next up previous
Next: Test Images Up: Parallel Algorithms for Image Segmentation Previous: Implementation Notes

Acknowledgements

 

We would like to thank the UMIACS parallel systems staff for their help and machine maintenance while developing this research on the 16-node IBM SP-2 and 32-processor UMIACS CM-5, and the CASTLE group at UC Berkeley, especially the help and encouragement from Arvind Krishnamurthy, Lok Tin Liu, David Culler, Steve Luna, and Rich Martin. Computational support on UC Berkeley's 64-processor TMC CM-5 and 8-processor Intel Paragon was provided by NSF Infrastructure Grant number CDA-8722788. We also thank Toby Harness and the Numerical Aerodynamic Simulation Systems Division of NASA's Ames Research Center for use of their 128-processor CM-5 and 128-node (all wide) IBM SP-2.

We recognize Charles Weems at the University of Massachusetts for providing the DARPA test image suite, and Nawal Copty at Syracuse University for providing additional test images.

Additional thanks goes to Argonne National Labs for allowing use of their 128-node IBM SP-1, and to the Maui High Performance Computing Center for use of their 400-node IBM SP-2 machine. William Gropp, from the Mathematics and Computer Science Division of Argonne National Labs, provided significant help with the IBM SP-1 message passing interface. Also, Klaus Schauser, Oscar Ibarra, and David Probert of University of California, Santa Barbara, provided access to the 64-node UCSB Meiko CS-2. The Meiko CS-2 Computing Facility was acquired through NSF CISE Infrastructure Grant number CDA-9218202, with support from the College of Engineering and the UCSB Office of Research, for research in parallel computing.

Arvind Krishnamurthy provided additional help with his port of Split-C to the Cray Research T3D. The Jet Propulsion Lab/Caltech Cray T3D Supercomputer used in this investigation was provided by funding from the NASA Offices of Mission to Planet Earth, Aeronautics, and Space Science. We also acknowledge William Carlson and Jesse Draper from the Supercomputing Research Center for writing the parallel compiler AC on which the T3D port of Split-C has been based.

Please see http://www.umiacs.umd.edu/research/EXPAR for additional performance information.



next up previous
Next: Test Images Up: Parallel Algorithms for Image Segmentation Previous: Implementation Notes



David A. Bader
dbader@umiacs.umd.edu