 Live from the Sands Convention Center in Las Vegas, Nevada, it's theCUBE at AWS re-invent 2014. Brought to you by headline sponsors, Amazon and Trend Micro. Okay, welcome everyone. We are live in Las Vegas for Amazon Web Services re-invent conference. This is theCUBE. We go out to the advanced district to see the noise again on our fifth season, second year now doing the Amazon events. And again, we bring the anchor desk, we bring the cube here to extract the signal from the noise and I'm joined this week by three awesome people, David Floyer, who will be coming on. He's an analyst at Wikibon. Stu Miniman, my co-host and Jeff Kelly, big data analyst and big data angle here as well as a lot of infrastructure. Again, the full stack. This is Amazon's disruption and we're here on the ground covering all the action keynotes today. So Stu Miniman, Jeff Kelly here. Let's get right into it. What a great show. We love bringing the cube. And I got to say in watching the keynote, they highlighted MLB, it's a lot of slew of the announcements but we'll get to some of the meaty ones. But the most favorable for me was the MLB ones too because we love sports, we call this the ESPN of tech and Andy Jassy is a huge sports fan. So bringing the cube down, the ESPN of tech if you will was fun. But to see the tech innovation going on in the cloud, the agility, the fact that the baseball example hits home for consumers is amazing. They give a great demo around stat free. But showing the graphics, showing the iterations, showing the real time data. And then they follow down with not so much a crowd pleaser but relevant, the hospital cancer thing. So you got two inch inspections, play and medical. So this brings the things, brings everything together. Amazon is rolling out a serious platform, Stu. So I want to get your take on one, what you think, what's going on with Amazon and let's get into some of the announcements. Yeah, John, great point, great to be back. And wow, what a difference a year makes. You look at, I mean last year's show there was a lot of buzz, it was really exciting to be here. But look at this show floor, it's twice the size it was. Last year it felt like everybody had little card tables and everybody was a mobile developer or some little gaming thing. And you look around here and it feels like an enterprise show, John. The stat was they sold out, capped it at 13,500 people. And the keynote was still a little bit of a mix of the traditional cloud companies, media company like Major League Baseball, people that are going to use the analytics, use lots of video, that's kind of traditional Amazon workload. Then you have somebody on like Johnson & Johnson, company that had huge data centers and looking to get the agility by using Amazon. So it's definitely, it's much bigger, a ton of announcements. I tell you, when we got the analyst session yesterday we just sat there like we had drank from the fire hose, felt like my brain was leaking out by the end of the day. So many different pieces, so many things. And as Dave Vellante said last year, it is rarely rare that you have the company that is the gorilla in the marketplace like Amazon that is also driving forth new innovations. They move fast and agile like a cheetah. When we talk to their executives, they say they keep the innovation going by keeping small teams, those what they call the two pizzas should be able to feed a group when they work on it. And they're loosely coupled. So keep building on what they're doing, keep adding new feature functionality, keep going after new markets. Like I mean the big announcement today is the Aurora database, which I know we're going to get into. Yeah, so Stu and Jeff, I want to just drive into some of the big announcements. But step back for a second and really let's discuss the theme of the show. Amazon obviously has made their bones in the public cloud, helping developers. And now you're starting to see what we've been tracking on theCUBE and what you guys have been detailing out in the research at Wikibon is the convergence and the confluence of big data in the cloud workloads. Again, the blurring of the lines, Jeff, around big data and cloud and the onboarding of this new consumption. Stu, I noticed the tweet you did in the keynote among your awesome tweets. One was the interview we did with Phillips. This speaks to the new consumption. So Stu, I want to go to you first and I want to get Jeff's perspective on the big data. What is this new consumption model? And how does that impact the existing traditional players like an Oracle, like an IBM? I mean, Oracle's under threat with this new Aurora announcements we'll get into. But what's the consumption pattern? What's the big mega trend? Yeah, so John, the big quote that we use out of that Phillips interview is, they said that 85% of their IT was undifferentiated heavy lifting. And that is the sin in IT today is that we spend way too much time building bespoke architectures for specific applications. For most companies, IT is not a differentiator. There are places where it absolutely can drive the business but how much do I need to really have those geek knobs and tweak everything versus where can I take advantage of services, where can I just go to a new way of doing things and change from really being an administrator to being an architect and someone that can just deliver value back to the business. And that's what Amazon's been building up. They started with their infrastructure