Abstract
As technical limits are approached on the speed of processors used in super-computers, we may expect that greater parallelism will be sought to enhance their power. The parallel execution of varied tasks is to some degree used in all modern mainframes. In the new class of parallel machines, however, parallelism is a central design concept, and its presence is much more apparent to the user. Efficient use of a two processor XMP, for example, to run a single, large task such as an atmospheric GCM, is a very different problem than that of running the same task on its predecessor, the CRAY-1. As the number of processors increases from two to say eight or sixteen, we cannot expect the total computing power to increase proportionally, but the difficulty of programming probably will. Furthermore, the increase in computational speed we get from additional processors will depend both on our task (and how we organise it) and on the organisation of the machine itself. It thus behooves us at these early stages in the development of parallel super-computers to consider how well our models could be made to perform on various computer organisations. At Goddard Space Flight Center we started a project to explore the possibilities of using machines like Goddard’s Massively Parallel Processor (MPP) in atmospheric modelling.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
Reference
Flynn M.J. (1972) Some computer organisations and their effectiveness. IEEE Transactions on Computers, 21: 948–960.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1988 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Suarez, M.J. (1988). Atmospheric Modelling on a SIMD Computer. In: Hoffmann, GR., Snelling, D.F. (eds) Multiprocessing in Meteorological Models. Topics in Atmospheric and Oceanic Sciences. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-83248-2_11
Download citation
DOI: https://doi.org/10.1007/978-3-642-83248-2_11
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-83250-5
Online ISBN: 978-3-642-83248-2
eBook Packages: Springer Book Archive