The resolution of Earth System models may be substantially increased in the next decade with the emergence of exascale supercomputer architectures. These architectures will likely support a range of floating-point precisions, and prompt us to reconsider our use of double-precision as a default. Single-precision arithmetic, for example, has been successfully applied to atmospheric models where 40% reductions of computational cost have been observed. Equally, the potential for reduction of numerical precision in ocean models has recently been explored (GMD 12, 3135–3148, 2019). Given that ocean models are, like atmospheric models, memory and communication bound, the use of single- or mixed-double-single-precision can be expected to deliver similar performance gains. For both atmosphere and ocean, there are however remaining questions regarding the impact on variables with long-timescale memory, the dependence on subtle differences in weak gradient scenarios of mixed-phase fluids, and more generally transport uncertainty, which would benefit from a systematic intercomparison of both algorithmic choices and their precision sensitivity.
Here we present preliminary results from running NEMO with single-precision arithmetic at ECMWF, focusing on its computational and forecast performance, with respect to the double-precision model. We focus on a double-gyre, mesoscale eddy-resolving test case and a global 1/4 degree ORCA configuration complete with sea-ice. We find that, in some cases, single-precision can deliver a greater than 2x speed-up with respect to double-precision.
