Fast GPU-Based Generation of Large Graph Networks From Degree Distributions

Alam, Maksudul; Perumalla, Kalyan

doi:10.3389/fdata.2021.737963

Fast GPU-Based Generation of Large Graph Networks From Degree Distributions

Journal Article · Fri Nov 26 00:00:00 EST 2021 · Frontiers in Big Data

DOI:https://doi.org/10.3389/fdata.2021.737963· OSTI ID:1833940

^[1]; ^[1]

Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)

Synthetically generated, large graph networks serve as useful proxies to real-world networks for many graph-based applications. The ability to generate such networks helps overcome several limitations of real-world networks regarding their number, availability, and access. Here, we present the design, implementation, and performance study of a novel network generator that can produce very large graph networks conforming to any desired degree distribution. The generator is designed and implemented for efficient execution on modern graphics processing units (GPUs). Given an array of desired vertex degrees and number of vertices for each desired degree, our algorithm generates the edges of a random graph that satisfies the input degree distribution. Multiple runtime variants are implemented and tested: 1) a uniform static work assignment using a fixed thread launch scheme, 2) a load-balanced static work assignment also with fixed thread launch but with cost-aware task-to-thread mapping, and 3) a dynamic scheme with multiple GPU kernels asynchronously launched from the CPU. The generation is tested on a range of popular networks such as Twitter and Facebook, representing different scales and skews in degree distributions. Results show that, using our algorithm on a single modern GPU (NVIDIA Volta V100), it is possible to generate large-scale graph networks at rates exceeding 50 billion edges per second for a 69 billion-edge network. GPU profiling confirms high utilization and low branching divergence of our implementation from small to large network sizes. For networks with scattered distributions, we provide a coarsening method that further increases the GPU-based generation speed by up to a factor of 4 on tested input networks with over 45 billion edges.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 1833940

Journal Information:: Frontiers in Big Data, Journal Name: Frontiers in Big Data Journal Issue: 1 Vol. 4; ISSN 2624-909X

Publisher:: FrontiersCopyright Statement

Country of Publication:: United States

Language:: English

References (23)

