 Live from Seattle, Washington, it's theCUBE at Tableau Conference 2014. Brought to you by headline sponsor, Tableau. Here are your hosts, John Furrier and Jeff Kelly. Okay, welcome back everyone. We are here live in Seattle, Washington for Tableau's User Technology Conference. Data 14, it's theCUBE, our flagship program. We go out to the events, extract the signal from the noise. I'm John Furrier, my co-host. Jeff Kelly, analyst with our big data group at Wikibon. And our next guest is Scott Barnison, business development manager at Amazon. We'll be at the Amazon re-invent conference coming up. It's going to be awesome. Great to have you back on theCUBE. Good to see you again. Thanks, John. So, Amazon, born in the cloud is now the buzzword of this year. Been around for a while, but truly you're seeing the acceleration of Amazon's model of cloud-based computing with all the integrated stacks really powering the developer market. Certainly now, even in the enterprise, people talk about Hybrid Club. At the end of the day, you guys have the model. You're doing some BizDev here at Tableau. Perfect fit for you guys, right? I mean, come on, you can't beat someone doing visualization, playing with data and needing resources. So you're here doing some BizDev, shaking hands. Smiling a lot, shaking hands, yeah. So the relationship with Tableau is an important one for us. They're growing really well. And most importantly, if you heard yesterday in the keynote with Christian, he talked a lot about experimentation. And experimentation for us is one of our core messages. If you can reduce the friction required to get your hands on the infrastructure so you can try new things and if you can lower the cost of failure, you'll increase the rate of experimentation. We believe there's a direct correlation between experimentation and invention. Invention's super important for the history and the lifeblood of companies. So Tableau gives this wonderful view into the customer, right? We have the back end infrastructure, they have the view into that infrastructure to extract the value and the insight. So it's very symbiotic, a great partner for us. We're very happy to do it. I mean, the transformation is obvious, right? So the freemium models on the business side for application developers, clearly working, give someone some taste and they get addicted or they learn, they use it. Tableau's a great example, that perfect example. You guys have the same model. You have one server, you have three, now you got all those other goodness there in Amazon. So the question I have for you is, how tightly coupled are you guys with Tableau? Because as a user of Amazon, of course, we have our crowd chat and our back end analytics all running on Amazon, awesome, awesome service. But I want to push a button and get Tableau. I don't want to have to do a lot of work. So what are you guys doing to kind of get more tightly coupled with Tableau? Yeah, it's a great question and we've been getting that feedback here at the show. Today we talk about a couple different stories with Tableau. One is the back end integrations to our data services. So you take Tableau desktop on the front end, pick your data source on the back end, whether it's the relational database service, elastic map produce, redshift, et cetera, and plug in and get access to the data. The other side of that is running the Tableau server on EC2 on the compute service. Something that I've heard repeatedly at the show, this is a way for folks to get their reports out to the end users in a scalable way at the time when they need it the most. So I think there's a strong story today. We have some technical assets to help people get started and have a good experience. Over time I think what you'll see is we're going to continue to focus on enabling and empowering the users to do this stuff themselves. Do it on your own. Yeah, and you guys had microstrategy in the past. We've used that. I'm kind of not a big fan of that because it's too hard to use, but it should be easy. That's the goal, right? Yeah, I mean I think, again, going back to experimentation, if you make it too hard, people give up. Or if someone needs to stand over your shoulder and do the clicks for you, they might not, so. So is there a lot of overlap in terms of customer base between Tableau? How much would you say is Amazon versus non-Amazon? I could say we have great success without sharing the numbers. We're hearing from a bunch of different markets, whether it's- More than 50%. Classic, classic. Yeah, so across the number of industries with some great customer stories, we had a session here, I don't know if you guys, I know you're busy here in theCUBE, but from noon to one, we had a great customer telling their story about Tableau on Redshift. So we've been thrilled with the success so far. I think it's the fastest growing product in AWS history. Is that accurate? It is. It is the fastest growing service in our history. We are growing pretty fast in general, so it's been overwhelming. And I think it's really because this is a technology that historically has been reserved for the few. So if you reduce the cost, the investment required to get started and democratize access to a data warehouse, the count of use cases that are applicable goes up dramatically. So we're really proud of the fact that we're providing more access to more customers and more parts of the organization to solve problems on behalf of customers. So I do want to dig into a little bit more about how Tableau is being applied to things like Redshift, but maybe just take a little bit of a step back. The whole concept of big data in the cloud. We at Wikibon done a lot of research around this. And we think that's definitely where the market is going for a number of reasons, just abstracting away a lot of that complexity, the underlying complexity of the systems. But in the shorter term, there's challenges, right? There's regulated industries, there's compliance rules in some industries where you can't have your data in the cloud. How do you see this playing out? I mean, are those kind of concerns? Do you find that a lot with