as a service and then they're building other services on top of it. I knew just going to talk about the bread shift and everything else. Yeah, but before we get to Jeff, you saw a cloud, they got the cloud migration tools. You got Aurora, the new high performance database. You have cloud trails, there's a lot of the auditing. They're kind of leaving the breadcrumbs to the enterprise, right? So you're seeing this talk about another full scale migration global in two months with Intuit, that is the theme. You're starting to see the, not totally taking the gloves off on the enterprise but certainly stepping into the ring. Yeah, well John, one of the things they actually were maybe a little bit strong against the enterprise. They're like, you do things wrong. You suck and you need to come to the cloud because this is the way IT should be done. And today they're a little bit more embracing. It's hybrid IT and we're going to do certifications. I mean, one of the reasons that early cloud people said is we don't want people that spend their entire career doing the yearly certifications updating here. There's a DevOps certification here, John. I mean, Andy Jackson, a big sports fan, we know him pretty well and have a one-on-one. He will use a sports metaphor. Amazon is moving the goalposts. I mean, right when Amazon is thinking that they have a good comfortable lead, a slew of new announcements come out. These new services, they're constantly shipping new products, new services, and they're shifting the goalposts, Jeff. They're like, just when EMC or Oracle or other guys, they can get into the game with Pivotal or with Cloud Foundry, you're thinking, oh, great. We're going to get back and we're going to win the game. And also, the goalposts change and they're kicking into the stand. So what's your take, Jeff? What is your analysis here so far? Obviously, a big data consumption piece of the cloud is here. What's your analysis? Well, I think it was interesting that in the keynotes this morning, two of the customers they brought up on stage, Johnson & Johnson and Phillips Healthcare, both talked about how they're using big data analytics services from AWS to drive their business. It's all about making better decisions with the data you're collecting from your systems, your transactional systems, your machines, whatever it might be. The interesting thing, I think, for AWS and their play in the big data space is that one of the challenges we hear consistently from practitioners and members of the Wikibon community around big data is that it's still very difficult to get the systems operating in a way that is effective for the business. IT is doing a pretty good job of standing up to the clusters, making sure they're running, but then actually getting the business users the data they need, the tools they need to manipulate that data and then operationalizing some of those insights is a challenge. And what AWS does, one is it abstracts away some of the, a lot of the complexity around standing up systems, but it also provides a number of ways to deliver that content, deliver those insights to business users because, and do it in a way that's much faster than in the traditional way because from an IT perspective, you're not worried about the infrastructure, you're worried now about how do I get this in front of people that are making decisions. So, okay, so I got to ask you the, I got to ask you the hard question here because Hortonworks has filed their S1 to go public, Cloudera did the big Intel deal. So, obviously there's no real IPO market, it seems the material, well, there is one that filed the S1, but you're starting to see big data. Will big data as a sector collapse? Do you see big data being a market where there's a standing in a leaderboard, top one, two, three market share leaders? Is Hadoop a one-trick pony in one part of the sector of the kingdom of big data? So, is big data going to be a real business model or is it just a feature of an overall cloud platform? Well, it's definitely an important, I think feature is too strong word, it's a very important component of this new digital fabric that's emerging that we've been researching at Wikibon that also includes cloud, it includes things like DevOps. So, it's not, big data is not the be all end all, it's not a cure all either for a lot of business challenges, but, so let me back up. So, the question around the Hadoop market, Hadoop is a very important segment, foundational technology in the big data stack and that's not going to change, I don't think. That said, there is a market above Hadoop when you look up higher in the stack around analytics, around database, around I think tools to allow people to automate and operationalize insights that's going to also drive a lot of value. You've got to differentiate between the different types of big data. You've got what we're calling inline analytics where you're actually operationalizing insights and then you've got deep analytics which is what Hadoop's all about and they're different markets and I think the real interesting thing is how those two interact. What use cases are markets? Would you say use cases? Well, they're both different, no, they're related use cases but they are different markets. Things like Hadoop is becoming more real time with some of the capabilities that are being driven around Spark and yarn making it more multi-application aware but when you really want to do the operational, well you