Report: Google's Finalizing a $1 Billion Investment in SpaceX Internet
It looks like Google is going to throw a ten-figure sum into Elon Musk's space internet dream. Both The Information and The Wall Street Journal report that the search giant is finalizing a $1 billion investment in SpaceX's recently revealed effort to offer global internet service through small satellites. It's a crazy idea that just might work.
Musk recently revealed some details about his as yet unnamed space internet project to Bloomberg. A team of about 5o employees at the new SpaceX office in Seattle are working on a project that would use about 700 small satellites to provide internet access to the entire planet, with a strong focus on rural and developing areas. SpaceX declined to comment on the plans and the Google investment in an email to Gizmodo.
It's a popular and ambitious effort in which other billionaires like Richard Branson and Mark Zuckerberg are also investing sizable sums. (Branson likes the satellite idea, while Zuck is taking a drone-focused approach.) Google itself has been thinking about building a balloon-borne space internet for years, though most people think that the so-called Project Loon is a bit loony—or just a UFOs.
Sooooo… Who will be the first billionaire to build a space internet?! A better question: Who would've believed ten years ago we'd be asking that question in 2015? Not this guy. [WSJ, The Information via Gizmodo ES]
Image via SpaceX
The following replies are approved. To see additional replies that are pending approval, click Show Pending. Warning: These may contain graphic material.
Can't be any worse than Huges net....400ms delay.
As one of the proud builders of HNS-1 Jupiter (and the recently acquired TerreStar-1) I can't help but agree with you on this. By 2018 you'll have two more Jupiters up there so...hang tight.
Proud builder? Of a satellite? You lucky bastard.
It was fun while I was there but I was one of about 200 people with their hands on that bird. Still kinda amazing that you worked on something that is 30,000+ miles above you.
Hanging tight won't matter. You're not going to improve Latency which is the real killer. Unless Musk is planning on installing sats in a low orbit that might only last a year or two, standard Geosynchronous insertions for satellites is I think 26,000 miles? Which means round trip you're sending data 46,000 miles. You're going to have a minimum latency of about 500ms at best. Typically, my sat connections hitting 1000ms (1 second) aren't uncommon.
This is a great idea that someone is willing to do this to support rural areas, but I doubt even if they offer a gig per subscription it will come close to 100mbit broadband in observed performance (they will have to offer it at prices far, far below that of Google fiber imo). But for the copper/fiber connected? This will >not< be a replacement.
I suppose you could change everything so such a small MTU wasn't required. Larger MTUs would reduce the number of packets sent, thus reducing the number of packets actually sent (but increasing the odds of requiring a retransmit), this would reduce the effects of latency. If I were betting, I'd say they'd almost certainly would have to encapsulate the entire stream within a ipsec VPN tunnel just to get consistent and controllable packet sizes.
The delay will make it slow.
No comments:
Post a Comment
Please leave a comment-- or suggestions, particularly of topics and places you'd like to see covered