prospects? They're thinking about it, kicking the tires, they'd like to move, they've got some concerns around regulation, security, privacy, how is that impacting your big data business? Yeah, I think when we think about big data, it's really across industries, it's across vertical segments, geographies. And the areas where you see some of the slower adoption is really, and what we've seen is this is really an education thing. It's a communication thing. So for example, earlier this year at our summit in New York, we had Siemens on stage, Siemens being a large conservative company, launching a diagnostics cloud offering. So this is basically taking medical information from a whole bunch of people to try to triage and get to the right treatment faster to improve healthcare and the betterment of all mankind. This is the thing that you tell your folks over dinner so they're happy with you. So Siemens is a conservative company with a product in a space that's highly regulated, launching a big data initiative on AWS. Let's take them for example, how were they able to overcome some of those challenges? Is it just a mindset or do they have to do specific things to make this possible? I think it goes back to that comment around the conversation in the education. So as we engage with Siemens to show them, hey, here's the security operations and controls that we have and how we take it as our first priority all the time, as well as really digging in and building subject matter expertise so that we can speak the language that they're used to hearing. So we have folks in the team that are healthcare centric, they understand the high tech and HIPAA laws and can then translate those requirements through infrastructure architecture diagrams. Hey, here's how you can do it. This is the way you can maintain compliance and maintain security. So it's a, you know, I think over time, those direct engagements and communications, you know, the friction will be reduced as the word gets out. Hey, this is a, the cloud is open for business or big data of all types. So on the flip side of that, who are you seeing, who are your customers that are, you know, rapidly adopting the product? What are the industries that are kind of forward thinking around big data in the cloud? It's a good question. And of course we're seeing these cases all across the board. I think some patterns that we're seeing are really around customer behavior analytics and that comes in many different forms, whether it's in the gaming world, you know, we talk about Supercell, Nintendo. These guys are looking at how users are interacting with games or using Redshift on the back end to do the analytics. Airbnb is using Redshift again when they launch new features to the website and their business is, right, this web app of how do you connect the inventory of property with people who need a place to stay? That's a big data problem, right? So that's really a behavior, customer behavior analytics job and one market's moving really fast. So within gaming, within hospitality, within other sectors, we see that pattern emerge. Although, you know, the use cases are as diverse as you can get. Yeah, there really isn't an industry that's not going to be impacted by this, whether it's, you know, the example I like to use is agriculture. Not necessarily a high tech business, but absolutely, you see companies like John Deere putting sensors on their tractors and crunching data to help farmers to understand when and where to plant certain crops and there's really no vertical that's not going to be impacted. So talk a little bit about the portfolio a little bit. So you've got Redshift's fastest growing service. You've also got Elastic MapReduce, your Hadoop offering. Talk a little bit about kind of those two services, how they relate, interact. You're seeing a lot of uses where, you know, customers maybe using EMR for some of the big data crunching and then moving things into Redshift. What are some of the patterns emerging there? Sure, so, you know, one of our fundamental philosophies at AWS and specifically in the data services business is that there is no one tool to rule them all, right? So we want to give developers and decision makers a variety of tools they can apply appropriately to their requirements and the problem they're trying to solve. So if you start with, we have our own services, things like DynamoDB for a non-relational database. You mentioned Elastic MapReduce and Redshift. There's also a relational database service. There's a lot of relational databases out there. Yeah, they're not going away. They're not going away. So we have a managed offering there and then Kinesis recently launched our streaming service. And then we work for the ecosystem too. You want to work with Elastic MapReduce and run our distribution, that's great. If you want to use folks in the ecosystem like Cloudera, we're happy to have that as well. So we're really trying to give flexibility in order to align the tools with the customer so it's really, really cool. And I've had some great conversations here at the show is to start seeing those pipelines build together. You might start on the front of the non-relational database when you need high throughput and low latency. But over time, the economics of that data store, you wrote that the timeliness of the data is not commensurate with the cost of storing it there. So you might move it over to a place like Redshift to do longer term analytics. So I think the pipeline that's enabled there is a powerful one. I know you can't talk about the numbers because you won't answer the question, so I'm not going to even ask it to, but... Thank you, John. Andrew Sloan Horowitz just put out a great post called Why Amazon Has No Profit in Why It Works, which is a great story, something that we've been tracking. They have three lines of business. This is Amazon, the company, media electronics, general merchandise, and other. The other category is growing significantly, the other being AWS. The other comment in here is about the... Amazon's the master bundler. Bundling stuff has been a key part. Can you comment from a product standpoint