Low-Carbon Computing Platform from Old Smartphones
Google/DeepMind
University of California San Diego
Researchers at the University of California, San Diego, with support from Google, are creating computing clusters from motherboards of retired smartphones. They plan to deploy a data center of 2000 Pixel devices to provide low-carbon cloud computing, reducing the need for new hardware and cutting carbon emissions.
The carbon footprint of computing equipment consists of operational emissions (energy consumption) and embodied emissions (equipment manufacturing). Researchers at the University of California, San Diego, with support from Google, are developing a method to extend the lifespan of smartphones: motherboards are extracted from used phones, combined into clusters, and used as a universal computing platform. The university plans to deploy a data center of 2,000 Pixel smartphones, which will provide hundreds of researchers and students with low-cost, low-carbon-footprint cloud computing. The performance of a single modern smartphone processor is comparable to a server core; a cluster of 25–50 phones is equivalent to one modern server according to SPEC tests. Before deployment, batteries, displays, and cases are removed from the phones, and Android is replaced with a standard Linux system. Tasks are orchestrated using Kubernetes in containers. A pilot cluster of 20 phones has already demonstrated its ability to handle peak loads for a class of 75 students with latency no worse than Amazon Web Services. The full system of 2,000 devices will go live in fall 2026.
Source: Google Research —
original
