GViewer: GPU-Accelerated Graph Visualization and Mining
Visualization is an effective way of identifying the patterns of interests (such as communities) in graphs including social networks and Web [8,6]. There have been a number of tools developed for graph visualizations, e.g., Tulip, Gephi and GMine . All of these tools use the CPU as the main power to calculate the graph layouts for visualization, such as force-directed layout . However, the layout calculation is usually computation intensive, for example, the force-directed layout has the complexity of O(N 3), where N is the number of vertexes in the graph. In our experiments, the CPU-based solution takes more than half one hours on the CPU to layout a graph with 14.5 thousand vertexes.
KeywordsShared Memory Bandwidth Utilization Spring Force Graph Visualization Graphic Processor
Unable to display preview. Download preview PDF.
- 1.Beckma, B.: Theory of Spectral Graph Layout. Technical report, MSR-TR-94-04 (1994)Google Scholar
- 2.Fruchterman, T.M.J., Reingold, E.M.: Graph drawing by force-directed placement. Softw. Pract. Exper. 21(11) (1991)Google Scholar
- 3.Grivet, S., Auber, D., Domenger, J.-P., Melancon, G.: Bubble tree drawing algorithm. In: International Conference on Computer Vision and Graphics (2004)Google Scholar
- 4.He, B., Yang, K., Fang, R., Lu, M., Govindaraju, N., Luo, Q., Sander, P.: Relational joins on graphics processors. In: SIGMOD (2008)Google Scholar
- 5.He, B., Yu, J.X.: High-throughput transaction executions on graphics processors. In: Proc. VLDB Endow., vol. 4, pp. 314–325 (February 2011)Google Scholar
- 6.Koenig, P.-Y., Zaidi, F., Archambault, D.: Interactive searching and visualization of patterns in attributed graphs. In: Proceedings of Graphics Interface (2010)Google Scholar
- 8.Rodrigues Jr., J.F., Tong, H., Traina, A.J.M., Faloutsos, C., Leskovec, J.: Gmine: a system for scalable, interactive graph visualization and mining. In: VLDB (2006)Google Scholar