want to drive operational processes that are making you money or saving you money as a business, at least certainly in the short and mid term that's not going to be done on Hadoop, that's going to be done in operational environments. So they're related but they're different markets in that sense. Look, I think a lot of comments have been made around the Hortonworks S1, they're spending a lot of money, the revenue is smaller than a lot of people thought but this is part of their business model. Their business model is taking a long term view, it's about laying the foundation now, spending the money to build out the partner relationships, the reseller relationships, continue to beef up Hadoop itself and they're laying the groundwork so that when the hockey stick adoption comes, they're prepared for that. Now, if you're an investor and you buy that long term strategy, the spending shouldn't really bother you right now. If you don't buy it, that's not a question. I buy the strategy, so I'll tell you, well this morning I was actually texting DMing Peter Fenton at Benchmark about just specifically what's going on here at Amazon, specifically teasing around Hortonworks. He's the investor at Hortonworks, he's also an investor in New Relic, basically two companies that Peter Fenton has filed to go public in the same week, it's kind of like a VC watershed, well he called himself a waiter, I call him a matriot D, kind of shepherding that thing out there, but Peter Fenton's a smart investor and I think what you're seeing, he's also going to the analyst's day at Twitter today, so you look at companies like Twitter, Hortonworks, some of these big data companies like Cloudera and even Amazon, the most compelling companies in my opinion, I want to get your guys' take on this, are the ones that are misunderstood, right? The misfit toys, if you will. Twitter right now is misunderstood, I think Hortonworks is clearly misunderstood, people are trashing them for their revenue shortfalls that they kind of put out in their S1, but they're not, they're the real deal because this new cloud marketplace, Stu, I want to get your take on this and Jeff, Stu, is the new business model. So what's the analysis? Why is Twitter a winner? What are people missing about Twitter? What are people missing about Hortonworks? Is it the go-to market? Is it the land and expand? I was trying to tease this out with the Illumial guys yesterday. So let me give you my take on Hortonworks. The bottom line is, as I just mentioned, they've got to invest heavily in their early days to lay the foundation so that they're ready to grow. Their product essentially is a subscription support service. That's what it is. That only becomes valuable or interesting to enterprise practitioners, enterprise customers, when they're ready to move from test and dev and pilot projects to full-scale production. And the reality is, there aren't a lot of those companies out there right now that are using Hadoop in full-scale, large-scale production environments. Wikibon's latest survey around adoption found that about 30% of practitioners were using big data in production. Now, but the important copy out of that is that the survey was of early adopters. So that's just early adopters. If you look at the overall enterprise market, I'd say less than, you know, single-digit percentage of companies are using Hadoop in production. When that goes up, that's when Hortonworks is preparing itself to capitalize on that. Now, that said, there's a lot of competition. There's Flutter, there's MapR, and others, but that's there. But if you look at the SaaS model, all the companies that went public like HubSpot has won, they just recently went public. It's a flat curve, and then it kicks up in the hockey stick, classic hockey stick. So, Stu, what's your take on this? Is it the business model? Is it the market evolution? Are you seeing a much more accelerated ramp once they can kick up into the hockey stick, or what's your analysis? So, John, to bring it to Amazon, Amazon definitely is misunderstood. Everybody, because first of all, we don't get the real numbers, right? We all have to kind of squint through that other category on Amazon. People looked at their growth of a little bit of a blip earlier this year because there was a big price cut. I'm not sure if Amazon's misunderstood. They're definitely a little bit underestimated because John, this wave is coming fast. When you talk to the customers here, they're happy with Amazon. They're adding more services to it. Many of them are going kind of all in, especially if they're a company that started in the last five years or so because it is just so much easier to get started, and Amazon is keeping those customers happy. So, it's really easy for people to throw stones and say, oh, well, Amazon doesn't make any money and therefore AWS must be a bad business. They're throwing three to four billion dollars into infrastructure every year. They're adding into their key markets and they're adding the services that their customers are asking for. So, it is impressive to watch them move. And if I was in kind of the traditional enterprise market, I'd be worried that Amazon is coming over after my business next, if they haven't already. So, the question, Dave, I was just kind of talking about their income statement, but this is ultimately the, is it a flat blade, then hockey stick, I guess something in the shaft? It really is what I'm talking about in here. And