ExaHyPE · Project

Supercomputer Software That Simulates Earthquakes, Tsunamis, and Cosmic Events Faster

digitalPrototypeTRL 4Thin data (2/5)

Imagine you need to predict how an earthquake will ripple through a city, or how two neutron stars collide — but your computer would take months to crunch the numbers. ExaHyPE built a simulation engine that runs these massive calculations on the world's most powerful supercomputers, squeezing out far more speed from the hardware. Think of it like upgrading from a hand calculator to a spreadsheet — same math, but orders of magnitude faster. The engine is open-source and works for any problem described by the same type of equations, from seismic waves to blood flow in arteries.

By the numbers

EUR 2,795,000

EU funding for development

consortium partners

countries involved (DE, IT, RU, UK)

total project deliverables

software demo deliverables released

The business problem

What needed solving

Companies in geophysics, insurance, and energy exploration need to simulate earthquakes, seismic waves, and other large-scale physical phenomena — but current software either cannot use modern supercomputers efficiently or produces results too slowly to be useful for risk assessment and decision-making. As supercomputer hardware becomes more energy-constrained and parallel, legacy simulation codes fail to keep up.

The solution

What was built

ExaHyPE delivered an open-source simulation engine for solving large systems of conservation-law equations on exascale supercomputers. Concrete outputs include a petascale prototype, optimized vectorized solvers and Riemann solvers, dynamic load balancing software, multi-PDE and multi-parameter support, and a final open-source release with permanent archiving.

Audience

Who needs this

Geophysical consulting firms running earthquake and seismic hazard simulationsReinsurance companies modeling natural catastrophe riskOil and gas companies doing seismic exploration and subsurface imagingHPC cloud service providers offering simulation-as-a-serviceNational meteorological and geological survey agencies

Business applications

Who can put this to work

Seismic hazard assessment

enterprise

Target: Geophysical consulting firms and reinsurance companies

If you are a geophysical services company struggling with earthquake risk models that take weeks to run — this project developed an open-source simulation engine that handles seismic wave propagation on next-generation supercomputers. It models dynamic rupture processes and regional wave patterns, giving you faster and more detailed hazard maps. The engine was built by 7 partners across 4 countries with EUR 2,795,000 in EU funding.

Oil and gas exploration

enterprise

Target: Seismic data processing and reservoir modeling companies

If you are an exploration company that relies on seismic imaging to find resources underground — ExaHyPE's engine solves the same type of wave equations your models use, but on exascale hardware with adaptive mesh refinement that focuses computing power where it matters most. The final open-source release includes optimized solvers and dynamic load balancing across massive compute clusters. This means higher-resolution subsurface models without proportionally higher compute costs.

High-performance computing services

mid-size

Target: HPC cloud providers and simulation-as-a-service vendors

If you are an HPC service provider looking to offer turnkey simulation capabilities — ExaHyPE delivered a general-purpose engine with 35 deliverables including vectorized solvers, multi-PDE support, and fault-tolerant computing designed for energy-efficient supercomputers. The software is open-source and designed to relieve scientists from building their own exascale codes. You could package it as a managed service for academic and industrial customers.

Frequently asked

Quick answers

What would it cost to use ExaHyPE's simulation engine?

The ExaHyPE engine was released as open-source software, so there are no licensing fees. Your costs would be supercomputer access (cloud HPC or on-premise clusters) and the engineering effort to integrate your specific equations into the engine. The project was funded with EUR 2,795,000 in EU contribution over 4 years.

Can this handle industrial-scale simulations, not just academic demos?

The engine was designed specifically for exascale computing — the largest supercomputers in existence. Deliverables include a petascale prototype release and a final release with dynamic load balancing and multi-parameter support. However, the demonstrated applications are in geophysics and astrophysics research, not commercial products.

What is the IP situation — can we use this commercially?

The final deliverable is explicitly described as a 'Final Open Source Release' with permanent archiving on the project website. This means the code is freely available. You should check the specific open-source license on the project repository to confirm commercial use terms.

How mature is this technology — is it ready to deploy?

The project delivered working software including optimized solvers, dynamic load balancing, and multi-PDE support across 35 deliverables. However, it remains a research-grade engine primarily validated on scientific benchmarks in seismology and astrophysics. Adapting it for a specific commercial application would require additional engineering.

What kind of technical team would we need to use this?

Based on the project's technical scope — exascale computing, Discontinuous-Galerkin numerical methods, spacetree grids — you would need computational scientists or HPC engineers familiar with large-scale parallel computing. The consortium was 5 universities and 1 industrial partner, reflecting the deep technical expertise required.

Is there ongoing development or support after the project ended?

The project ended in September 2019. The final open-source release was archived on the project website (exahype.eu). Based on available project data, ongoing maintenance depends on the academic teams at Technische Universität München and partner institutions continuing development independently.

Consortium

Who built it

The ExaHyPE consortium of 7 partners across 4 countries (Germany, Italy, Russia, UK) is heavily academic — 5 of 7 partners are universities, with only 1 industrial partner and 1 SME (14% industry ratio). Led by Technische Universität München, a top-tier German technical university, the team has deep computational science expertise but limited commercial orientation. For a business considering this technology, the low industry involvement means you would likely need to invest in adapting the research-grade engine for production use, but the strong academic base ensures the underlying mathematics and algorithms are rigorous.

TECHNISCHE UNIVERSITAET MUENCHENCoordinator · DE
STIFTUNG FRANKFURT INSTITUTE FOR ADVANCED STUDIESparticipant · DE
LUDWIG-MAXIMILIANS-UNIVERSITAET MUENCHENparticipant · DE
UNIVERSITY OF DURHAMparticipant · UK
UNIVERSITA DEGLI STUDI DI TRENTOparticipant · IT
BAYERISCHE FORSCHUNGSALLIANZ BAVARIAN RESEARCH ALLIANCE GMBHparticipant · DE

How to reach the team

Technische Universität München, Germany — reach out via their computational science department

Order a report on this consortium See TECHNISCHE UNIVERSITAET MUENCHEN profile →

Next steps

Talk to the team behind this work.

Want to explore if ExaHyPE's simulation engine fits your seismic modeling or HPC needs? SciTransfer can arrange a technical briefing with the development team.

Order a report on this topic More in Digital & ICT