Abstract
For well-resolving extreme weather events, running numerical weather prediction model with high resolution in time and space is essential. We explore how efficiently such modeling could be, using NURION. We have examined one of community numerical weather prediction models, WRF, and KISTI’s 5th supercomputer NURION of national HPC. Scalability of the model has been tested at first, and we have compared the computational efficiency of hybrid openMP + MPI runs with pure MPI runs. In addition to those parallel computing experiments, we have tested a new storage layer called burst buffer to see whether it can accelerate frequent I/O. We found that there are significant differences between the computational environments for running WRF model. First of all, we have tested a sensitivity of computational efficiency to the number of cores per node. The sensitivity experiments certainly tell us that using all cores per node does not guarantee the best results, rather leaving several cores per node could give more stable and efficient computation. For the current experimental configuration of WRF, moreover, pure MPI runs gives much better computational performance than any hybrid openMP + MPI runs. Lastly, we have tested burst buffer storage layer that is expected to accelerate frequent I/O. However, our experiments show that its impact is not consistently positive. We clearly confirm the positive impact with relatively smaller problem size experiments while the impact was not seen with bigger problem experiments. Significant sensitivity to the different computational configurations shown this paper strongly suggests that HPC users should find out the best computing environment before massive use of their applications
Highlights
Numerical weather prediction models have been developed since 1920s and have been practically applied to produce weather forecast based on computer simulations since 1950s [1]
KISTI NURION used for this study includes two different CPUs: Skylake and Knight Landing (KNL), we have examined only KNL cores in this study
We have tested the sensitivity of WRF simulation performance to the number of CPU cores per node, under the various settings of the number of nodes, such as 16/32/64/68 cores per node with 2/4/8/16 KNL nodes
Summary
Numerical weather prediction models have been developed since 1920s and have been practically applied to produce weather forecast based on computer simulations since 1950s [1]. For predicting weather for the days and for understanding and analyzing weather phenomena, we employ the numerical models, based on the natural laws of atmospheric physics/dynamics, to produce the meteorological data of interest. Those models provide 4-dimensional atmospheric states (zonal, meridional, vertical, and temporal, 4 dimensions) for the period of interest, as users configure. Increasing resolution of the numerical weather prediction model is very demanding these days, but it could not be possible without the considerable computing power This is why most popular supercomputers in the world are significantly consumed by the field of weather/climate modeling and simulation these days. It shows a significant portion of weather/climate research occupying the total supercomputer use over time
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have