my analysis and my take is very simple. The misunderstood companies are the ones that are most likely going to be the unicorns, in my opinion. If you look at it, it's always those misfit toys that people don't truly understand because it's a little bit different. The question is when the accelerated scale kicks in? When does the fly wheel kick in? When's the escape velocity kick in? So, it's a value market. So, when the value is established, then the ramp hits. So, everyone's looking for that quick, bolt on some sales, where's the pop versus the land that expands. So, to me, that's the analysis. So, John, Andy Jackson this morning said that they're growing at over 40% year over year. This is, for a company that's in a multi-billion dollar run rate already, this is quite impressive. They don't seem to be slowing down. There obviously is some tough competition from the likes of Microsoft and Google who keep bombing the price. The question is how far will Amazon go through the race down to zero? One thing, John, we didn't hear, unless I missed it towards the end of the keynote, there was no price drop yet. Google dropped their prices last week and the expectation is that Amazon will match those price cuts. Okay, so we got to get the break here, but before we go, I know we're going to squeeze a little bit into the next section, which is okay, so I want to get this out of the way. Key news here at Amazon, obviously the big presence, showcase customers at MLB, they talked about cancer and medicine, Andy Jassy, literally talking 100 miles a minute, he couldn't spit out all the new announcements. It was as if it was a machine gun of announcements and as Dave Vellante, who's not here, always says it's the horses on the track, if that's the case, then they're secretariat, they're thundering away and pounding and 30 length lead, horse length lead on everyone else. They announced Amazon Aurora, which is their new big database, that's an Oracle killer. This service catalog, we just had a chat on this morning about a new company that's coming out the Woodworks startup called Catalogic Software. We're going to disrupt the backup and data, that's a service catalog, so you're seeing the traditional stuff from backup and recovery to full on compute. Key management services, I think this is going to be a very interesting point, we're talking about security, cyber threats, don't lose your keys, don't slide into first base as we learned on the MLB, and obviously the code deploy, you're starting to see the GitHub, you're starting to see the operationalizing of Amazon into the enterprise. Guys, your comment, give them a letter grade, Stu, A minus, A plus, I think they did pretty good today. Yeah, John, I give them an A minus because there was a little bit too much for us all to swallow. There's so many different announcements and there's more coming tomorrow, some of the really cool stuff they've saved for tomorrow, the big announcement being the database today. John, the question I have is, Amazon is not an open source company, can they win the developers by only giving them an API that's kind of the big looming question that we've had for them for a long time. Does Amazon become, as Andy Jazzy said, the new normal and the environment? Is this the future of IT or will there be enough choice and option? When we talk to practitioners, they're all using Amazon but all of them are using other solutions. Even the guys that are all in on Amazon are also using Google Docs and Salesforce and other SaaS applications. So Amazon's not going to go in all of it and they're trying to say that they've got a big ecosystem and they are partnering with a lot of people here. Jeff, real quick, letter grade and what's the importance of big data at this show? Is it subtle? Is it apparent? Is it front and center? I think it's maybe not front and center but it's just off stage. It's important. As I mentioned earlier, you're hearing from customers and a lot of the use cases they're talking about are big data and analytics. And I think from the perspective of big data, AWS, the cloud makes consuming and creating and delivering on those insights from big data much easier. So we thought of data formation yesterday. Those guys are a new startup. There's a data fabric developing. You see that true? Absolutely, there's so much data being created throughout the enterprise, outside the enterprise and to think that you're going to have a spot solution here or there that's going to solve all these problems, that's the wrong way to look at it. You need a platform that builds out a data pipeline and allows you to get data where it needs to be, when it needs to be there to perform all the different kind of analytics you have to do to be a data driven business and the cloud helps you do that. Okay, we are here live. We're kicking off two days, two and a half days. We had a little mini burst yesterday morning. We did some, yesterday afternoon, we did some interviews. But this is theCUBE. We're live on the ground in Las Vegas at Amazon re-invent. We want to thank Amazon for giving us some space here. One of the shout out for those guys. And if you want to follow the conversation go to crowdchat.net slash re-invent. We get the threaded conversations there recording all the comments on the record. If you want to be on the record go to crowdchat.net slash re-invent. I'm John Furrier with Stu Miniman and Jeff Kelly will be right back with our first guest after this short break.