News: Google unveiled Project Suncatcher to explore space-based AI compute using solar-powered satellites.
About Project Suncatcher
- It is a research initiative exploring how constellations of solar-powered satellites could host data centres in space.
- Initiative of: it is an initiative of Google.
- Aim: To examine whether future AI computing can be powered by solar energy harnessed directly from space and scaled effectively.
- Objective: To process AI workloads in orbit, lowering the environmental footprint of terrestrial data centres while benefiting from uninterrupted solar exposure.
- Working Mechanisms
- Constellation design: Google plans a constellation of compact satellites, each carrying custom Tensor Processing Units (TPUs) to run AI workloads in orbit.
- Power supply: Each satellite draws electricity from high-efficiency solar panels to provide continuous onboard power for computing.
- Data transport in space: Satellites connect using free-space optical communication beams, creating high-speed links between satellites and to ground stations on Earth.
- Networked operation: Data is relayed across the satellite network and then back to Earth, enabling coordinated processing rather than isolated tasks.
- Key Features
- Environmental advantage: Operating beyond the atmosphere avoids many cooling, land, and water needs of Earth-based data centres.
- Throughput targets: Early ground tests achieved bidirectional speeds up to 1.6 Tbps, with plans to scale to tens of terabits per second.
- High power generation: Solar panels on these satellites could generate enough electricity to be eight times more powerful than panels on Earth.
- Radiation tolerance: Trillium TPUs were tested to withstand a total ionizing dose equivalent to a five-year mission without permanent failures.
