Fast shared-memory streaming multilevel graph partitioning

Jafari, Nazanin; Selvitopi, Oguz; Aykanat, Cevdet

doi:10.1016/j.jpdc.2020.09.004

Fast shared-memory streaming multilevel graph partitioning

Journal Article · Sat Sep 12 00:00:00 EDT 2020 · Journal of Parallel and Distributed Computing

DOI:https://doi.org/10.1016/j.jpdc.2020.09.004· OSTI ID:1844373

Jafari, Nazanin ^[1]; Selvitopi, Oguz ^[2]; Aykanat, Cevdet ^[3]

Univ. of Massachusetts, Amherst, MA (United States)
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Bilkent Univ., Ankara (Turkey)

In this report we show that a fast parallel graph partitioner can benefit many applications by reducing data transfers. The online methods for partitioning graphs have to be fast and they often rely on simple one-pass streaming algorithms, while the offline methods for partitioning graphs contain more involved algorithms and the most successful methods in this category belong to the multilevel approaches. In this work, we assess the feasibility of using streaming graph partitioning algorithms within the multilevel framework. Our end goal is to come up with a fast parallel offline multilevel partitioner that can produce competitive cutsize quality. We rely on a simple but fast and flexible streaming algorithm throughout the entire multilevel framework. This streaming algorithm serves multiple purposes in the partitioning process: a clustering algorithm in the coarsening, an effective algorithm for the initial partitioning, and a fast refinement algorithm in the uncoarsening. Its simple nature also lends itself easily for parallelization. The experiments on various graphs show that our approach is on the average up to 5.1x faster than the multi-threaded MeTiS, which comes at the expense of only 2x worse cutsize.

View Accepted Manuscript (DOE)

Research Organization:: Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)

Sponsoring Organization:: USDOE; USDOE Office of Science (SC)

Grant/Contract Number:: AC02-05CH11231

OSTI ID:: 1844373

Alternate ID(s):: OSTI ID: 1668421

Journal Information:: Journal of Parallel and Distributed Computing, Journal Name: Journal of Parallel and Distributed Computing Vol. 147; ISSN 0743-7315

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (17)

