the well-known sequential algorithm for $k$-clique listing by Chiba and Nishizeki and new parallel algorithms for producing unbiased estimations of clique counts are designed. Expand
SISA: Set-Centric Instruction Set Architecture for Graph Mining on Processing-in-Memory Systems
TLDR
The key design idea is to alleviate the bandwidth needs of SISA instructions by mapping set operations to two types of PIM: in-DRAM bulk bitwise computing for bitvectors representing high- degree vertices, and near-memory logic layers for integer arrays representing low-degree vertices. Expand
  • 2020
High-Performance Parallel Graph Coloring with Strong Guarantees on Work, Depth, and Quality
TLDR
This work develops the first parallel graph coloring heuristics with strong theoretical guarantees on work and depth and coloring quality, and designs a relaxation of the vertex degeneracy order, a well-known graph theory concept, to introduce a tunable amount of parallelism into the degeneracy ordering. Expand
Ordering heuristics for k-clique listing
TLDR
A comprehensive comparison of all the state-of-the-art k-clique listing and counting algorithms is presented and a new color ordering heuristics based on greedy graph coloring techniques which is able to significantly prune the unpromising search paths is proposed. Expand
...
1
2
3
4
5
...