The projects SD3 currently designs, builds, and operates.
An open source research data management system built at NCSA. Clowder handles large, heterogeneous datasets, automatically extracts metadata and previews with a pluggable extractor framework, and is used by research groups and institutes across many disciplines.
The University of Illinois campus generative AI service, giving students, faculty, and staff secure access to large language models and the ability to build custom assistants grounded in their own documents.
The campus-wide program that gives University of Illinois researchers access to computing, storage, and expert support at no cost. SD3 contributes to the software and services that Illinois Computes offers to campus.
The Illinois Extended Reality testbed, an open source end-to-end XR system and research platform that brings together perception, visual, and audio pipelines so researchers can study and improve the full XR stack.
The Interdependent Networked Community Resilience Modeling Environment, developed with the NIST Center of Excellence for Risk-Based Community Resilience Planning, lets researchers model hazards and the resilience of interdependent physical, social, and economic systems.
Formerly the Illinois Computes Research Notebooks (ICRN). Hosted Jupyter Notebooks, RStudio, and VS Code with guaranteed CPU and shared GPU access, offered free of charge to Illinois researchers through the Illinois Computes program.
A self-hosted AI gateway that provides a web chat interface and an OpenAI-compatible API proxy, load balancing across model endpoints and skipping unhealthy ones.
An ARPA-H funded effort led by Stephen Boppart that combines optical imaging with AI screening to confirm, in real time during surgery, that all tumor tissue has been removed. NCSA builds the software and data infrastructure that connects imaging, analysis, and the operating room.
The Predictive Ecosystem Analyzer is an open source ecoinformatics toolbox and workflow system that manages the flow of information in and out of ecosystem models, enabling data assimilation, uncertainty analysis, and forecasting.
A data delivery service developed for the IRIS-HEP institute that filters and transforms petabyte-scale Large Hadron Collider datasets on Kubernetes and streams columnar results directly to physicists' analysis code.