Parallel Space-Time Adaptive Processing On a Cluster of Personal Computers

• Fernando Silva Wright-Patterson AFB
• Gary B. Lamont Wright-Patterson AFB

DOI: https://doi.org/10.5753/sbac-pad.1999.19783

Resumo

This work evaluates the capabilities and the performance of the Air Force Institute of Technology's Pile-of-PCs for parallel digital signal processing using space-time adaptive processing (STAP) under the Linux OS. The MITRE RT_STAP Benchmark version 1.1 is ported and executed on it, as well as on a cluster of six Sun SPARC workstations connected by a Myrinet network (the AFIT NOW), and on a IBM SP for comparison. Modifications to the RT_STAP benchmark source code are performed to accommodate the BLAS routines obtained from the US Department of Energy's Accelerated Strategic Computing Initiative, and the FFTPACK from the Netlib repository, allowing improvements in the sustained Gflops/sec rales. However, the Pile-of-PCs also reveals limited scalability as a result of severe communication overheads imposed by RT ST AP cornerturn operations. Analysis of experimental data indicates that the PC Cluster outperforms AFIT NOW but needs interconnection network improvements to be globally competitive to multicomputers such as the IBM SP.

Palavras-chave: Cluster of PCs, parallel signal processing, STAP, real-time benchmarking, Linux

Referências

CAIN, K. et al. RT_STAP: Real-Time Space-Time Adaptive Processing Benchmark. Technical Report MTR 96B0000021, MITRE Corporation. Bedford. Mass,. Feb. 1997.

HWANG, K. et. al. Benchmark Evaluation of the IBM SP2 for Parallel Signal Processing. IEEE Transactions on Parallel and Distributed Systems. vol. 7, no. 5. pp. 522-536, 1996.

LAMONT, Gary. B.: GALLAGHER. David, M. Scalable Distributed Multi-Dimensional FFT Algorithm Design, Implementation, and Analysis. Interim Report 2. Air Force Institute of Technology. Summer/Fall 1997.

The MITRE Corp. Benchmark Results, available at http://www.mitre.org/research/hpc, 1999.

NETLIB Repository at UTK. available at http://www.netlib.org, 1999.

SANDIA National Laboratories, available at http://www.sandia.gov, 1999.

SILVA, Fernando. Parallel Digital Signal Processing on a Network of Personal Computers - Case Study: Space-Time Adaptive Processing. MSCS Thesis. AFIT/GCS/ENG/99J-01, Graduate School of Engineering, Air Force Institute of Technology, Wright-Patterson AFB OH, June 1999.

WANG, C.J. et. al. STAP Benchmark Evaluation of the T3D, SP2, and Paragon. Proceedings of the 10th lnternational Conference on Parallel and Distributed Systems, New Orleans, Oct. 1997.

WARD, J. Space-Time Adaptive Processing for Airborne Radar. Technical Report 1015, MIT Lincoln Laboratory, DTIC n. AD-293032, 1994.