and a business standpoint as you go out and talk to all the Clare Cloudera's, the Hortonworks, the MapR's, the Tableau's? You're doing some biz dev. You want to be in good standings with whatever your customers want to use. So clearly you want to bundle all this stuff in. So comment on that other category, Amazon. You guys are obviously doing well, but the bundling specifically. Yeah, so I do believe, so let's take big data. We're at a big data conference. There is a value chain that exists, right? From ingest, to process, to storage, to analytics. And we're trying to look at that value chain in a way in which we start from the customer. What are they trying to solve? And what bits do they need? What parts of the solution are required in order to derive a solution? So packaging matters. And we look to both the technology partner ecosystem, the software providers you mentioned, but also on the consulting partner side. There's a lot of folks who need help to get this stuff implemented. And they may be the ones that are putting the right pieces together in that final package and solution and delivering it to the customer. So bundling is part of, you guys are bundled, basically. I think people want solutions. Yeah, which are bundled, packaged. Which are caught in the cloud. So you can rent a server by the hour. We try to get to the bundled in the cloud, but that's what people want. They want to pay by the drink and have whatever they need to work, right? That's right. Talk about Kinesis and any updates there in terms of success. Obviously Redshift, we love. Big fans of what you guys announced at the last re-invent. Really close the loop on the stack side. The integrated stack has proven to be a great winning formula to you guys. Talk about Kinesis and give us an update on successes, any stats or anything. Yeah, no stats or numbers to share. I'll tell you that Kinesis is very much a developer focused offering today. And we're still, every day, getting new stories of ways people are taking this effectively giant pipe that could stream any number of data sources or any number of end points and doing interesting things with it. So I think you'll hear more about that at re-invent. I can guarantee you. Some really good stories and good customers. You're good. I'm telling you, you got the good message. Mary's trained you well on the media training. I like my job. Exactly. Yeah, we know the culture over there. We love the culture, but let's talk about the Kinesis. I think that's a great example. We're here in the same kind of things that tab below this fast experimentation kind of concept. My take on what you just said was, look at, it's a cool tech, it's getting great reviews. People, we don't really know yet the full magnitude of what this could be. Is that kind of? I think that's true in general of how AWS has gone to market. If you go back to the very early days and the decisions, should we be building real offerings that are packaged in a box versus primitives charged for on the most granular basis we can support? We chose option B because we believe customers will decide how best to put these things together. Whereas if, and by definition, we didn't have a big budget either. But that was the building blocks, right? You start with computing and storage. That's basic. And then you guys are adding more and more building blocks. That slide is getting crowded. Slide is crowded. Okay, so we're looking forward to some updates. What other pre-show data can you share with us for reinvent? What's the drumbeat going into the show? So I think if you, let's take this from a historical perspective. I think you've seen a trend where we've launched a number of services that are more focused on the enterprise. We've done a lot of talking about the enterprise this year. Many folks and ourselves included as well as the community are showing this is the year where enterprise goes mainstream in the cloud. So things like workspaces as an example and some of the integrations around there. Zocalo for collaboration and file sharing. So a lot of enterprise offerings baked into AWS hitting the market this year. You can expect that trend to continue simply because that's where the customers are pulling us. So let's talk a little bit about Tableau. So one of the two areas I was really interested to hear about leading up to this week were announcements around cloud and mobile from Tableau's perspective. My analysis was that Tableau was a little behind the curve when it came to cloud and mobile. So I was really excited to hear about the project elastic announcement yesterday. But how do you grade Tableau on their cloud and mobile approach? Obviously the BI visualization space is getting very crowded with startups that are cloud first or mobile first or mobile only. So they've got competitors coming from come that end of the market as well as they've got the old guard that are trying to adapt. How would you grade Tableau in terms of their cloud approach and their mobile approach? So when I look at the strategy that Tableau has taken it's really about getting the tool in the hands of users listening to what they have to say. And I think there's a group of users that are looking for more cloud first technology, more cloud first functionality. There's a segment of the market that really wants that. They don't want anything to do with the underlying infrastructure, right? And so I think over time, if that group gets louder you'll see more and more from a product perspective from Tableau. But today, taking Tableau online and having a direct live connection to Redshift I think is a big step. And that unlocks a lot of power and value for a number of use cases for customers. On the mobile side, it's funny, we talk, this is a big data show but at the same time, what isn't big data anymore, right? And mobile and connected devices and this is about data generation and then data generation just becomes a burden unless you can do something with it. So analytics becomes a really important story. So for them I think rapid prototyping as well as seeing the value of the visualizations in mobile is important, but I think there's a