The Earth-2 Weather Analytics Blueprint is a reference implementation of a geospatial data analysis service comprised of multiple features from NVIDIA's Earth-2 platform. This repository provides a starting point for you to build your own weather / climate systems accelerated by AI and Omniverse. This blueprint will demonstrate the following core features and their integration to create a complete weather analytics service:
- Using the Nvidia Omniverse platform to aggregate and visualize data at scale
- Deploying and scaling AI weather models through Nvidia Inference Microservices (NIMs)
- Using components of the Nvidia Earth-2 platform to implement an adapter framework that connects to partner and public data stores
- Earth-2 Command Center (E2CC): An Omniverse Kit application with extensions that enable visualization of geo-spatial data. This is the front-end of the blueprint.
- Data Federation Mesh (DFM): Glue code or orchestration layer that processes pipelines from different data stores. Connects the different data sources to E2CC and runs the needed pipelines to generate textures.
- FourCastNet NIM (FCN NIM): An AI weather model, packaged as an Nvidia Inference Microservice, for global weather forecasting.
This repository is specifically designed for developers. Follow these steps to clone the repository:
git lfs install
git clone [email protected]:NVIDIA-Omniverse-blueprints/earth2-weather-analytics.git
cd earth2-weather-analytics
# just in case
git lfs fetch --allUse the blueprint guide to build and deploy the blueprint on your infrastructure. This guide requires intermediate knowledge of Python, Docker, and beginner knowledge of Kubernetes / Helm Charts.
- Workflow Overview
- Prerequisites
- Quickstart
- Deployment Guide
- Earth-2 Command Center Guide
- Data Federation Mesh Guide
- Sequence Diagram
- Troubleshooting
Below are a few samples of what the following blueprint can produce.
| Multivariant Analysis | Regional Data | RTX Rendered |
|---|---|---|
 |
 |
 |
The Earth-2 Weather Analytics Blueprint is provided under the Omniverse License Agreement, please see LICENSE.md for full license text.
The Earth-2 Weather Analytics Omniverse Blueprint is shared as reference and is provided "as is". The security in the production environment is the responsibility of the end users deploying it. When deploying in a production environment, please have security experts review any potential risks and threats; define the trust boundaries, implement logging and monitoring capabilities, secure the communication channels, integrate AuthN & AuthZ with appropriate access controls, keep the deployment up to date, ensure the containers/source code are secure and free of known vulnerabilities.