Read more
Numerical weather prediction on the one hand needs a very large number of floating point calculations, but on the other hand is very time-critical. Therefore, the largest computers available, i.e., the "supercomputers", have usually been acquired by the national meteorological services long before they were used in other fields of research or business. Since the available technology limits the speed of any single computer, parallel computations have become necessary to achieve further improvements in the number of results produced per time unit. This book collects the papers presented at two workshops held at ECMWF on the topic of parallel processing in meteorological models. It provides an insight into the state-of-the-art in using parallel processors operationally and allows extrapolation to other time-critical applications. It also shows trends in migrating to massive parallel systems in the near future.
List of contents
Introduction: The Case for Multi-Processing in Meteorological Applications.- Workshop I: 3.-6. December 1984.- State of the Art of Multi-Processors.- Programming Languages for Multiprocessor Systems.- Practical Concerns in Multitasking on the Cray X-MP.- (r?, n½, s½) Measurements on the 2-CPU Cray X-MP.- MIMD Processing and the Denelcor HEP.- Using an IBM Multiprocessor System.- Computer Requirements for Atmospheric Modelling.- The Current State of Meteorological Modelling.- Aspects of Using Multiprocessors for Meteorological Modelling.- Atmospheric Modelling on a SIMD Computer.- The Multitasking Spectral Model at ECMWF.- U. K. Meteorological Office's Plans for Using Multiprocessor Systems.- Canadian Meteorological Service's Plans for Using Multi-Processor Systems.- Multiprocessors: Requirements and Consequences: Summary of the Discussion.- Workshop II: 8.-11. December 1986.- Tools for Assessing Multiprocessing.- Dataflow Architectures and Implicit Parallel Programming.- Transposing Large Arrays in Extended Memory.- An Approach to Automatic Parallel Processing.- The Floating Point Systems T Series.- Parallel Processing on an IBM 3090 with Vector Facility.- Multi-Tasking with a Memory Hierarchy.- The Myth of Performance for Parallel Machines.- The ECMWF Model: Past, Present and Future.- Microtasking an Atmospheric Model at Environment Canada.- Application of Multi-Processors to Operational Analysis and Forecasts at the French Weather Service.- Parallel Integration of the Norwegian Prediction Model.- A Comparative Study of the ECMWF Weather Model on Several Multiprocessor Architectures.- Multiprocessors: Requirements and Perspectives: Summary of the Discussion.
