Imagine your company collects so much data that no single computer — or even a cluster of them — can handle it fast enough. That's the "exascale" problem. PROCESS built open-source software that lets organizations crunch enormous datasets by orchestrating computing power across multiple data centres, almost like a traffic controller for supercomputers. They tested it on real problems: analyzing medical records, optimizing airline ticket pricing, and mapping global disaster risks.
Companies in healthcare, aviation, and risk analytics are drowning in data that exceeds the capacity of their current computing infrastructure. Processing terabytes or petabytes of information — patient records, flight booking patterns, satellite disaster imagery — requires orchestrating resources across multiple supercomputing centres, which is technically complex and expensive to build from scratch.
The project built 5 exascale data service prototypes, progressing from first prototype through alpha and beta releases to a first packaged software version. Deliverables include a data orchestration system, an integrated user interface portal, and benchmark workflows — all open source and tested on real-world use cases in medical informatics, airline revenue management, and disaster risk reduction.
If you are an airline or travel platform dealing with massive booking and pricing datasets that overwhelm your current analytics — this project developed exascale data services tested specifically on airline revenue management. The open-source tools orchestrate processing across distributed computing resources, letting you run pricing optimization models on datasets too large for conventional systems.
If you are a healthcare organization struggling to process large volumes of patient records, imaging data, or genomic datasets — this project built and validated data service prototypes using medical informatics as one of its 5 real-world pilot applications. The packaged software release handles data orchestration across high-performance computing centres without requiring deep HPC expertise.
If you are a reinsurance or risk analytics company needing to process global disaster data for risk modelling — this project specifically tested its exascale data platform on open data for global disaster risk reduction. The 8-partner consortium across 6 countries built tools that integrate heterogeneous data sources at scales current solutions cannot handle.
PROCESS used an open-source strategy, meaning the software releases are freely available. Your costs would be infrastructure (HPC or cloud computing resources) and integration effort. The project invested EUR 2,972,250 in EU funding across 8 partners to develop the platform over 3 years.
Yes — the entire point was exascale computing, which means processing datasets at the largest scales currently possible. The project delivered 5 very large data service prototypes validated in real-world settings including industry pilot deployments. Use cases ranged from medical data to airline revenue optimization.
The project explicitly chose an open-source strategy to maximise uptake and reuse. This means you can adopt and modify the software without licensing fees. Check the project repository and deliverables for specific license terms (typically Apache or similar for EU-funded open-source projects).
PROCESS was designed with a modular architecture and mature software engineering practices specifically to minimise setup and maintenance effort. The project delivered a user interface portal and data orchestration tools to ease the learning curve for the broadest possible range of users.
The project ended in October 2020. Based on available project data, long-term maintenance depends on the consortium partners, particularly Ludwig-Maximilians-Universität München as coordinator. The open-source codebase may have community contributions, but you should verify current activity on the project website.
The project produced a first packaged version of the PROCESS software, followed by alpha and beta releases of the data service. This suggests a deployable product, but integration timelines depend on your infrastructure. The modular design was intended to reduce deployment time.
The consortium of 8 partners across 6 countries (Germany, Netherlands, Spain, Poland, Switzerland, Slovakia) is heavily academic — 5 universities and 1 research organization versus only 2 industry partners (25% industry ratio), with zero SMEs. This is typical for infrastructure-level computing research. The coordinator, Ludwig-Maximilians-Universität München, is a top German research university. The low industry presence means the technology was validated primarily in research settings, and commercial adoption would benefit from a systems integrator or cloud provider picking up the open-source tools. The geographic spread across Central and Western Europe gives good coverage of HPC centres but limited direct links to commercial end-users.
The coordinator is Ludwig-Maximilians-Universität München (Germany). Use SciTransfer's coordinator lookup to find the project lead's contact details.
Want to explore whether PROCESS tools can solve your big data bottleneck? SciTransfer can connect you directly with the development team and help assess fit for your use case.
