High Performance Computing
Supercomputers accessible to (almost) all.
© Inria / Photo C. Morel
Supercomputers and digital simulation lie at the heart of the research carried out by the Datamove project team. They are cutting-edge technologies that, in 50 years' time, should become more generalised and used in numerous professional fields. This is what Bruno Raffin, the head of the team, predicts.
Bruno Raffin: The Datamove team focuses on HPC, high-performance computers. This is what enables very large-scale digital simulations on enormous machines to be carried out. In 50 years' time, we can almost certainly say that these machines will still exist, and on an even larger scale.
Their use will most certainly evolve. Today, digital simulation remains quite limited to a few areas of application. We can imagine that, in the coming years, digital simulation will become - for the scientist and the engineer - a commonly used tool that will be interconnected with other experimentation instruments. Let’s take the example of the electronic microscope. Combined with a supercomputer, it will be able to interactively process the acquired data, identify areas of interest, and superimpose finer molecular models - that cannot be seen under the microscope - on them.
Today, supercomputing takes place almost exclusively in dedicated centres or within large companies with the resources and expertise to use the supercomputers.
I believe that these machines will become even more generalised and be of use in new fields.
For example, simulation will be used in urban planning and, as a result, to assess the impact of the planned developments on mobility, urban growth and the environment, a lot more quickly and accurately than is currently the case. A mayor will therefore be able to dispose of quite simple tools in order to plan the urban policy of his/her municipality.
Agriculture is another field where digital simulation is still quite rare, when it could make major contributions in order to go from mechanised agriculture as we know it to digital and robotic agriculture enabling, for example, a reduction in inputs.
Medicine should also be affected. There are many highly-sophisticated measurement devices, but it remains static. Combining these devices with digital simulation tools should make it possible to observe the potential evolution of an injury or illness. A doctor will be able to simulate – dynamically - prosthesis on a model of the patient, built using the different data available. The aim is to have a made-to-measure prosthesis that is perfectly adapted to the person's morphology.