 From Gillette Stadium in Foxborough, Massachusetts, it's theCUBE, covering VTUG Winter Warmer 2019, brought to you by SiliconANGLE Media. Hi, I'm Stu Miniman, and this is theCUBE at VTUG Winter Warmer 2019 at Gillette Stadium, home of the New England Patriots, the AFC championship team going to the Super Bowl, third year in a row, yet again, Randall, right? Yep, playing against my Los Angeles Rams. So, happy to welcome to the program Randall Hunt, who's a software engineer with AWS. Did a keynote this morning. I believe it was 100 AWS features in 50 minutes and felt like you added a couple more than 100 and went a little over 50 minutes, but... I think we probably hit 57 minutes. That was what the slide counter said, but yeah, I added in a couple of the updates since re-invent, you know, re-invent is not the end of our innovation. We continue releasing new stuff after that. All right, so our program, we're not gonna be showing JavaScript. We're gonna take a deep breath and slow down a little bit because, you know, our audience absolutely knows Amazon. I tell you, this show, remember like four years ago, first time AWS presented, we actually had Microsoft and AWS here and people heard Cloud 101. And I was like, come on, I could've given this presentation and they were walking around like, oh my God, I just found out that, you know, horseless carriages and I can do them and things like this. So, you know, Cloud, we've been there for a decade, but we're still, I believe, you know, day zero, day one is what Amazon always likes to say. It's always day one. It's always day one. So, there's no way we can shove the entire re-invent and keynote into this discussion. So, you know, let's start first. Tell us, Rand, a little bit about yourself, your role, what you work on and what customers you talk to. Sure, so I studied physics and then I found out physicists don't really make any money. So I became a software engineer and I worked at NASA, I worked at SpaceX and worked at this company called MongoDB. Back then it was called Tengen. And then I'm at Amazon. It's my second time around at Amazon. I'm a software engineer there but I'm also a technical evangelist and what that means is I get to travel around the world and make all of the demos and chat with all of our customers and kind of solicit feedback from them and then kind of try to act as the voice of the customer for the service teams. Whenever I can get them to listen. Yeah, so probably not going to go into open source versus software licensing of things with you because we want to make sure that we can publish the interview, but I tell you, space is one of those things. Oh, I love it. When I've interviewed people that have been in space, I've talked to lots of companies that have our code in space. Amazon, you have, I loved, you know, robotics and space are hard and we make it easy and I kind of laugh because I was an engineer as an undergrad. I studied a little bit of, you know, what it takes to break gravity and understand and I always love watching, you know, all the shows about space and track SpaceX, would you work for and things like that? Give me a break. You haven't made space easy. Well, I think space as a whole is getting easier. This industry is becoming more approachable. One of the things that we launched to reinvent this year was ground station and this is something where if you have an S-band or UHF satellite in Leo, which is low Earth orbit or MEO, which is medium Earth orbit, you can basically downstream that data to one of these ground stations, which is essentially attached to a region. In this case, USC's two, which is an Ohio area, and you can go and say, hey, just stream this data into S-3 for me or let me access this from my VPC, which is pretty gnarly if you think about it. You have an IP address, which is a satellite in space. Yeah, I love, I worked on replication technology 15 years ago and it was like, okay, can the application take the ping off the satellite or how do we do this? So, look, we're leveraging satellites a little bit more. I understand it's a great tagline to make those useful and more readily. Just, you know, it's amazing. You think about, when you think about my availability zones and regions, now, you know, things aren't just on the terra firma. Well, I'm looking forward to the first availability zone on the moon or on Mars. That'll be, you know, when we have Utopia Planitia 1A, that'll be, you know, the really cool AZ. All right, we heard of first blue origins working on taking AWS to Mars. No, but... Well, the latency, you know, you have 350,000 kilometers on average between the Earth and the moon. So, you know, you can go around the Earth with speed of light, 7.5 times every second. To go to the moon is a full, I think it's like six, seven seconds or so. So the latency requirements become a little bit harder there. Right now, where am I wrong pin? I have the Grace Hopper nanosecond pin. Which is the... Which is, you know, curled up and if you follow the white thing, it's how long light would take to travel that and it does it in a nanosecond. So, you got me, I'm a physics lover and love space as does a lot of our audience. So bring it down to the thing, one of the things that