OpenDreamKit · Project

Ready-to-Use Open Source Toolkit for High-Performance Mathematical Computing and Data Analysis

digitalPilotedTRL 7Thin data (2/5)

Imagine you need to do heavy-duty math and simulations but your tools don't talk to each other — your spreadsheet can't handle millions of equations, your code runs on one processor when you have sixteen, and sharing your work means emailing files back and forth. OpenDreamKit took the best open-source math software (like SageMath, Jupyter notebooks, and specialized algebra libraries) and made them work together seamlessly, run faster on modern hardware, and live in a shared online workspace. Think of it as building a well-organized, turbocharged workshop where all the power tools are compatible and everyone on the team can use them at once.

By the numbers

consortium partners across Europe

countries represented in consortium

total project deliverables completed

demonstrator deliverables showcasing working applications

universities contributing to development

The business problem

What needed solving

Companies and organizations that rely on heavy mathematical computation — risk modeling, engineering simulation, cryptography, data analysis — often struggle with fragmented tools that don't interoperate, code that only runs on a single processor core, and no easy way for teams to share and reproduce computational work. This slows down R&D cycles and creates bottlenecks when problems outgrow desktop-scale computing.

The solution

What was built

The project delivered parallelized math libraries (PARI/GP, Singular, LinBox, MPIR) optimized for modern multi-core and SIMD processors, a unified Virtual Research Environment architecture built on Jupyter notebooks with HPC integration, collaborative document editing via MathHub.info, interactive textbook demonstrators, and a continuous integration platform for cross-platform testing — totaling 59 deliverables.

Audience

Who needs this

Quantitative finance firms running large-scale risk and pricing modelsEngineering consultancies performing physics and structural simulationsEdTech companies building interactive STEM learning platformsPharmaceutical companies with computational chemistry pipelinesResearch computing centers modernizing their infrastructure

Business applications

Who can put this to work

Financial Services & Insurance

mid-size

Target: Quantitative trading firms and actuarial departments

If you are a financial services firm dealing with slow computation times for risk models and portfolio optimization — this project developed parallelized exact linear algebra algorithms and multi-core math libraries that speed up large-scale numerical computations. The 59 deliverables include production-ready parallel polynomial arithmetic and optimized processor-level instructions (AVX/SIMD) that can accelerate the matrix operations underpinning your pricing models.

Engineering & Simulation

any

Target: Engineering consultancies running physics simulations

If you are an engineering firm struggling with simulation bottlenecks and scattered computational tools — this project built a Virtual Research Environment that connects algebra, simulation, and visualization tools through Jupyter notebooks with HPC integration. Their demonstrators include physics problem-solving workflows and parallel computing across clusters, letting your engineers run bigger models without switching between disconnected software.

EdTech & Online Education

any

Target: Universities and e-learning platform providers

If you are an education provider wanting interactive, computation-enabled course materials — this project created interactive textbooks (demonstrated with Linear Algebra and Biology courses) and collaborative document editing through MathHub.info. The Jupyter notebook integration means students can run real calculations inside their learning materials, turning static PDFs into living documents.

Frequently asked

Quick answers

What would it cost to adopt these tools?

All software produced by OpenDreamKit is open source, meaning zero licensing fees. Your costs would be integration, training, and infrastructure (servers or cloud compute). Since the tools build on widely-used platforms like Jupyter and SageMath, most technical teams can get started without specialized consulting.

Can these tools handle industrial-scale workloads?

Yes — a core focus was parallelization and HPC integration. Deliverables include parallel sparse polynomial algorithms, distributed linear algebra on clusters and accelerators, and SIMD-optimized assembly code for modern processors. The architecture was designed to scale from single-core laptops to massively parallel machines.

What is the IP and licensing situation?

All code and tools are released as open source. This means you can use, modify, and integrate them freely. There are no patents or proprietary restrictions. The project explicitly committed to open-source delivery across all 59 deliverables.

How mature is the software — is it production-ready?

The project delivered production code integrated into established tools like SageMath, Jupyter, PARI/GP, and LinBox. A continuous integration platform was built for multi-platform testing. These are not prototypes — they are released versions of software already used by thousands of researchers worldwide.

Can we integrate this with our existing IT systems?

The component-based architecture was specifically designed for flexibility. Jupyter notebooks serve as the front-end interface, which integrates with most data science and engineering stacks. HPC and grid service interfaces are built in, so connecting to existing cluster infrastructure is straightforward.

Who maintains these tools now that the project has ended?

The tools are maintained by their respective open-source communities (SageMath, Jupyter, GAP, PARI/GP, Singular, LinBox). These communities existed before the project and continue active development. The project invested in sustainability by improving existing ecosystems rather than building throwaway software.

Consortium

Who built it

The consortium of 20 partners from 7 countries is overwhelmingly academic, with 16 universities, 2 research organizations, and just 1 SME — resulting in a 0% industry ratio. This is typical for research infrastructure projects. For a business buyer, the upside is deep technical expertise from top European math and computer science departments (including Université Paris-Saclay as coordinator). The downside is that no industrial partner validated the tools in a commercial setting during the project. However, the reliance on established open-source communities (Jupyter, SageMath) provides a practical bridge to industry adoption that pure academic projects often lack.

UNIVERSITE PARIS-SACLAYCoordinator · FR
UNIWERSYTET SLASKI W KATOWICACHparticipant · PL
FRIEDRICH-ALEXANDER-UNIVERSITAET ERLANGEN-NUERNBERGparticipant · DE
THE UNIVERSITY OF SHEFFIELDparticipant · UK
UNIVERSITEIT GENTparticipant · BE
SIMULA RESEARCH LABORATORY ASparticipant · NO
THE UNIVERSITY COURT OF THE UNIVERSITY OF ST ANDREWSparticipant · UK
CONSTRUCTOR UNIVERSITY BREMEN GGMBHparticipant · DE
UNIVERSITY OF LEEDSparticipant · UK
UNIVERSITE GRENOBLE ALPESparticipant · FR
UNIVERSITE DE BORDEAUXthirdparty · FR
RHEINLAND-PFALZISCHE TECHNISCHE UNIVERSITATparticipant · DE
UNIVERSITY OF SOUTHAMPTONparticipant · UK
UNIVERSITAT ZURICHparticipant · CH
UNIVERSITY OF WARWICKparticipant · UK
THE CHANCELLOR, MASTERS AND SCHOLARS OF THE UNIVERSITY OF OXFORDparticipant · UK
CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRSparticipant · FR
EUROPEAN X-RAY FREE-ELECTRON LASERFACILITY GMBHparticipant · DE
UNIVERSITE DE VERSAILLES SAINT-QUENTIN EN YVELINESparticipant · FR

How to reach the team

Coordinator is at Université Paris-Saclay, France. SciTransfer can facilitate an introduction to the research team.

Order a report on this consortium See UNIVERSITE PARIS-SACLAY profile →

Next steps

Talk to the team behind this work.

Want to explore how these open-source computational tools could accelerate your engineering or analytics workflows? SciTransfer can arrange a technical briefing with the development team and help you evaluate fit for your specific use case.

Order a report on this topic More in Digital & ICT