Stream2HPC 2026: Orchestrating High Performance Computing with Real-Time Scientific Data in the Age of AI

As High-Performance Computing (HPC) enters the Exascale era, the scientific community faces a critical data paradox: our ability to generate data via high-fidelity simulations and next-generation instruments vastly outpaces our ability to store it. In many domains—from climate modeling and cosmology to experimental physics—simulation outputs and sensor inputs now frequently reach the Petabyte range per run. The traditional "compute-store-analyze" paradigm, which relies on dumping massive datasets to parallel file systems for post-processing, has become an unsustainable bottleneck. It introduces unacceptable latency, wastes energy on data movement, and often renders high-frequency data analysis practically inaccessible.

Stream2HPC 2026 addresses this challenge by exploring the architectural shift from file-based workflows to high-performance streaming. In this new paradigm, storage is treated not as a buffer for post processing once simulations are completed, but as a sink for final scientific results, while analysis, visualization, and reduction occur in-transit or in-situ.

This workshop is particularly timely given the rapid convergence of HPC and Artificial Intelligence. Modern scientific workflows increasingly rely on AI to steer simulations, train surrogate models on-the-fly, and integrate Digital Twins with real-time data from the edge. These "live" interactions require data to flow continuously between compute nodes and inference engines, bypassing the latency of the file system entirely.

Stream2HPC 2026 will serve as the premier forum for researchers, developers, vendors and facility operators to discuss the software stacks, protocols, and hardware architectures required to make streaming a first-class citizen in HPC. The workshop will move the conversation beyond ad-hoc, application-specific implementations toward standardized, interoperable frameworks.

Key topics of discussion will include:
* Architectures: High-throughput streaming middleware and effective coupling of MPI-based simulations with modern data streaming frameworks.
* Convergence: Streaming pipelines for training AI models in-situ and enabling real-time inference for simulation steering.
* Sustainability: The role of streaming in "Green HPC" by reducing the energy footprint associated with massive data movement and redundant storage IO.
* Edge-to-Exascale: End-to-end data pipelines connecting experimental facilities (synchrotrons, telescopes, microscopes) or existing data archives directly to supercomputing centers for immediate processing.
* Advanced Use Cases: Showcasing domain-specific workflows where streaming is mandatory, such as urgent computing for disaster response, transient event detection in high-energy physics, and closed-loop control for fusion energy or imaging techniques.

By convening experts from the traditional HPC I/O community alongside pioneers in Big Data and AI, Stream2HPC 2026 aims to define the roadmap for a future where time-to-solution is limited by compute capability, not by the speed of the disk.

---

Organizers:

• Bjoern Enders
• Maxime Martinasso
• Adam Thompson
• Thea Klaeboe Aarrestad