Amazon has done really well is I don't need to be a physics geek to be able to use this technology. We're having arguments as to, you know, if I'm starting out or if I want to restart my career today, do I go code or heck, you know, let me just use Lambda and all these wonderful things that Amazon have and I might not even need to know traditional coding. I mean, when I learned programming, you know, it was you learned logic and wrote lines of code and then when you went to coding, it's pulling pieces and modifying things and in the future, it seems like serverless goes even further along that spectrum. I definitely think there's opportunities for folks who have just, you know, I don't want to say modest coding abilities, but people who are kind of, you know, industry adjacent, scientists, app, you know, data scientists, folks like that who may not necessarily be software engineers or have the, they couldn't recite big O notation for merge sort and things like that from scratch, you know, but they know how to write basic code. There's a lot of opportunity now for those developers and I'll call them developers to go and write a Lambda function and just have it accomplishing a large portion of their business logic for their whole company. I think the, you know, you have a spectrum of compute options. You have, you know, EC2 on the one side and then you have containers and then as you move towards serverless, you get this, you know, spectrum between Fargate and Lambda and Lambda being the chief level of abstraction, but I think in a couple cases, you can, you know, even go further than that with things like Amplify, which is a service that, well, it's an open source project that we launched and it's also a service that we launched and it takes together a bunch of different AWS services, things like AppSync and Cognito and Lambda and it merges them all together with one CLI call you can go and say, hey, spin up a static site for me, like a Hugo static site or something and it'll build the code pipeline, build all that stuff for you without you having to, you know, worry about all the stuff. And if developers are starting new today, you know, I remember when I started, I really had to go deep on some of the networking stuff, you know, I had to learn all these different routers and like how to program them and these are like the industry routers, you know, the million dollar ones and having to rack and stack this stuff and the knowledge is not really needed to operate a large scale enterprise, you know, if you know a route table and you know VPCs, you know, you can run, you know, a multi-billion dollar company if you want. Yeah, it's been interesting to watch too and, you know, I think the last five years, the proliferation of services in AWS got to a point where it was like, oh my gosh, if I wanted to kind of configure a server for my data center or configure an equivalent, something that I wanted at AWS, there was more choices in the public cloud than there was there and people were like, oh my gosh, how do I learn it? How do I do this? But what we start to see is it's more, I don't need to do that because what do I want to do if there's an application that I can run or services that will help make it easier for me to do that because the whole, it's not let me replicate what I was doing here and do it there, but I have to kind of start with a clean sheet of paper and say, okay, well, what's the goal? What data do I need? What applications do I need to build and start there? I'm curious what you see and how do you help companies through that? So this is a really common scenario. So this is kind of a key point here is enterprises and companies have existed since before the cloud was really around. So why do we keep seeing so much uptake? Why do we keep seeing so many customers moving into the cloud? And how do we make it easier for customers to get into the cloud with their existing workloads? So along that same spectrum, if you have Greenfield projects, if I were running my own company and I were doing everything, I would absolutely start in the cloud and I would build everything as kind of cloud native. And if you want to migrate these existing workloads, that's part of one of the things that we launched this year in partnership with VMware is VMware kind of interface for AWS. So you can use your native vCenter and vSphere kind of control plane to access EBS, to access Route 53 and EC2 and all the other kind of underlying stuff that you are interested in running. And I think he could even do RDS on VMware in my environment. So that line is definitely blurring between my stuff and my stuff somewhere else. And when people are talking about migrating workloads, you can take the lowest hanging fruit, the most orthogonal piece of your infrastructure and you can say, hey, let me take this piece as an experimental proof of concept workload and kind of lift and shift it into the cloud. And then let me build the accoutrement, the glue and all the other stuff that kind of is associated with that workload, cloud native. And you'll get additional agility. One ops person can manage this whole suite of things across 19, 20 regions of AWS and it's kind of global availability and all of this kind