Portable Software Environment for Ultrahigh-Resolution ELM Development on GPUs  

作　　者：Dali Wang Peter Schwartz Fengming Yuan Franklin Eaglebarge Danial Riccuito Peter Thornton Chris Layton Qinglei Cao Dali Wang;Peter Schwartz;Fengming Yuan;Franklin Eaglebarge;Danial Riccuito;Peter Thornton;Chris Layton;Qinglei Cao(Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA;Pellissippi State Community College, Knoxville, TN, USA;Computer Science, Saint Louis University, St. Louis, MO, USA)

机构地区：[1]Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA [2]Pellissippi State Community College, Knoxville, TN, USA [3]Computer Science, Saint Louis University, St. Louis, MO, USA

出　　处：《Journal of Computer and Communications》2025年第2期28-36,共9页电脑和通信(英文)

摘　　要：This paper presents our endeavors in developing the large-scale, ultra-high-resolution E3SM Land Model (uELM), specifically designed for exascale computers furnished with accelerators such as Nvidia GPUs. The uELM is a sophisticated code that substantially relies on High-Performance Computing (HPC) environments, necessitating particular machine and software configurations. To facilitate community-based uELM developments employing GPUs, we have created a portable, standalone software environment preconfigured with uELM input datasets, simulation cases, and source code. This environment, utilizing Docker, encompasses all essential code, libraries, and system software for uELM development on GPUs. It also features a functional unit test framework and an offline model testbed for comprehensive numerical experiments. From a technical perspective, the paper discusses GPU-ready container generations, uELM code management, and input data distribution across computational platforms. Lastly, the paper demonstrates the use of environment for functional unit testing, end-to-end simulation on CPUs and GPUs, and collaborative code development.This paper presents our endeavors in developing the large-scale, ultra-high-resolution E3SM Land Model (uELM), specifically designed for exascale computers furnished with accelerators such as Nvidia GPUs. The uELM is a sophisticated code that substantially relies on High-Performance Computing (HPC) environments, necessitating particular machine and software configurations. To facilitate community-based uELM developments employing GPUs, we have created a portable, standalone software environment preconfigured with uELM input datasets, simulation cases, and source code. This environment, utilizing Docker, encompasses all essential code, libraries, and system software for uELM development on GPUs. It also features a functional unit test framework and an offline model testbed for comprehensive numerical experiments. From a technical perspective, the paper discusses GPU-ready container generations, uELM code management, and input data distribution across computational platforms. Lastly, the paper demonstrates the use of environment for functional unit testing, end-to-end simulation on CPUs and GPUs, and collaborative code development.

关 键 词：E3SM Land Model Ultrahigh-Resolution ELM Portable Software Environment GPU-Ready Environment 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]