 Okay, we're live back here, you know, towards the end of the day here in Santa Clara, for the Cassandra Summit, I'm John Furrier, the founder of SiliconANGLE.com. This is our theCUBE, our flagship telecast. We go out to the event, talk to the smartest people you can find, and extract a signal from the noise, and we love data, and Cassandra is the big data show. Hadoop, Cassandra, all these stuff's changing the world, no sequels, really driving on the scene, and data is part of it, and our next guest is Eric and Russ, who, Eric's the CEO of Simple Reach. CTO. CTO, I'm sorry, CTO, it's okay. Yep. From New Jersey, so I'm not going to hold it against you. What exit, I got to ask, did I grow up in New Jersey? 18, but I live in New York City now. Yeah, New York City's hot right now, good start-up environment. Guys, welcome to theCUBE, and you're a Chief Architect? Yeah, Principal Architect. Principal Architect, great. So, Eric and Russ, Eric, let's start with you. Honestly, Cassandra, you guys have to play with data, you offer a service, first tell the folks out there, what is Simple Reach product? You guys are a start-up, you're growing really fast. Talk about what your company does, and the product is going to jump into the conversation. Sure, so Simple Reach is a social intelligence tool for content creators. We track all the social actions that happen across the web in real time. So we're talking likes, pins, tweets, digs, shares, you name it, we got our hands on it. And we take people, we take content creators, anything from video, articles, we track it from the instant that it's published, and we assign it a score. This way, content creators can see how well their articles are doing socially. So you guys, what just went to your site, seems like you guys are in beta, or only member-only at this point? Yeah, so we're in probably- It's like a Quantcast meets PageRank, right? Yeah, I mean, we like to call ourselves the PageRank for social. It's a little bit of a big statement, but we know we can do it. So obviously the elusive term is engagement, right? Everyone wants to know engagement, that's the holy grail. We've been covering social for a long time at Silicon Angle. We have a labs group that we've been playing with it. We covered this moment, which is the last independent, kind of social media platform out there, Buddy Media got bought, and it's a hot area, right? Yes, absolutely. Buddy Media Wildfire just got sold to Google. Yep, it's crazy. Radiant Six is gone. They were one of the early pioneers in monitoring, kind of Twitter, if you will, but now this is a paradigm that's not about just social, it's about user behavior. So advertisers want to know what's going on too, right? Like, who should I buy? Page is worth anything? So that or advertising is shifting to more engagement in social networks. You guys track any of that stuff too, or how does that relate to that trend? So we actually started out with AdTech. AdTech is obviously a very interesting space. There's been a lot of major AdTech acquisitions in the social space, as you mentioned, Buddy Media. We happen to know a bunch of those guys, New York City being the mecca for AdTech. There's clearly a lot going on there, but we started out in AdTech and we realized that what people were really interested in was the social aspect of it. People would say, well, this is great, but how engaged are people socially? Are they, is this ad something that they may want to tweet? Is this something that they're going to want to share with their friends? And we found that this question kept coming up over and over and over, and we realized that there was no real way to get at this data. So we started aggregating all the social data that we could find for our publishers, the ones that we were working with, and we found that giving them the ability to mine, it was really what they were after. It's not something that they know how to do and that they're capable of doing. So are you guys funded? BC funded, Angel funded? Yes, we have a seed round. We are very shortly going to be looking for our Series A sometime in the next six months or so. Can you say who the, in the Angel Investors are? So I can tell you that as of right now, we are led by High Peaks and the other major investors, Village Ventures. And that's about as much as I can say for right now. We're talking to a bunch of people, a bunch of people. It's a hot market for startup right now. I mean, big data, just you spell Hadoop or Cassandra, you're in, right? Yeah, actually we're both here to speak. I spoke this morning and Russ is going to be speaking a little bit later today. So we're not only working with Cassandra, but we're really sort of helping push the industry forward. Great, so let's get into some of the meat and potatoes because it's a fun market. Big data is changing the world and getting insights is great, right? So but the hard part is building a system that's specialized yet general purpose because you got diversity of data sets. So one of the things that Jeff, Kelly and I have been talking about is, you know, we believe in is that it's not about the data, it's about the mashups, right? And so you have an ingestion issue. I need to ingest the data. I need to act on the data. And to be real time, you've got to be fast. So you can't just park it out, then run algorithms on it. You've got to do a lot of things and on the ingestion point and then kind of when you store it. So require some semantic analysis. So take us through how you guys look at that and what are you guys doing with Cassandra to address that? Yeah, so I'm going to start out and then I'm going to let Russ talk a little bit here. One of the things that we're actually here to talk about today is that because we have so many different aspects of our system, we've built what we like to call polyglotany. Like we have a bunch of different types of data stores to service different parts of our application. So we use Cassandra for big data, high volume, high velocity ingestion. And when we want to see the deeper analytics, we use a MySQL column store, which is Infobrite. When we want to show our real time data, we use Mongo and we have a front end caching engine and we use Redis for that. And all of that stuff is only really possible because of what this guy did. What this guy did was he built, there was already an existing Node.js driver for Cassandra, but it really wasn't capable of doing everything we needed to do. So he rewrote the Node.js driver for Cassandra, basically single-handedly, and put us in such a position so that we can talk to any data store and provide, or act on any request in a manner that would be fitting for the request type. So I'm going to let them talk a little bit about our Edge servers and our internal. Before you jump in, Russ, that's fantastic, congratulations by the way, Russ. So we did theCUBE at Node Summit. So we love Node, Node has become really functional. And a lot of new ways that people weren't even, that were clever. So we love it. So I want you