The Alliant FX/Series: A language driven architecture for parallel processing of dusty deck fortran
The Alliant FX/Series− multiprocessor mini-supercomputers represent a major advance in commercial computing by bringing the benefits of vectorization and parallelism transparently to user applications. By providing compiler support for the automatic discovery of “concurrency” in existing Fortran programs, the FX/Series is often able to achieve time-to-solution speedups proportional to the number of processors applied to the problem. To achieve this computational performance advantage, the instruction set architecture supports scalar, floating point, vector, and concurrency instructions. The hardware architecture and operating system were designed and optimized to deliver both high throughput and computational performance to user programs.
KeywordsFloating Point Alliant System American National Standard Institute Float Point Operation Concurrent Optimization
Unable to display preview. Download preview PDF.
- 1.American National Standards Institute, Inc., American National Standard Programming Language FORTRAN, American National Standards Institute, Inc., X3.9-1978, New York, N.Y., 1978.Google Scholar
- 2.American National Standards Institute, Inc., American National Standard Programming Language FORTRAN, American National Standards Institute, Inc., X3J3/S8.102, November 1986.Google Scholar
- 3.“Proceedings of the 1986 International Conference on Parallel Processing,” IEEE Computer Society, August 19–22, 1986.Google Scholar
- 4.Dongarra, J., “Performance of Various Computers Using Standard Linear Equations Software in Fortran Environment,” Argonne National Laboratory, Technical Memorandum No. 23, September, 1986.Google Scholar
- 5.Hwang, K., Ed., Supercomputers: Design and Applications, IEEE Computer Society Press, Silver Spring, Md., 1984.Google Scholar
- 6.Hwang, K., Su, S., and Ni, L. M., “Vector Computer Architecture and Processing Techniques,” Advances in Computers, Vol. 20, Academic Press, 1981.Google Scholar
- 7.Kuck, D. J., et. al., “Measurements of Parallelism In Ordinary Fortran Programs,” presented at the Sagamore Computer Conference on Parallel Processing, 1973.Google Scholar
- 8.McMahon, F. H., Lawrence Livermore National Laboratory, December, 1985.Google Scholar
- 9.Metcalf, M., Fortran Optimization, Academic Press Inc., 1982.Google Scholar
- 10.Sorensen, D. C., “Advanced Computing Research at Argonne National Laboratory,” presented at the Alliant User Group Conference, September, 1986.Google Scholar
- 11.Stephenson, T., “Applications of the Alliant FX/8 and the Pixar 2D to Image Processing, Image Understanding, and Image Generation,” presented at the Alliant User Group Conference, September, 1986.Google Scholar