IOWARP / APPLICATIONS
Where the stack meets the science.
For scientists and partners asking whether IOWarp fits their work. The stack is built with and for scientific teams: materials scientists, the National Data Platform, DOE laboratories in the Genesis Mission, and the facilities that run the nation’s research computers. Each story covers the science, where context breaks, the parts of the stack involved, and what the work has produced, with sources.
Six stories across the stack.
| Application | 01Accelerators | 02Operating system | 03Storage systems | 04Context runtime | 05Agents and tools | 06Applications | 07Research tooling |
|---|---|---|---|---|---|---|---|
| A1 · Materials scienceAgents in the materials workflow | Not used | Not used | Not used | ||||
| A2 · Data infrastructureAgents on the National Data Platform | Not used | Not used | Not used | Not used | Not used | ||
| A3 · DOE Genesis · SPOTTER-AIProvenance and integrity for AI-driven science | Not used | Not used | Not used | Not used | Not used | ||
| A4 · DOE Genesis · PORTUSAgentic workflows across HPC and cloud | Not used | Not used | Not used | Not used | Not used | ||
| A5 · National infrastructureCLIO on national research systems | Not used | Not used | Not used | Not used | |||
| A6 · Accelerated computingGPU direct I/O and compression | Not used | Not used | Not used |
Layers follow the CLIO stack: accelerators, operating system, storage systems, context runtime, agents and tools, applications, and research tooling.
Application stories
- A1
Agents in the materials workflow
Design on HPC, additive manufacturing, X-ray tomography, and fatigue testing: the Utah team’s reference workflow, and the agent tools built for it.University of Utah · Illinois Tech · The HDF Group
Active collaboration - A2
Agents on the National Data Platform
MCP tools that let an agent search, stage, and analyze datasets federated through the National Data Platform, first shown on EarthScope data.National Data Platform · University of Utah · Illinois Tech
Demonstrated on EarthScope data - A3
Provenance and integrity for AI-driven science
A DOE Genesis Mission project at Argonne National Laboratory that traces integrity attacks on scientific AI workflows to their source. CLIO Agent provides the investigating agents.Argonne National Laboratory · Oak Ridge National Laboratory · Illinois Tech · Hewlett Packard Enterprise
Project began August 2026 - A4
Agentic workflows across HPC and cloud
A DOE Genesis Mission project at PNNL on performance-aware agentic workflows for the Genesis Mission Platform. The IOWarp team works on workflow motifs and performance models.Pacific Northwest National Laboratory · Princeton Plasma Physics Laboratory · Iowa State University · University of Delaware · Illinois Tech · collaborating with Oak Ridge National Laboratory and HPE
Project began August 2026 - A5
CLIO on national research systems
Where CLIO Core has been built and run: ALCF Aurora, DeltaAI, and the Ares cluster.Argonne Leadership Computing Facility · NCSA · The HDF Group · Illinois Tech
Built and run on national systems - A6
GPU direct I/O and compression
CUDA kernels that read and write IOWarp from the GPU, GPU payloads offloaded to host memory and NVMe, and compression on the device. A research thread built on NVIDIA technologies.Illinois Tech Gnosis Research Center · Argonne co-authors on KV-cache research
Research thread across the stack
Next step
Bring a workflow
that strains.
The most useful collaborations start from a real pipeline that is slow, fragile, or hard for an agent to operate: a characterization campaign, a federated catalog, a simulation campaign on a national system. The team works best with a concrete dataset, a baseline, and a question.