Implementation of Minimum Spanning Tree Algorithm on Cell BE

Minimum spanning tree (MST) computation on a general graph is an irregular algorithm. Regular, data-parallel algorithms map well to the SIMD architecture of cell BE. Irregular algorithms on discrete structures like graphs are harder to map to them. Obtaining efficient parallel implementations for irregular graph problems always remains a challenge. The contribution of this paper lies in presenting a parallel implementation of MST on multi-core Cell Broadband Engine Architecture by utilizing the cell architecture potentials including DMA transfer, double buffering, mailboxes and at the same time considering the architecture limitations including memory constrains and thread synchronization. The implementation achieves up to 7 times speedup over sequential implementation.

[1]  Anne Condon,et al.  Parallel implementation of Bouvka's minimum spanning tree algorithm , 1996, Proceedings of International Conference on Parallel Processing.

[2]  Ibm Redbooks,et al.  Programming the Cell Broadband Engine Architecture: Examples and Best Practices , 2008 .

[3]  Fabrizio Petrini,et al.  Challenges in Mapping Graph Exploration Algorithms on Advanced Multi-core Processors , 2007, 2007 IEEE International Parallel and Distributed Processing Symposium.

[4]  Kurt Mehlhorn,et al.  Review of algorithms and data structures: the basic toolbox by Kurt Mehlhorn and Peter Sanders , 2011, SIGA.

[5]  N.R. Malik,et al.  Graph theory with applications to engineering and computer science , 1975, Proceedings of the IEEE.

[6]  Frank Dehne,et al.  Practical parallel algorithms for minimum spanning trees , 1998, Proceedings Seventeenth IEEE Symposium on Reliable Distributed Systems (Cat. No.98CB36281).

[7]  David A. Bader,et al.  Techniques for Designing Efficient Parallel Graph Algorithms for SMPs and Multicore Processors , 2007, ISPA.

[8]  Kam-Hoi Cheng,et al.  A fast graph search multiprocessor algorithm , 1997, Proceedings of the IEEE 1997 National Aerospace and Electronics Conference. NAECON 1997.