● The study compared the environmental impact of terrestrial and orbital datacentres and sought to validate the technical feasibility of such a project.
● According to Damien Dumestier, the end-to-end systems architect who supervised the study for Thales Alenia Space, if Europe succeeds in designing a launcher that is ten times less emissive over its entire lifecycle, the first orbiting datacentre could be deployed by 2050.
What are the technical arguments for datacentres in space? Can they really provide European companies with a competitive advantage and reinforce data sovereignty?
Damien Dumestier. Most of the datacentres used by Europeans are not located in Europe. The drive to repatriate them will require substantial investment, not only in new centres but also in increased electricity production and distribution capacity, and sufficiently powerful communications infrastructure. Ascend would enable Europe to deploy datacentres without placing any supplementary burden on the continent’s electricity grid. At the same time, space-based datacentres could provide a more robust solution in the event of natural disasters that might disrupt terrestrial services.
Yet another advantage is that European companies deploying datacentres in space would not have to comply with legislation that may undermine digital privacy rights like the Cloud Act, which obliges cloud service providers to share hosted data with US authorities.
Deployment in space could significantly reduce the carbon footprint linked to energy consumed by datacentres
From an environmental point of view the project seems to be something of paradox…
At first glance, locating datacentres in space appears to be more damaging to the environment than leaving them on the ground. However, to look at the big picture you have to take into account the entire lifecycle of these systems, and that is what we did in the Ascend study. When we analysed and compared all the parameters involved in the production, deployment and operation of ground and space-based systems, we found that the operation of terrestrial datacentres, which also draws heavily on terrestrial energy resources, is responsible for 80% of their CO2 emissions. Clearly, if these facilities were deployed in space, we could certainly reduce the carbon footprint of this operational phase. At the same time, there is also a trade-off because we have to be sure that the added environmental cost of deployment in space does not wipe out any operational gains. That is why we concluded that future feasibility will depend on the development of an eco-designed space launcher.
You talk about developing “a launcher that is ten times less emissive over its entire life cycle”. Is it possible to do that?
According to industry representatives, it should be feasible to develop a launcher with an overall carbon footprint that is only one-tenth that of current systems. The Protein study, conducted for ESA, has begun to investigate the feasibility of a reusable launcher that can deploy large-scale in-space infrastructure. However, launchers of this kind don’t yet exist, so they will have to be developed in the coming years. We are talking about a heavy-lift launcher with the capacity to transport 35 tonnes to an altitude of 1,400 km, which can be reused up to 50 times. The plan to develop such an ambitious project could represent a major opportunity for Europe, which would also be in line with the goals of net zero emissions by 2050 and reducing the ecological footprint of the space sector.
What about the risk of collisions?
The orbit selected for the project, which is at an altitude of 1400 km, has a relatively low space debris population. At the same time, the design for the Ascend datacentres is based on a distributed, modular architecture with the various elements connected to each other using optical communication terminals. External communications are to be handled by inter-satellite laser links, and space-ground transmissions will be via RF with support from the on-board optical terminals.
Like all future European space missions, the Ascend system will be equipped with collision avoidance and prevention mechanisms.
What are your plans now that the study has been completed?
The study brought together a consortium of 11 partners from different technical backgrounds and countries. Working in full collaboration for more than a year, this multi-disciplinary team succeeded in demonstrating very encouraging results on the technical and economic feasibility of the project and on minimizing the environmental footprint of space-based datacentres.
Further studies on the validation of estimates and roadmaps for the development of technologies needed for the Ascend project over the period 2024 and 2025 are currently under discussion with various European agencies. Thereafter, we will be initiating demonstration programmes for key technologies as well as the definition and deployment in 2031 of an architectural proof of concept for the space-based datacentres. Once that has been done, we will move ahead with the design and construction phase of an initial space datacentre, which will be deployed by 2036 before the roll-out of the large-scale system. The aim is to deploy 1GW of computing power by 2050.
Picture: conceptual image of the Thales Alenia Space data centre – credit: Thales Alenia Space_MasterImageProgrammes
Damien Dumestier
