Profiling Methodology and Performance Tuning of the Met Office Unified Model for Weather and Climate Simulations

Abstract

Global weather and climate modelling is a compute-intensive task that is mission-critical to government departments concerned with meteorology and climate change. The dominant component of these models is a global atmosphere model. One such model, the Met Office Unified Model (MetUM), is widely used in both Europe and Australia for this purpose. This paper describes our experiences in developing an efficient profiling methodology and scalability analysis of the MetUM version 7.5 at both low scale and high scale atmosphere grid resolutions. Variability within the execution of the MetUM and variability of the run-time of identical jobs on a highly shared cluster are taken into account. The methodology uses a lightweight profiler internal to the MetUM which we have enhanced to have minimal overhead and enables accurate profiling with only a relatively modest usage of processor time. At high-scale resolution, the MetUM scaled to core counts of 2048, with load imbalance accounting a significant fraction the loss from ideal performance. Recent patches have removed two relatively small sources of inefficiency. Internal segment size parameters gave a modest performance improvement at low-scale resolution (such as are used in climate simulation); this however was not significant a higher scales. Near-square process grid configurations tended to give the best performance. Byte-swapping optimizations vastly improved I/O performance, which has in turn a large impact on performance in operational runs.