to expand out specifically what you did with Node and why was Node so instrumental in making that happen? Well, Node.js is extremely great at doing the, I guess kind of doing multiple requests concurrently, and doing this with IO bound sources. Since we're in the Amazon Cloud, everything we do is IO bound. So anything we do like sending requests off to data stores, writing to disk, writing log files, et cetera, is a lot more efficient using Node.js. We do this all on our Edge servers to kind of- In the cloud. In the cloud, yes. Our Edge servers in the cloud that do the actual data ingestion are doing thousands of requests per second on a very few amount of servers. How's your EC2 bill look like? It's actually, it's pretty amazing what we do. Is you're using S3 and EC2 on Node.js? Actually, what I'm using S3 for nearly anything. All the, everything that we do is we have, give or take, a little over 50 instances. About 30 of those are dedicated to database engines. And what's amazing is that what handles all of our data processing happens on about 10 servers in total. All of our real time, all of our batch processing, everything handles on, is handled on about 10 servers. And like I said, a lot of it's- So where are you storing all the data? On Prim? Cassandra on a Femeral Drive. Oh, okay, cool. Yeah, it's because it's, you know, there's a couple terabytes here and there's no big deal. But they continue, so go ahead, continue to tell us more. So all the data ingestion happens on Node.js side servers because it's extremely fast. And the key thing that we like about it is we can respond back down to the client saying that the request is complete before it even gets written to the data source. We don't have to write it to each individual data store which there's probably about 10 to 15 writes that happen for every single page view request we get. In addition to data ingestion and just high volumes concurrency, we also use it as an API layer for our service-oriented architecture. We kind of wrap all of our data stores in this abstraction, which is powered by Node.js so that we can kind of return to all of the various clients. Almost every language can speak in JSON and we use multiple languages for what they're good for. So it kind of makes that a lot easier. And it's very high performance for what it does without having to drop down to something that's going to take a lot of development time like C or Java. So take me through the value purposes. Now I'm a customer. We were at silkenangle.com. We have a website and we have no ads on our site so we're not really in the ad tech business because we essentially were free. We collect data though. We have an HBase implementation. We store all the data of our site and also other third party, our external data of our audience. But one of the things that we say to people who work with us, hey, you know, we're very effective in social networks. How do you guys measure that? So is it because, is there a overhang in terms of the metrics? Because it's not just per tweets, there's gravity like in the social networks, word of mouth has legs, the time is a dimension. So there's a time series aspect. Do you guys look at that? Yeah, so we actually have a data science team that does do all sorts of valuable insights. We actually have a section of our site that we're going to be launching very shortly called Insights. Which is going to be push insights. Because the great thing about analytics is that as long as you know which question to ask, it's a fantastic set of data. Problem is, is a lot of people don't know what questions to ask, especially when it comes to the social media. So that's what one of the values that we provide in our data science team provides. But in terms of what our value prop is, from the moment that an article is published, we will take a look at it across every social network. We're going to track every action and we're going to track the value of that action. So do I have to put in something, an API key in my analytics? So all you have to do is throw some code, throw some JavaScript on your page. And you know the low friction entry point that we're doing for most of our clients is most people use WordPress. So we've written a WordPress plugin and we say just throw this on there and it gives us the title, the URL, the author, and some other basic data. How do we demo it? Or how much does it cost? Content publishers aren't very, they don't like to pay. Yes, we know it's like getting water from a stone. We've experienced this. Actually- We want it for free, right, Mark? Everybody wants it for free. But actually- I know Forbes is doing some work with Word, WordNIC and others, they're trying to figure out how to do better. Because online is not like a book, there's no beginning and end. Online behavior is not like a magazine. So people want insight on how to best serve their audience. Right, so one of the things that we do is when we provide you this page rank for social, we give you the ability to actually catch content as it's trending up. I mean, there's no use in trying to put ad spend behind an article or push it out on Facebook when it's on its way down. We know when it's going up, we know when it's going down. We have this beautiful thing that we call economy of scale. And because we know and have seen so many articles and we've seen what the pads are and we know when the trajectory is going up and trajectory is going down, we can tell you. And we can tell you, don't paddle up river with this. You know, put your spend behind this article or now would be a great time to post this to Facebook. We know you have 1.5 million fans on your Facebook page. If you want to catch this going up, push it now. We can do things like that for you. Yeah, I think it's compelling. I think you don't know it yet, but I think you're in the ad check business. We know we're in the ad check business. Okay, good. We absolutely know we're in the ad check business, but it's just not our first pitch. Yeah, yeah, well, they're not ready for you. Right. It's too early. But that's good. If those such thing is too early or too late, you just have to be early enough so that you don't miss the phone. Yeah, yeah, exactly. All right, get in front of the wave or your driftwood is Pat Gelsinger would say. Yes, yes. Ross, Eric, great. Thanks for coming on theCUBE. I look forward to following up on a follow you guys, plus the Jersey connection, nice to end New York. Love these coasts. Love that mojo. And New York's hot right now. Love the scene in New York. Yes. Got to love how New York's just trumped Boston in terms of the entrepreneurialness. Although Boston's got a little good big data mojo coming back on Boston too. So you got to watch out. So, okay, we'll be right back. Thanks, Andrew. A lot of action, a lot of startups. So these guys are speaking. SimpleReach.com. Go join, check out their service, cutting edge work on the cloud, node.js, this is what it's all about. We'll be right back more live from Silicon Valley right after this.