Wide Interest Seminar: V. Balaji
« Climate computing: the state of play »
By V. BALAJI
NOAA/GFDL and Princeton University
Climate models represent a large variety of processes on a variety of time and space scales, a canonical example of multi-physics multi-scale modeling. In addition the system is physically characterized by sensitive dependence on initial conditions, and natural stochastic variability, and very long integrations are needed
to extract signals of climate change. Algorithms generally possess weak scaling. Weak-scaling, I/O and memory-bound, multi-physics codes present particular challenges to computational performance.
In this talk I will present trends in climate science driving models toward higher resolution, greater complexity, and larger ensembles, all of which present computing challenges. I will also discuss the
prospects for adapting these models to novel hardware and programming models. Finally, I will introduce a set of metrics that can be used for the comparative study of computational performance of Earth System models. These measures do not require specialized software or specific hardware counters, and should be accessible to anyone. They are independent of platform, and underlying parallel programming models. We might use these measures as a basis for a CPMIP, a computational performance MIP.