Better Performance on Ensemble Kalman Filter with C++ senders/receivers

To deal with the chaotic nature of the real world, researchers use data assimilation as part of any realistic simulation. Data assimilation allows us to modify the simulation results based on the real-world observations and deal with the exponentially growing error.

In complex simulations for large problems, like predicting a contamination spread in a city, an Ensemble Kalman Filter (EnKF) is one established data assimilation component. In order to accelerate a simulation with an Ensemble Kalman Filter concurrency is necessary, as EnKF involves computations, communications and I/O.

Watch the presentation by Dr. Yuuichi Asahi from Japan Atomic Energy Agency to learn how C++ senders/receivers approach helps improve the performance by up to +39%. Watch now.