lot more. There's a lot more there. We'll keep pumping ourselves up and by saying that we coined the term data first because cloud first, mobile first was a big buzzword, then cloud first and even Microsoft adopted that. I think you guys coined cloud first first. But data first is something that David Anvalonte and Jeff and I have been talking about because data is native. So I don't think there is a big data conference anymore. I think this isn't more an application conference. Just think about software in the cloud, if you will. But data is in everything. So it's not, you can't really peg the data market, right? It's like fast data, little data, internet of things. So I think it's pretty clear. Has the internet of things and these new big data, new data areas opening up, propelled Redshift more? I mean, Redshift seems like a great fit for the internet of things kind of things. Obviously Dropcam, who we interviewed at Reinvent, so awesome and then seeing the acquisition was not a camera company, it was a cloud company. I mean, they built, they innovated with Glacier and Storage, some really badass process to create a great pricing structure. They, no one figured that out. Founders said I couldn't get funding. He almost went out of business. But then he ended up crossing over. That innovation was in Amazon. Right. Yeah, I think when you look at internet of things or connected devices, what you need is ubiquitous connectivity, place to store the data and then compute power to do something with it. So it's a natural fit for the cloud. And I think also, because of where we are in the phase of maturity with IoT and connected devices, there's a lot of experimentation happening right now. People are trying to figure out what is the right way to do this? And to do this with purely an operating cost and a low one and a low risk of failure, it's just a natural fit. So I think we'll continue to see. And I think just, you know, just kind of to talk about some of the things that we experienced with you guys just to share for the audience here more data about what we've done is that we've had a similar experience with, I like Drop Can with our CrowdChat app. And I was talking to a venture capitalist to like, it's just a chat application. I can put a database together. I can do that in a second. I go, no, no, you don't understand. It's not a chat engine. It's a DevOps play. Right. And they're like, it's a DevOps play. I thought it was a big data play. No, what is in the cloud because it's horizontally scale. So when they see that level of large scale engineering, they go, oh, I get it. I can see the big picture now. So what things can you share along those lines, similar use cases where, wow, the innovation is not so much what you're looking at. It's in the cloud. Yeah, that's a great question and a long list. I'll take the example from our session here today. So UBM's big event organizer, one of the largest event organizers in the world. And they're responsible for figuring out who goes to the Black Hat Security Conference every year. Okay, it's highly coveted, invite only. So they turned to Redshift. They put all this data in Redshift and they start slicing it to figure out what is our profile of person that we should bring to Black Hat this year? And then how do we actually communicate to those folks and get them to come? So that is a kind of cloud-first strategy, data-centric, using analytics to better understand what is the profile? What is the customer? How do we get the right people in a room? So kind of an interesting story and one that's using both Tableau and Redshift. Okay, to summarize then, we learned that you're here doing Bizdeb with Tableau, great partner, great fit. Obviously, we agree with that. Two, the drumbeat leading up to re-invent is going to be enterprise. And just business as usual for Amazon. Right? We're iterating as fast as we can. Redshift has been on there. You guys always add new stuff. I mean, you guys are always adding new products. I'm expecting to see more new stuff at the show. You can expect to see more stuff at the show. Good, confirm. We're going to see new stuff at the show. Any, anything else you'd like to share? Yeah, I think one thing for folks who are interested in Tableau and trying to understand what this cloud thing is all about, we do have a program we call Test Drive, which is basically a disposable environment with no risk, no cost for people to play around with Tableau connected to Redshift and some sample data. So if you had aws.amazon.com slash Test Drive, there's a Tableau offer in there. And- aws.com slash what? aws.amazon.com slash Test Drive. Be a great opportunity for those who are in the community to try and understand what this means for them to get their hands on the keyboard and try it out. So risk-free for what? Tableau on Amazon? Tableau on Redshift, okay. One question I meant to ask you. Help me understand, help our audience understand the, you've got the, I think the Redshift, we think we understand, but what are you seeing in terms of the usage of EMR and Hadoop in your, among your customers? Are there some use cases emerging? Is it still mostly Test and Dev and POCs? What are you seeing there? Same, it's the same story. Very broad range of use cases. Some long running clusters with very predictable analytic jobs, you know, you're running the same types of jobs every day. Others that are pure experimentation. It's all over the map. We've got some great enterprise customers that are using EMR as part of their core analytic workflow. FINRA is one that we talk about publicly. So, you know, there's some really good stories out there and I think that's the way in which we try to, you know, communicate to the market. There's others doing this. There are ways to find success at the same time. You know, we want to help them find the right tool for the job, so. Okay, Scott, thanks for coming on theCUBE. Really appreciate you stopping by. We'll see you at the show. This is theCUBE live in Seattle, the Tableau User Technology Conference, data 14, hashtag data 14. I'm John Furrier, Jeff Kelly. We'll be right back after this short break.