Computing the Full Earth System at 1 km Resolution

TL;DR

This paper reports the first global simulation of the full Earth system at 1.25 km resolution, leveraging massive GPU computing resources to achieve unprecedented detail and performance in modeling energy, water, and carbon flows.

Contribution

It introduces a high-resolution, full Earth system simulation at an unprecedented scale, utilizing advanced heterogeneous computing and optimization techniques.

Findings

Achieved 145.7 simulated days per day of real time.

Outperforms earlier atmosphere-only simulations at similar resolution.

Reduces code complexity while increasing performance and portability.

Abstract

We present the first-ever global simulation of the full Earth system at 1.25 km grid spacing, achieving highest time compression with an unseen number of degrees of freedom. Our model captures the flow of energy, water, and carbon through key components of the Earth system: atmosphere, ocean, and land. To achieve this landmark simulation, we harness the power of 8192 GPUs on Alps and 20480 GPUs on JUPITER, two of the world's largest GH200 superchip installations. We use both the Grace CPUs and Hopper GPUs by carefully balancing Earth's components in a heterogeneous setup and optimizing acceleration techniques available in ICON's codebase. We show how separation of concerns can reduce the code complexity by half while increasing performance and portability. Our achieved time compression of 145.7 simulated days per day enables long studies including full interactions in the Earth system and even outperforms earlier atmosphere-only simulations at a similar resolution.