NextSim · Project

Faster Aircraft Design Simulations Using Next-Generation Supercomputing Software

transportPilotedTRL 7

Designing an airplane wing today requires running massive computer simulations that can take days or even weeks — like trying to predict weather for every square centimeter of a plane's surface. The problem is that the software engineers use hasn't kept up with today's supercomputers, like driving a race car in first gear. NextSim rebuilt the core simulation engine (called CODA) from the ground up so it can fully exploit modern supercomputers, making aerodynamic simulations dramatically faster and more accurate. AIRBUS is already adopting this as their new go-to tool for aircraft design.

By the numbers

consortium partners

countries involved (DE, ES, FR)

total project deliverables

demo deliverables including software prototypes and mini-apps

29%

industry ratio in consortium

The business problem

What needed solving

Aircraft manufacturers spend enormous time and computing resources on aerodynamic simulations, but current industrial solvers cannot fully exploit modern supercomputing hardware. This means design engineers either wait too long for results or accept lower accuracy — both of which slow down aircraft development and increase costs. Problems that need to be solved for competitive aircraft design remain computationally unaffordable with existing tools.

The solution

What was built

The project built CODA, a next-generation aerodynamic flow solver using Finite Volume and high-order discontinuous Galerkin schemes, optimized for extreme-scale parallel computing. Concrete deliverables include software prototypes (M18 and M36), optimization software based on feature detection and ROM, and freely distributed mini-apps demonstrating the core algorithms, plus a Knowledge Capture and Preservation System for the aeronautical community.

Audience

Who needs this

Aircraft OEMs running large-scale aerodynamic simulations (e.g., AIRBUS, Dassault, Embraer)Tier-1 aerospace suppliers designing aerostructures and engine componentsHPC centers and cloud providers serving aerospace engineering clientsCAE software companies developing CFD products for industrial usersMotorsport and automotive companies with high-fidelity aerodynamic simulation needs

Business applications

Who can put this to work

Aerospace manufacturing

enterprise

Target: Aircraft OEMs and tier-1 aerostructure suppliers

If you are an aircraft manufacturer struggling with simulation bottlenecks that slow down your design cycles — this project developed CODA, a next-generation aerodynamic solver built for extreme-scale parallel computing. It tackles market-relevant problems that current industrial solvers cannot handle at the required accuracy and affordable computational cost. AIRBUS has already designated CODA as their new reference solver for aerodynamic applications.

High-performance computing services

enterprise

Target: HPC cloud providers and supercomputing centers

If you are a supercomputing center or HPC service provider looking to attract aerospace clients — NextSim's mini-apps and software prototypes demonstrate how to fully exploit streaming processors and many-core platforms for industrial simulation workloads. The project involved 7 partners across 3 countries, including Barcelona Supercomputing Center, proving these methods work at scale.

Engineering simulation software

mid-size

Target: CAE/CFD software vendors

If you are a simulation software company whose products underperform on modern HPC hardware — NextSim developed algorithms for numerical efficiency, data management, and optimized HPC implementation that are available through freely distributed mini-apps. These demonstrate the mathematical methods and algorithms that make industrial-scale aerodynamic simulation practical on next-generation hardware.

Frequently asked

Quick answers

What would it cost to adopt this simulation technology?

The CODA solver is being developed as the new reference solver inside AIRBUS group, so direct licensing terms are not publicly disclosed. However, the project released freely distributed mini-apps that demonstrate the core mathematical methods and algorithms. Access to the full solver would likely require a partnership or licensing agreement with the consortium.

Can this handle full-scale industrial aircraft simulations?

Yes — that is the explicit goal. AIRBUS defined a series of market-relevant problems that current industrial solvers cannot solve at the required accuracy and affordable computational cost. The project delivered software prototypes at M18 and M36 demonstrating the algorithms work on extreme-scale parallel computing platforms.

What about intellectual property and licensing?

The CODA solver itself is positioned as AIRBUS's proprietary reference tool. However, NextSim also delivered mini-apps — small software pieces seeking open-source components — that are freely distributed to the scientific community. The optimization software based on feature detection and ROM was also delivered as a separate package.

What computing hardware does this require?

CODA is specifically re-engineered for extreme-scale parallel computing platforms, including streaming processors and many-core architectures. Based on available project data, the software targets leading-edge emerging HPC architectures, meaning you would need access to modern supercomputing infrastructure or equivalent cloud HPC resources.

Is this production-ready or still experimental?

The project ran as an Innovation Action from 2021 to 2024 and delivered software prototypes at both mid-term (M18) and final stage (M36). AIRBUS has designated CODA as their new reference solver, indicating it has moved beyond experimental into industrial adoption within the AIRBUS group.

Can I access the results without being in aerospace?

The mini-apps are freely distributed and demonstrate the mathematical methods and HPC algorithms developed in CODA. The Knowledge Capture and Preservation System was specifically designed to ensure maximum accessibility and longevity of project outcomes for the broader aeronautical community. Non-aerospace HPC users could benefit from the algorithmic approaches.

Consortium

Who built it

The 7-partner consortium across Germany, Spain, and France brings together the key European aerospace and HPC players. With 4 research organizations (including Barcelona Supercomputing Center as coordinator), 2 industrial partners, and 1 university, the consortium has a 29% industry ratio. The presence of AIRBUS as an industrial end-user defining market-relevant test problems is a strong commercialization signal — this is not academic research looking for a customer, but an industrial pull project where the buyer helped design the product. The three-country spread covers Europe's main aerospace hubs.

BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACIONCoordinator · ES
AIRBUS OPERATIONS SASparticipant · FR
OFFICE NATIONAL D'ETUDES ET DE RECHERCHES AEROSPATIALESparticipant · FR
DEUTSCHES ZENTRUM FUR LUFT - UND RAUMFAHRT EVparticipant · DE
CENTRE EUROPEEN DE RECHERCHE ET DEFORMATION AVANCEE EN CALCUL SCIENTIFIQUEparticipant · FR
CENTRE INTERNACIONAL DE METODES NUMERICS EN ENGINYERIAparticipant · ES
UNIVERSIDAD POLITECNICA DE MADRIDparticipant · ES

How to reach the team

Barcelona Supercomputing Center (BSC-CNS), Spain — one of Europe's leading HPC centers

Order a report on this consortium See BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACION profile →

Next steps

Talk to the team behind this work.

Want to explore how NextSim's HPC simulation methods could accelerate your engineering workflows? SciTransfer can arrange an introduction to the consortium team.

Order a report on this topic More in Transport & Mobility