Performance report and optimized implementations of Weather & Climate dwarfs on multi-node systems

TL;DR

This paper reports on optimized implementations of Weather & Climate dwarfs for multi-node CPU and GPU systems, achieving significant performance improvements and demonstrating the importance of hardware-specific optimizations and high-bandwidth interconnects.

Contribution

It introduces optimized multi-node CPU and GPU implementations of Weather & Climate dwarfs, highlighting performance gains and hardware considerations for exascale weather prediction models.

Findings

Up to 30% performance improvement on CPU multi-node systems.

Up to 10X speedup on multi-GPU systems with data residency and fast communication.

High-bandwidth interconnects like NVLink/NVSwitch enhance multi-GPU performance.

Abstract

This document is one of the deliverable reports created for the ESCAPE project. ESCAPE stands for Energy-efficient Scalable Algorithms for Weather Prediction at Exascale. The project develops world-class, extreme-scale computing capabilities for European operational numerical weather prediction and future climate models. This is done by identifying Weather & Climate dwarfs which are key patterns in terms of computation and communication (in the spirit of the Berkeley dwarfs). These dwarfs are then optimised for different hardware architectures (single and multi-node) and alternative algorithms are explored. Performance portability is addressed through the use of domain specific languages. Here we summarize the work performed on optimizations of the dwarfs focusing on CPU multi-nodes and multi-GPUs. We limit ourselves to a subset of the dwarf configurations chosen by the consortium. Intra-node optimizations of the dwarfs and energy-specific optimizations have been described in Deliverable D3.3. To cover the important algorithmic motifs we picked dwarfs related to the dynamical core as well as column physics. Specifically, we focused on the formulation relevant to spectral codes like ECMWF's IFS code. The main findings of this report are: (a) Up-to 30% performance gain with CPU based multi-node systems compared to optimized version of dwarfs from task 3.3 (see D3.3), (b) up to 10X performance gain on multiple GPUs from optimizations to keep data resident on the GPU and enable fast inter-GPU communication mechanisms, and (c) multi-GPU systems which feature a high-bandwidth all-to-all interconnect topology with NVLink/NVSwitch hardware are particularly well suited to the algorithms.