Multiprocessing in Meteorological Models pp 307-317 | Cite as
The Floating Point Systems T Series
Conference paper
Abstract
The Floating Point Systems T-Series is a massively parallel supercomputer for scientific and engineering applications. Each processing node combines the performance of fully pipelined vector oriented floating point arithmetic with the connectivity and parallel processing capability of the INMOS Transputer. Because each processor has its own memory rather than a switched path to a global memory, very large homogeneous configurations are possible. This paper describes the new architecture and gives an example of distributed processing with associated performance.
Keywords
Very Large Scale Integration Direct Memory Access Result Matrix Arithmetic Unit System Board
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Preview
Unable to display preview. Download preview PDF.
References
- Floating Point Systems Inc., 1987: “Programming the FPS T Series: Program Development Guide.” FPS Publication No. 860–0001–003.Google Scholar
- Fox, G.C., and Otto, S.W., 1984: “Algorithms for Concurrent Processors.” Physics Today, 37, 5, pp. 50–59.CrossRefGoogle Scholar
- Hawkinson, S., 1986: “The FPS T Series: A Parallel Vector Supercomputer.” Floating Point Systems, Beaverton, Oregon.Google Scholar
- Gustafson, J., 1986: “Programming the FPS T Series.” Checkpoint, FPS Technical Journs, 4, 6, pp. 2–8.Google Scholar
- Inmis Ltd., 1984: “OCCAM Programming Manual.” Prentice/Hall International.Google Scholar
Copyright information
© Springer-Verlag Berlin Heidelberg 1988