 from Las Vegas. It's theCUBE, covering AWS re-invent 2018. Brought to you by Amazon Web Services, Intel. And their ecosystem partners. Hey, welcome back, everyone. LiveCube here in Las Vegas for AWS re-invent. I'm John Farrier with my coach Dave Vellante. Day three of wall-to-wall coverage, holding our voices together. Exciting for our next guest, Shane Hawthorne, general manager at AWS for the exciting project around the ground station, partnership with Lockheed Martin. Really kind of an outside-the-box announced on Tuesday, not at the keynote, but this is a forward-thinking, real project where satellites can be provisioned like cloud computing resources. Totally innovative and will change the nature of edge computing, feeding connectivity to anything. So thanks for joining us. Thank you guys for having me. You're right, my voice is going out this week too. We've been doing a lot of talking. Great service. If this is really compelling because it changes the nature of the network, you can feed connectivity, because power and connectivity drive everything. Power, you got battery. Connectivity, you got satellite. Totally obvious now that you look at it, but not before this. Where did it come from? How did it all start? You know, it came from listening to our customers. Our customers have been talking with us and they had a number of challenges in getting the data off of their satellites and down to the ground. So we listened to these customers and we listened to the challenges they were experiencing in getting their data to the ground, having access to ground stations, having the ability at the network level to move the data around the world quickly to where they wanted to process it. And then also having a complex business process logic and other things that were required to help them run their satellite downlinks and uplinks. And then finally, the ability to actually have AWS services right there where the data came down into the cloud so that you could do great things with that data within milliseconds of it hitting the ground. So it's essentially satellite as a service with a back end data capability, data ingestion analytics and management capability that had that idea come about. I mean, it just underscores the scale of AWS and I'm thinking about other things that you might be able to, where did the idea come from? How was it germinated? Well, and actually let me just say one thing. We actually would call it ground station as a service. It's the ground station on the surface of the earth that communicates with the satellite and allows us to get the data off the satellite or send commands up to it. And so like I was saying, we came up with the idea by talking to our customers. And so we went into, I think this is like an incredible part of working at Amazon because we actually like followed through with our leadership principles. We worked backwards from the customer. We actually put together a press release in a frequently asked questions document, a PRFAQ, in a traditional six page format. And we started working it through our leadership and it got all the way to the point that Andy and the senior leadership team within AWS made the decision that they were going to support our idea and the concept and the architecture that we had come up with to meet these customers' requirements. We actually were able to get to that by about March of 2018. By the end of March, Andy had even had us go in and talk with Jeff. He gave us the thumbs up as well. And after six months, we've already procured 24 antennas. We've already built two ground stations in the United States and we've downlinked over hundreds of contacts with satellites, bringing earth imagery down and other test data to prove that the system works. Get it ready for preview. Some believe because you're basically taking the principles of AWS, which is eliminating the heavy lifting, applying that to building ground stations, presumably, right? So the infrastructure that you're building out, partners that you're working with, are there critical players there that are enabling this? Yeah, it's really neat. We've actually had some really great partnerships both with helping us build AWS ground station as well as partners that helped us learn what the customers need. Let me tell you first off about the partnership that we've had with Lockheed Martin to develop a new innovative antenna system that will collaboratively come together with the parabolic reflectors that AWS ground station uses. They've been working on this really neat idea that gives them the ability to downlink data all over the entire United States in a very resilient way, which means if some of their ground station antennas in verge don't work due to man-made reasons or due to natural occurrences, then we're actually able to use the rest of the network to still continue to downlink data. And then we complimentary bring in AWS Astra for certain types of downlinks and then also to provide uplink commanding to other satellites. The other customer partnership that we've worked with was working with the actual customers who are going to use AWS ground station like Digital Globe, BlackSky, Compelisar, Hawkeye360, who all provided valuable inputs to us about exactly what do they need in a ground station. They need the ability to rapidly downlink data. They need the ability to pay by the minute so that they're actually able to use variable expense to pay for a satellite downlinks, instead of capital expenses to go out and build it. And then by doing that, we're able to offer them a product that's 80% cheaper than if they'd had to go out and build a complete network similar to what we built. And they're able to, like I said before, access great AWS services like recognition or SageMaker so that they can make sense of the data that they bring down to the earth. It's a big idea and I'm just curious as to how and if you sort of validated it, how'd you increase the probability that it was actually going to deliver a business return? Can you talk about that process? Well, we were really focused on like validating that we could meet customer challenges and really give them the data securely and reliably with great redundancy. So we validated first off by we built our antennas into ground stations and into preview software. We finished over a month and a half ago and we've been rigorously testing it with our customer partners and then letting them validate that the information we provided back to them was 100% as good as what they would have received on their own network. And we tested it out and we've actually got a number of pictures and images downloaded over at our kiosk. They were all brought in on AWS ground station and it's a superb products over there. So Shane, how does it work? You write this press release, this working backwards document. Describe that process, was that process new to you? Had you done it at other companies? How did you find it? Was it a useful process? Obviously it was because you got the outcome that you were looking for, but talk a little bit more about that approach. Yeah, I mean it's actually very cool. I've only been at AWS for a year and a half. And so I would say that my experience at AWS so far completely validates working backwards from customers. We were turned on to the idea by talking to our customers and the challenges they said. I started doing analysis after the job was assigned to me by Dave Nolton, my boss, and I started putting together the first draft of our PRFAQ, started engaging with customers immediately. Believe it or not, we went through 28 iterations of the PRFAQ before we even got to Andy. Everybody in our organization took part in helping to make it better, add in, ask hard questions, ensure that we were really thinking this idea through and that we were obsessing on the customer. And then after we got to Andy and we got through approving that, it probably went through another 28 iterations before we got to Jeff. And then we went through talking with him. He asked additional hard questions to make sure that we were doing the right thing for the customer and that we were putting together the right kind of product. And finally, we've been iterating in on it ever since until we launched it a couple of days ago. Sounds like you were iterating, raising the bar, and it resonated with customers. Totally. And even as part of getting out of it. That sounds language of love. And then your engineering resource, you know, if people are asking you hard questions, you obviously need engineering folks to validate that it's doable. At what point do you get that engineering resource? How does that all work? Well, it's neat. In my division, Region Services Division, we actually were supporting it completely from within the division all the way until we got approval from Andy. And then we actually went in and started hiring very good skills to show you what kind of incredible people we have at Amazon. We only had to hire about 10% space expertise from outside of the company. We were actually able to bring together 80 to 90% of the needed skills to build AWS ground station from people who've been working at Amazon.com and AWS. And we came together. We really learned quickly. We iterated, failed fast, put things together, changed it, and we're able to deliver the product in time. Whole cloth made from our own expertise. So just to summarize, from idea to actual, we're going to do this, how long did that take? I'd say that took about three months. From idea to making a decision three months, from decision to have a preview product that we could launch at re-invent six months. That's unbelievable. It is. You think about something of this scope. And it was a joy. I mean, it was incredible to be a part of something like this. It is the best work I've ever done in my life. Yeah, space is fun. It is. Shane, thanks for coming on theCUBE, sharing your story and insight. We love this. We're going to keep following it. And then we're going to see us at the public sector summits and all the events you guys are out. So looking forward to seeing and provisioning some satellite. I'm looking forward to showing you what we do next. So thank you for having me. We're going to speak peak. This is theCUBE here in Las Vegas. We're back with more coverage after this short break.