Fast multilevel implementation of recursive spectral bisection for partitioning unstructured problems Barnard, Stephen T.; Simon, Horst D. Concurrency: Practice and Experience, Vol. 6, Issue 2 https://doi.org/10.1002/cpe.4330060203	journal	April 1994
Parallel static and dynamic multi-constraint graph partitioning Schloegel, Kirk; Karypis, George; Kumar, Vipin Concurrency and Computation: Practice and Experience, Vol. 14, Issue 3 https://doi.org/10.1002/cpe.605	journal	January 2002
An Efficient Heuristic Procedure for Partitioning Graphs Kernighan, B. W.; Lin, S. Bell System Technical Journal, Vol. 49, Issue 2 https://doi.org/10.1002/j.1538-7305.1970.tb01770.x	journal	February 1970
On the validity of a front-oriented approach to partitioning large sparse graphs with a connectivity constraint Ciarlet, P.; Lamour, F. Numerical Algorithms, Vol. 12, Issue 1 https://doi.org/10.1007/BF02141748	journal	March 1996
Shape-optimized mesh partitioning and load balancing for parallel adaptive FEM Diekmann, Ralf; Preis, Robert; Schlimbach, Frank Parallel Computing, Vol. 26, Issue 12 https://doi.org/10.1016/S0167-8191(00)00043-0	journal	November 2000
PT-Scotch: A tool for efficient parallel graph ordering Chevalier, C.; Pellegrini, F. Parallel Computing, Vol. 34, Issue 6-8 https://doi.org/10.1016/j.parco.2007.12.001	journal	July 2008
A Combined Evolutionary Search and Multilevel Optimisation Approach to Graph-Partitioning Soper, A. J.; Walshaw, C.; Cross, M. Journal of Global Optimization, Vol. 29, Issue 2 https://doi.org/10.1023/B:JOGO.0000042115.44455.f3	journal	June 2004
Collective dynamics of ‘small-world’ networks Watts, Duncan J.; Strogatz, Steven H. Nature, Vol. 393, Issue 6684 https://doi.org/10.1038/30918	journal	June 1998
Near linear time algorithm to detect community structures in large-scale networks Raghavan, Usha Nandini; Albert, Réka; Kumara, Soundar Physical Review E, Vol. 76, Issue 3 https://doi.org/10.1103/PhysRevE.76.036106	journal	September 2007
Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication Catalyurek, U. V.; Aykanat, C. IEEE Transactions on Parallel and Distributed Systems, Vol. 10, Issue 7 https://doi.org/10.1109/71.780863	journal	July 1999
A Multilevel Memetic Approach for Improving Graph k-Partitions Benlic, Una; Hao, Jin-Kao IEEE Transactions on Evolutionary Computation, Vol. 15, Issue 5 https://doi.org/10.1109/TEVC.2011.2136346	journal	October 2011
Partitioning Sparse Matrices with Eigenvectors of Graphs Pothen, Alex; Simon, Horst D.; Liou, Kang-Pu SIAM Journal on Matrix Analysis and Applications, Vol. 11, Issue 3 https://doi.org/10.1137/0611030	journal	July 1990
A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs Karypis, George; Kumar, Vipin SIAM Journal on Scientific Computing, Vol. 20, Issue 1 https://doi.org/10.1137/S1064827595287997	journal	January 1998
Mesh Partitioning: A Multilevel Balancing and Refinement Algorithm Walshaw, C.; Cross, M. SIAM Journal on Scientific Computing, Vol. 22, Issue 1 https://doi.org/10.1137/S1064827598337373	journal	January 2000
An Efficient Parallel Algorithm for Matrix-Vector Multiplication Hendrickson, Bruce; Leland, Robert; Plimpton, Steve International Journal of High Speed Computing, Vol. 07, Issue 01 https://doi.org/10.1142/S0129053395000051	journal	March 1995
Distributed GraphLab: a framework for machine learning and data mining in the cloud Low, Yucheng; Bickson, Danny; Gonzalez, Joseph Proceedings of the VLDB Endowment, Vol. 5, Issue 8 https://doi.org/10.14778/2212351.2212354	journal	April 2012
A property of eigenvectors of nonnegative symmetric matrices and its application to graph theory [A property of eigenvectors of nonnegative symmetric matrices and its application to graph theory] Fiedler, Miroslav Czechoslovak Mathematical Journal, Vol. 25, Issue 4 https://doi.org/10.21136/CMJ.1975.101357	journal	January 1975

Cited By (2)

Buffered Streaming Graph Partitioning Faraj, Marcelo Fonseca; Schulz, Christian ACM Journal of Experimental Algorithmics, Vol. 27 https://doi.org/10.1145/3546911	journal	October 2022
Streaming Hypergraph Partitioning Algorithms on Limited Memory Environments Taşyaran, Fatih; Demireller, Berkay; Kaya, Kamer arXiv https://doi.org/10.48550/arxiv.2103.05394	preprint	January 2021

Similar Records

Applying graph partitioning methods in measurement-based dynamic load balancing

Technical Report · Mon Sep 26 00:00:00 EDT 2011 · OSTI ID:1114706

Multilevel Graph Partitioning and Embedding

Software · Wed Jun 05 20:00:00 EDT 2019 · OSTI ID:code-28744

Multilevel graph embedding

Journal Article · Tue Sep 22 20:00:00 EDT 2020 · Numerical Linear Algebra with Applications · OSTI ID:1669230

Related Subjects

97 MATHEMATICS AND COMPUTING
graph partitioning
multilevel graph partitioning
parallel graph partitioning
streaming algorithms

Fast shared-memory streaming multilevel graph partitioning

Citation Formats

References (17)

Cited By (2)

Similar Records

Related Subjects