Connected Components in Random Graphs with Given Expected Degree Sequences Chung, Fan; Lu, Linyuan Annals of Combinatorics, Vol. 6, Issue 2 https://doi.org/10.1007/pl00012580	journal	November 2002
Defining and evaluating network communities based on ground-truth Yang, Jaewon; Leskovec, Jure Knowledge and Information Systems, Vol. 42, Issue 1 https://doi.org/10.1007/s10115-013-0693-z	journal	October 2013
Parallel Algorithms for Generating Random Networks with Given Degree Sequences Alam, Maksudul; Khan, Maleq International Journal of Parallel Programming, Vol. 45, Issue 1 https://doi.org/10.1007/s10766-015-0389-y	journal	October 2015
Novel Parallel Algorithms for Fast Multi-GPU-Based Generation of Massive Scale-Free Networks Alam, Maksudul; Perumalla, Kalyan S.; Sanders, Peter Data Science and Engineering, Vol. 4, Issue 1 https://doi.org/10.1007/s41019-019-0088-6	journal	March 2019
Stochastic blockmodels: First steps Holland, Paul W.; Laskey, Kathryn Blackmond; Leinhardt, Samuel Social Networks, Vol. 5, Issue 2 https://doi.org/10.1016/0378-8733(83)90021-7	journal	June 1983
Scalable generation of scale-free graphs Sanders, Peter; Schulz, Christian Information Processing Letters, Vol. 116, Issue 7 https://doi.org/10.1016/j.ipl.2016.02.004	journal	July 2016
Communication-free massively distributed graph generation Funke, Daniel; Lamm, Sebastian; Meyer, Ulrich Journal of Parallel and Distributed Computing, Vol. 131 https://doi.org/10.1016/j.jpdc.2019.03.011	journal	September 2019
An introduction to exponential random graph (p*) models for social networks Robins, Garry; Pattison, Pip; Kalish, Yuval Social Networks, Vol. 29, Issue 2 https://doi.org/10.1016/j.socnet.2006.08.002	journal	May 2007
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
Error and attack tolerance of complex networks Albert, Réka; Jeong, Hawoong; Barabási, Albert-László Nature, Vol. 406, Issue 6794 https://doi.org/10.1038/35019019	journal	July 2000
Community structure in social and biological networks Girvan, M.; Newman, M. E. J. Proceedings of the National Academy of Sciences, Vol. 99, Issue 12 https://doi.org/10.1073/pnas.122653799	journal	June 2002
The average distances in random graphs with given expected degrees Chung, F.; Lu, L. Proceedings of the National Academy of Sciences, Vol. 99, Issue 25 https://doi.org/10.1073/pnas.252631999	journal	December 2002
Markov Graphs Frank, Ove; Strauss, David Journal of the American Statistical Association, Vol. 81, Issue 395 https://doi.org/10.1080/01621459.1986.10478342	journal	September 1986
Highly optimized tolerance: A mechanism for power laws in designed systems Carlson, J. M.; Doyle, John Physical Review E, Vol. 60, Issue 2 https://doi.org/10.1103/physreve.60.1412	journal	August 1999
Efficient generation of large random networks Batagelj, Vladimir; Brandes, Ulrik Physical Review E, Vol. 71, Issue 3 https://doi.org/10.1103/physreve.71.036113	journal	March 2005
Power laws and the AS-level internet topology Siganos, G.; Faloutsos, M.; Faloutsos, P. IEEE/ACM Transactions on Networking, Vol. 11, Issue 4 https://doi.org/10.1109/TNET.2003.815300	journal	August 2003
Emergence of Scaling in Random Networks Barabási, Albert-László; Albert, Réka Science, Vol. 286, Issue 5439 https://doi.org/10.1126/science.286.5439.509	journal	October 1999
Generating Massive Scale-Free Networks under Resource Constraints Meyer, Ulrich; Penschuck, Manuel 2016 Proceedings of the Eighteenth Workshop on Algorithm Engineering and Experiments (ALENEX) https://doi.org/10.1137/1.9781611974317.4	conference	January 2016
A large time-aware web graph Boldi, Paolo; Santini, Massimo; Vigna, Sebastiano ACM SIGIR Forum, Vol. 42, Issue 2 https://doi.org/10.1145/1480506.1480511	journal	November 2008
Fast random graph generation Nobari, Sadegh; Lu, Xuesong; Karras, Panagiotis Proceedings of the 14th International Conference on Extending Database Technology - EDBT/ICDT '11 https://doi.org/10.1145/1951365.1951406	conference	January 2011
A high-level and scalable approach for generating scale-free graphs using active objects Azadbakht, Keyvan; Bezirgiannis, Nikolaos; de Boer, Frank S. Proceedings of the 31st Annual ACM Symposium on Applied Computing - SAC '16 https://doi.org/10.1145/2851613.2851722	conference	January 2016
On power-law relationships of the Internet topology Faloutsos, Michalis; Faloutsos, Petros; Faloutsos, Christos ACM SIGCOMM Computer Communication Review, Vol. 29, Issue 4 https://doi.org/10.1145/316194.316229	journal	October 1999
Groupies in random bipartite graphs Shang, Yilun Applicable Analysis and Discrete Mathematics, Vol. 4, Issue 2 https://doi.org/10.2298/AADM100605021S	journal	January 2010

Similar Records

Scaling and Quality of Modularity Optimization Methods for Graph Clustering

Conference · Wed Sep 25 00:00:00 EDT 2019 · OSTI ID:1580597

Parallelizing Single Source Shortest Path with OpenSHMEM

Conference · Mon Aug 07 00:00:00 EDT 2017 · OSTI ID:1567473

A Novel Multithreaded Algorithm For Extracting Maximal Chordal Subgraphs

Conference · Thu Oct 25 00:00:00 EDT 2012 · OSTI ID:1054479

Related Subjects

97 MATHEMATICS AND COMPUTING
GPU (graphic processing unit)
SIMT architectures
graph generation
large graph
random network

Fast GPU-Based Generation of Large Graph Networks From Degree Distributions

Citation Formats

References (23)

Similar Records

Related Subjects