The AI space race is under way as Google officially launched the first AI data center into orbit with some help from SpaceX — sort of. Project Suncatcher is a small satellite equipped with data center components, like Google’s custom Tensor Processing Units, assembled to see if AI can survive the brutal vacuum of space.Elon Musk has big dreams for the cosmos. Earlier this year, he famously unveiled plans to colonize the moon with the first-ever lunar city, dubbed Moonbase Alpha. Once established, it would become a launchpad for other galactic endeavors, including the first manned mission to Mars and orbital data centers meant to power AI platforms on Earth down below.Ultimately, Google wants to find out if its TPUs can survive sustained radiation without hardware failures.At the time, Google’s AI space ambitions were merely a footnote, with CEO Sundar Pichai stating, “We’ll send tiny, tiny racks of machines, and have them in satellites, test them out, and then start scaling from there.” The goal for this test project was originally 2027, but Google jumped ahead by putting its first data center into orbit on October 1.Ironically, Project Suncatcher was ushered to space on a ridesharing mission led by SpaceX. A Falcon 9 rocket — the same that will eventually take mankind to the moon to build Moonbase Alpha and establish Musk’s space data center empire — ultimately delivered Google’s satellite safely to its destination.The mission was deemed a success, and Project Suncatcher is currently collecting data to transmit back to Google.RELATED: The truth about how AI data centers affect local communities Matthew Fowler/Getty ImagesWhen diving into why AI innovators want to put data centers into space, the benefits are clear. According to a peer-reviewed report by Google, the sun generates a staggering 100 trillion times more power than every electrical grid around the world combined. In comparison, space has near-infinite power that can sustain a multitude of data centers without taking energy away from local communities or even each other. This alone is enough to spur spacefaring AI research.The second goal is to see how well data center components hold up in space. The great void beyond Earth’s atmosphere is teeming with radiation. In fact, there are two types of radiation to consider: Ionizing radiation consists of electromagnetic waves with the power to separate electrons from atoms and molecules in everything they touch. In short, they can wreak havoc on computer systems, processing chips, and other components, all leading to a mass hardware failure. Non-ionizing radiation, on the other hand, is less powerful and cannot separate electrons from atoms or molecules, though it can still be harmful in some situations.Google wants to find out if its TPUs can survive sustained radiation without hardware failures. So far, artificial tests here on Earth suggest Google TPUs can withstand up to five years of space-level radiation before failure, resulting in a replacement. The only way to know for sure, though, is to measure their performance in real space conditions.Third, Google is testing a flight system that guides a cluster of AI data center satellites orbiting together instead of one giant data center. The peer-reviewed paper calls them “satellite constellations.” Smaller, modular satellite clusters will ultimately be easier to deploy, maintain, and update as each satellite can be repositioned or pulled from orbit at any time without affecting the other satellites in the fleet.Ultimately, Project Suncatcher is a stress test to see if space-bound data centers are viable. Google will spend the next several weeks gathering data. If all goes well, we could have functioning AI data centers in orbit by the early to mid-2030s.Like Blaze News? Bypass the censors, sign up for our newsletters, and get stories like this direct to your inbox. Sign up here!
SpaceX's newest launch just took the first step to fill the skies with data centers
Full Article
Original Source
Read the full article at Theblaze →KhanList aggregates and links to publicly available news content. We do not host full articles from third-party sources. Always verify important information with original sources.