kmELM Reference Simulation README

This document provides instructions for configuring and running the kmELM reference simulation on the several machines (Frontier, Perlmutter, and Baseline). This README outlines the steps necessary to set up a high-resolution reference case using the TESSFA branch of the kmELM code.

Prerequisites

• Access to the CCSI/ORNL Linux environment.
• SSH client to log into the machine.
• Git installed to clone the repository.

Setup Instructions

1. Log into your machine

Use SSH to log into the Baseline system:

ssh wangd@baseline.ccs.ornl.gov

2. Create a new directory

Create a directory for the kmELM source code:

mkdir -p /gpfs/wolf2/cades/cli185/proj-shared/wangd/kmELM
cd /gpfs/wolf2/cades/cli185/proj-shared/wangd/kmELM

3. Clone the kmELM repository

Checkout the kmELM code from GitHub:

git clone git@github.com:daliwang/kmELM.git
cd kmELM
git submodule update --init --recursive

Note: Type yes when prompted to initialize submodules.

4. Checkout the required branch

Switch to the TESSFA branch:

git fetch origin
git checkout -b my-dev origin/TESSFA

5. Configure for your HPC system (Optional)

If you are running kmELM on a different HPC system, you will need to modify the configuration files in the kmELM/cime_config/machines directory. You may need to edit:

• config_machines.xml
• config_batch.xml
• config_pio
• config_workflow
• cmake_macros

Refer to the sections for CCSI-Baseline, SummitPlus, or Frontier for guidance on how to adjust these settings for your environment.

6. Navigate to case generation directory

Go to the appropriate case generation directory for your machine (e.g., ORNL_baseline or Frontier):

cd kmELM/case_gene/<machines>

7. Review case creation script

Examine the case generation script, uELM_caseGEN_TVA.sh, to understand its functionality.

8. Modify case generation script

You will need to modify the uELM_caseGEN_TVA.sh script with the appropriate paths:

E3SM_SRCROOT="/gpfs/wolf2/cades/cli185/proj-shared/wangd/kmELM"
CASEDIR="/gpfs/wolf2/cades/cli185/proj-shared/wangd/e3sm_cases/uELM_${EXPID}_I1850uELMCNPRDCTCBC"

9. Create the uELM case

Run the case generation script:

sh uELM_caseGEN_TVA.sh

This script will create the necessary case directories.

10. Build and run the case

Once the case has been created, you can build and run it:

./case.build
./case.submit

Following these steps will enable you to set up and run the kmELM reference simulation on the selected HPC systems. Adjust paths and configurations as needed based on your specific environment and requirements.

For any further assistance, refer to the E3SM documentation or contact the kmELM development team.