AI INFRASTRUCTURE
AI
may be the key to solving some of humanity’ s greatest challenges, but the computational power required to fuel this revolution comes at a staggering environmental cost.
The engine room of AI – the data centre – is grappling with an energy crisis. AI training models push thousands of Graphics Processing Units( GPUs) to their limits for weeks on end, generating enough waste heat to warm a small town. This thermal byproduct is typically treated as an expensive problem to be managed.
But what if that problem was actually a solution?
In East London, an innovative project by Queen Mary University of London( QMUL) is flipping this paradigm on its head. Instead of venting this valuable energy into the atmosphere, they are capturing it to heat their campus, turning a computational burden into a community asset.
When GPUs become the workhorses of AI The partnership with Schneider Electric has created something that sounds almost too good to be true: a data centre that helps the environment rather than just promising to offset its effects later.
“ The project at Queen Mary demonstrates how digital infrastructure can be a catalyst for net zero, allowing today’ s organisations to benefit from the power of advanced computing,” says Mark Yeeles, Vice President, Secure Power division, Schneider Electric UK and Ireland.
aimagazine. com 133