How does a supercomputer help Ducati build a 2027 MotoGP bike from scratch?

The 2027 rulebook rewrites the machine

MotoGP will introduce 850cc engines in 2027. The current 1000cc units disappear. Aerodynamics must be reduced. Ride-height devices — the hydraulic systems that lower the rear under acceleration and the front under braking — are banned. Fuel becomes 100 percent sustainable. A new tire supplier enters the paddock. Ducati Corse calls it the biggest technical reset in over a decade.

Every major element of the Desmosedici GP needs a fresh approach. The engine architecture changes. The aero philosophy changes. The thermal load changes. The tire interaction changes. At the same time Ducati must keep developing the 2026 bike to fight for this year's championship. Two completely different development programs run in parallel. One team. One factory. Two bikes that share almost nothing.

A cluster that runs two seasons at once

Lenovo expanded the Ducati Corse HPC infrastructure. The cluster now delivers over 108 TFLOPS of combined theoretical double-precision computing performance. Compared to the previous generation of compute nodes the latest expansion roughly doubled the available memory and added around 48 percent more processing power.

Engineers can now run more complex simulations. They evaluate a wider range of design ideas. They reduce the time required to turn analytical results into parts ready for on-track testing. The 3D engine CFD simulation times dropped by 50 percent. Structural analysis times dropped by 30 percent.

Only the most promising designs progress to the prototype stage. Physical manufacturing happens later. Development time shrinks. Manufacturing costs shrink. Accuracy and reliability rise.

When the wind tunnel is a screen

MotoGP's concessions system limits on-track testing. Simulation and predictive engineering fill the gap. The enhanced HPC infrastructure enables Ducati engineers to run increasingly sophisticated — and therefore more computationally demanding — simulation models. These models provide a more accurate representation of real-world conditions. They give a deeper understanding of the impact of aerodynamics on overall vehicle performance.

The new regulations require a completely redesigned engine. Computational power becomes vital for optimizing the intake and thermal systems, connecting rods, and the crankshaft. Halving computation time means more design iterations virtually before any metal is cut.

Where the data runs out

The announcement does not say how many engineers work on each program. It does not name the new tire supplier. It does not quantify how many design iterations were run versus previous seasons. It does not reveal whether the 2026 bike development has already peaked or whether resources will shift further toward 2027 as the season progresses. The concessions system that limits testing is mentioned but its specific allocation for Ducati in 2026 is not detailed.