 In San Francisco, covering the Spark Summit Silicon Angles, the Cube, we're here with all the action packed house in the moment on the ground, tons of people behind us. Packed House Spark is making its name for itself, its breakout moment, IBM's big announcement yesterday really changed the game, data bricks, tons of startups, it's a really growing ecosystem, really founded out of the growth of the Hadoop ecosystem, funded by the investors of CloudAir, the first pioneer and then the ecosystem just flourished and Hadoop really set the stage, now Spark's taking to the next level, my guess is Ed Dunbell, veteran in the industry, been there from the beginning now with Data Science, Silicon Valley Data Science, Ed, great to see you. It's great to be here, see you soon. So what's your thoughts? I mean I was kind of speculating, we talked about this earlier, Hadoop set the table, first before we get into kind of how this progressed, what's the action like here and what does it mean for the folks that are not here on site in this chaos of growth? I think there's certainly a lot of signal in the energy that we're seeing, there's no doubt that Spark gets so many things right, you know, at an economic level for the developer for productivity and while maybe not everyone here can say exactly what that is, the excitement taken together is a really important signal. So talk about the evolution of Hadoop, we were speculating yesterday on the cube about how Hadoop really incubated the market and that with no unstructured data, we saw that benefit. So the customers all had a sprinkle in the spark in their eye, if you will, pun intended, around the greatness and capabilities with the data. Do you see that as a factor? Is there other factors than driving this growth? Yeah, I think Hadoop definitely set the table from what we're seeing here with Spark at your technical level, you know, enable people's eyes to what's possible. They made the investment in the clusters, you don't need to buy a new cluster to deploy Spark, you just run it on the one that you bought, right? You can start leveraging your existing investment and Hadoop certainly at that level, particularly of HDFS, has set the table for the storage and what Spark gets right on top of that is it doesn't say my way or the highway says, here's all the ecosystem, let's play, let's unify, let's make a great development environment. So we don't have to glue together 10 different solutions, we can start to provide a unified platform and that is so important for skills, for maintainability, for productivity. You guys are showing some stuff in your booth, we did a little quick tour, really kind of a maker fear kind of product, but it really shows the power of Spark. Describe what you guys are demoing because it's very cool, you got a couple, you know, eight cores, is it? Describe what you guys are demoing. Yeah, so we have a fun little cluster, four node Spark cluster, running on low power arm, odroid boards, eight cores on each of them, they've each got two gigs of RAM. And we just like doing some stats on social media from the event. But what it shows is that Spark can go real small, you know, our developers run Spark on their laptops, no problem, right? They don't have to change tools from the small size when they go up to the big size, it's really important. So what's the big deal about Spark? For an enterprise person out there or a telco or a big customer, you okay, what's this buzz about Spark? It's kind of I got to pay attention to it. I mean IBM's announcement, I was talking to them last night, he's going to make every Wall Street analyst look at it, every and everyone who's covering the market and competitors of IBM even have to look at Spark now in a very serious way. It's going to come out of the the existing ecosystem now go mainstream. For those people that are in the mainstream market, what is Spark? Why is it important? I think it's real importance is as a unified platform for doing analytics at distributed level. You don't have to go outside of Spark if you want to do machine learning versus, you know, ETL or so on. And so that makes that makes it the first really integrated data science platform. And the thing it delivers over and above, you know, Hadoop it is it's performing, it's interactive, it's going to satisfy business users, and it's going to be fun and satisfying to work as a developer. What's your outlook for the ecosystem? And with Spark kind of coming in as a new engine of innovation, what's your outlook for the whole sector in the ecosystem? You know, I think we're going to see a time growing up. I'm happy about that. You know, there is a lot of fragmentation or tools here and there always is that's that's innovation. But I'm expecting to see the next couple years of consolidation on platform. And you know, these platforms will look various ways, but there'll be similar in various ways. I expect to be able to write to Spark and run on AWS and Google and IBM or wherever I want to run or on my own data center. And I'm ready for that. We want to get on and focus on the applications of data science, not endlessly worry about doing things together. So more. So with that, the next step is more tools. So plethora of tools on top of a platform, less multiple platforms, right? Yeah, I think we're stuck. We're ready to start see some of those things that used to be mature in the old analytics. Well, things like model management and so on come into come into our world. You know, and things like Docker really changed the game for how deployment works. So we're looking to kind of get more grown up, more ready, more, more productive. The developer story is an interesting one. We'll see how that plays out. You have been a big executive producer of the O'Reilly Stratoconference over the years. And so I got to ask you the DevOps cloud question and how that's going to relate to big data. How do you see that coming together? I'll see, you know, a lot of talk we've been speculating about how, you know, the killer app for DevOps is now analytics with cloud power. So with Spark, it adds more, more fuel to the fire because you got in memory, you got flashless persistence in the infrastructure. So it seems as DevOps is kind of coming together. What's your thoughts on that? And do you think it's going to power more growth, more startups, more innovation? Well, clearly, the ideal for us is to make most of that go away as much as possible. So we're talking about deploying apps and containers and really moving to an era of utility computing. Personally, IOC DevOps is a transitional discipline. It's taking, you know, took us from very manual and it's automating more and more away. We're heading towards utility computing. It's an exciting time, particularly around containers. They're all game changer. Yeah, and how to orchestrate all that. Final word is what's the thoughts out there for the folks watching? What's happening here? What's it like here in the building? Describe the seeing what's happening in the moment right now. This is just such a tangible excitement. You know, I think they could have done anything but sparking it and the whole crowd would turn up. Notice that there's some great content here, but the excitement kind of transcends anything that's going on in one particular place. A lot of people watching and we're bringing that coverage at Dunnville here inside the cube on the ground, live in San Francisco, almost live. We're recording. We'll be live on SiliconANGLE.tv. Again, we're in the moment. Just bootstrapping here, joining as much content. Share that with you. I'm John Furrier. Thanks for watching.