of good stuff that typically comes with the cloud. And in addition to that, as you keep moving more and more workloads over, it's not like it's a static thing. You can evolve, you can adjust the application, you can add new features and you can build new stuff as you're moving these applications over to the cloud. Yeah, and it's interesting because just the dynamics are changing so much. So there's been, there's still so much movement to the cloud. And then, oh, while some people are pulling stuff back and then you see your AWS outposts. So later 2019, we expect Amazon to have, you know, footprint in people's environments and then, you know, just to make things even more complicated. Well, the whole edge computing IoT and the like, which, you know, everything from snowball and these pieces. So the answer is it gets even more complicated. But, you know, your AWS, I know, is trying to help simplify this for users across the board. I think if I can say anything at all about AWS, it's that if a customer is asking us to build something, we are going to do our best to make that customer happy. We take customer feedback so incredibly seriously in all of our meetings, all of our service team meetings, you know, that voice of the customer is very strong. And so if people are saying, hey, I want AWS in my own data center, you know, that's kind of the genesis of outposts and it's this idea that, well, we have this control plane, we have this hardware, let's figure out how we can get it to more customers and customers are saying, hey, I want it in my data center, I want to just be able to plug in some fiber and plug in some power and I want it to work and that's the idea. Miranda, when I think of every company that I've watched, there's usually something that people will gripe about and what I've been very impressed with Amazon, Amazon absolutely listens and moves pretty fast to be able to address things. And if you see, you know, if I'm a competitor of Amazon, I'm like, oh, well, you know, this is the way that we get in there or, you know, where we think we have an advantage. Chances are that Amazon is addressing it, looking to, you know, move past it and, you know, absolutely the Amazon of 2019 is sure not the Amazon of 2018 or, you know, when you thought about it, you know, 2015 and it's a big challenge for people as to because usually I think of something and you never get a second chance to make a first impression, but it changes so much. You know, everything changes that, you know, I need to revisit it, it's like, oh, well, this is the way I do things. Well, Amazon has five different ways you can do that now, you know, which one fits you best. And I think that's important is different applications can have different characteristics that you want to be able to pull in and run in different ways. You know, honestly, I'm a huge fan of serverless. I think serverless is where a ton of different workloads are going to move into the future and I just see more and more companies migrating their existing, you know, everything from elastic mean stock applications to like, you know, VMware images into the serverless environment. And I like seeing that kind of uptick and someone recently, I can't remember who it was, someone sent me a screenshot of their console with their EC2 instances in 2010. And maybe it was part of this 10 year challenge thing on Twitter where it's 2009 versus 2019, but they sent me, you know, their M1 large and the screenshot of the console from back then. And they sent me a screenshot of 2019 and I was like, wow, things really have changed. You don't really notice it as much when you're using it every day, but I can imagine, you know, they're ops teams where they haven't logged into the console in three years because, you know, everything is done kind of in an automated fashion. They set up their auto scaling group, you know, three years ago and then the only time they ever log in is to update to new instance types or something for the cost savings. And I get messages on Twitter sometimes from people who are like, whoa, console got an update, this is so cool. And then sometimes we get messages from people where, you know, we changed the EBS volume snapshotting things. We had somebody who had, he was like 130,000 EBS snapshots or something and they were like, hey, you removed my ability for me to select multiple snapshots at once. It's like, well, you have 130,000. So we went in into the UI and we added a little icon that works better for large groups of snapshots. You know, if there's a customer pain point, we will do everything we can to address it. All right, Reddle Hunt, really appreciate you sharing with us your experience, what's going on with customers and absolutely that 10 year challenge. We know things change fast. We used to measure in decades. I would say now it's usually more like, you know, 18 to 24 months before between everything, you know, changes. AWS in 2029 is going to be crazy. I can't imagine what it's going to look like then. All right, well, the Cube, we started broadcasting from events in 2010. We appreciate you staying with us through 2019. Check out thecube.net for all of our programming. I'm Stu Miniman, thanks so much for watching theCUBE.