https://www.reddit.com/r/spacex/comments/gxb7j1/we_are_the_spacex_software_team_ask_us_anything/What kind of CPUs does Crew Dragon run in comparison to regular desktop CPUs? I know that there are multiple CPUs for redundancy but how would one of those units compare to, let's say, a desktop i9 9900k?
We use a dedicated quad core processor, similar in power to a 5 year old phone. - Sofian
For some level of scope on Starlink, each launch of 60 satellites contains more than 4,000 Linux computers. The constellation has more than 30,000 Linux nodes (and more than 6,000 microcontrollers) in space right now.
What is the single craziest/most impossible thing management (aka Elon) has asked you to do?
I recall for F9-14 I was in Elon's cube telling him the news that there was no way we could get all of the new S1 landing code done in time for the upcoming launch in 2 weeks. After some thought, he looked over to Lars Blackmore who was there with us and asked if we implement the code, what was our probability of landing. Lars said around 90%. Paraphrasing, Elon looked at us and basically said "can you give me 50%". I said in 2 weeks we can definitely write enough of the logic to get to a 50% probability of landing! We didn't land F9-14 (you can see it on our blooper reel) but we learned a LOT from it, and it was instrumental in eventually landing F9-21. A critical part of our success is our willingness to fail in ways that won't compromise the mission, as long as we are constantly learning from our failures. - Jeff
https://iss-sim.spacex.com/On this kind of project, pace of innovation is everything. We've spent a bunch of time making it easier, safer, and faster to update our constellation. We tend to update the software running on all the Starlink satellites about once a week, with a bunch of smaller test deployments happening as well. By the time we launch a batch of satellites, they're usually on a build that already older than what's on the rest of the constellation! Our ground services are a big part of this story as well – they're a huge part of making the system work, and we tend to deploy them a couple times a week or more.