Scalable betweenness centrality on multi-GPU systems

Bernaschi, Massimo; Carbone, Giancarlo; Vella, Flavio

doi:10.1145/2903150.2903153

Cited by 21 publications

(13 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most distributed-memory implementations [8,16,19,20,61] of BC are based on Brandes's algorithm [13]; many of these implementations do level-by-level traversals of the graph to efficiently calculate dependency values. Maximal-Frontier BC [53] is formulated using communication-efficient matrix operations.…”

Section: Related Workmentioning

confidence: 99%

A round-efficient distributed betweenness centrality algorithm

Hoang

Pontecorvi

Dathathri

et al. 2019

Proceedings of the 24th Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

We present Min-Rounds BC (MRBC), a distributed-memory algorithm in the CONGEST model that computes the betweenness centrality (BC) of every vertex in a directed unweighted n-node graph in O (n) rounds. Min-Rounds BC also computes all-pairs-shortest-paths (APSP) in such graphs. It improves the number of rounds by at least a constant factor over previous results for unweighted directed APSP and for unweighted BC, both directed and undirected. We implemented MRBC in D-Galois, a state-of-the-art distributed graph analytics system, incorporated additional optimizations enabled by the D-Galois model, and evaluated its performance on a production cluster with up to 256 hosts using power-law and road networks. Compared to the BC algorithm of Brandes, on average, MRBC reduces the number of rounds by 14.0× and the communication time by 2.8× for the graphs in our test suite. As a result, MRBC is 2.1× faster on average than Brandes BC for real-world web-crawls on 256 hosts.

show abstract

Section: Related Workmentioning

confidence: 99%

A round-efficient distributed betweenness centrality algorithm

Hoang

Pontecorvi

Dathathri

et al. 2019

Proceedings of the 24th Symposium on Principles and Practice of Parallel Programming

View full text Add to dashboard Cite

show abstract

“…The complexity of our metric can be further reduced if the algorithm is parallelized, which is a matter of parallelizing the single-source shortest paths (SSSP) and the accumulation functions in Brandes' algorithm [33], considering unweighted networks. This is feasible [34], [35], [36], [37] and the graph traversal performed in the SSSP needs to be run ρ + 1 times to find all the paths we need to compute the ρ-geodesic betweenness. In addition, if only local knowledge is available, it is possible to modify a distributed algorithm as the one proposed by Lehman and Kaufman [38] to compute our metric.…”

Section: Random Walk Vs ρ-Geodesic Between-nessmentioning

confidence: 99%

The Power of Quasi-Shortest Paths: $\rho$ -Geodesic Betweenness Centrality

Medeiros¹,

Campista²,

Mitton³

et al. 2017

IEEE Trans. Netw. Sci. Eng.

View full text Add to dashboard Cite

“…Table 1), therefore a vertex-based parallelism would be more suitable for such instances. Other efficient data-thread mapping techniques, like active-edge parallelism [2,3] or other warp-centric strategies [18], seem to be not very effective for Energy games instances where the average degree is pretty low.…”

Section: Related Workmentioning

confidence: 99%

Accelerating Energy Games Solvers on Modern Architectures

Formisano

Gentilini

Vella³

2017

Proceedings of the Seventh Workshop on Irregular Applications: Architectures and Algorithms

Self Cite

View full text Add to dashboard Cite

Quantitative games, where quantitative objectives are defined on weighted game arenas, provide natural tools for designing faithful models of embedded controllers. Instances of these games that recently gained interest are the so called Energy Games. The fast-known algorithm solves Energy Games in O(EV W ) where W is the maximum weight. Starting from a sequential baseline implementation, we investigate the use of massively data computation capabilities supported by modern Graphics Processing Units to solve the initial credit problem for Energy Games. We present four different parallel implementations on multi-core CPU and GPU systems. Our solution outperforms the baseline implementation by up to 36x speedup and obtains a faster convergence time on real-world graphs.

show abstract

Scalable betweenness centrality on multi-GPU systems

Cited by 21 publications

References 14 publications

A round-efficient distributed betweenness centrality algorithm

A round-efficient distributed betweenness centrality algorithm

The Power of Quasi-Shortest Paths: $\rho$ -Geodesic Betweenness Centrality

Accelerating Energy Games Solvers on Modern Architectures

Contact Info

Product

Resources

About