Google Sends TPU Into Orbit for Project Suncatcher Test
Google launched a prototype satellite carrying one of its Tensor Processing Units aboard a SpaceX rocket from California on October 1, TechCrunch reports. It is the first time Google has sent one of its advanced chips into space.
The satellite, built by Planet Labs, will test whether the TPU can operate under orbital power, cooling and radiation conditions. Once commissioned, it will run models in 15 minute bursts to limit pressure on the satellite's power and thermal systems.
Google and Planet Labs plan to launch two satellites designed for more substantial computing workloads in 2027. The satellites will attempt to work together through a laser communications link, supporting Google's Project Suncatcher research into orbital computing clusters.
Google's research estimates that reducing launch costs to about $200 per kilogram by 2035 would require Starship to carry 370,000 tons into orbit. At 200 metric tons per mission, that would require about 1,800 launches over 10 years, or 180 launches annually.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.
Also, consider following us on social media:
More from Data Centers
Nov 19 Nvidia delays its quarterly SEC filing as the Huang investigation continues Nov 18 Cloudflare releases open-weight Clairic models for extracting data from documents Nov 18 Nvidia scandal complicates US-China AI talks ahead of Trump-Xi meeting, Wired reports Nov 18 Everpure revenue jumps 51% as it wins a second top-five hyperscaler Nov 18 San José pauses data centre approvals while it writes new standardsSilicon Brief
Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
Oracle to Offer Google TPUs Through OCI in 2027
Google adds TPU 8tx for confidential inference and TPU 8i Max for long-context work
Pichai Says Google Sacrifices Cloud Revenue for Frontier AI
OpenAI Plans to Buy Google's AI Chips, The Information Reports
NeuralSat raises €2.9M to build a data relay and compute network in orbit
Daily AI Brief: the AI news that matters, in your inbox.