 Winston Edmonton here, I'm hanging out with Kathy Chang with IMPITUS. Kathy, I want to find out about the services, a full suite of services that you offer. How is that different than some other traditional that maybe just offer one product and no complimentary suite of services? Sure. So IMPITUS is a software solutions and services company. We've been around since 1991 in India and the headquarter was established in San Jose in 1996. We did our first big data project in 2007, so we were one of the earliest big data players in the services space. So when you're saying big data is a little bit different in the traditional IT services business where the customers may already know what they want to implement, what technology they want to choose, they want to implement it, here the big data journey is a little bit more difficult. There are so many technologies available out there, and customers may not necessarily have the expertise to know what they need to solve the problems they want to solve. So what IMPITUS does is we start from the problem that the customers are trying to resolve and we help them select technologies that work the best for their environment and we are objective because we are technology agnostic, we work with a number of partners including Hortonworks, who's a host of this Hadoop Summit here, one of the hosts of Hadoop Summit here, to work out the best solution for their environment and how we're different from the rest is that we're not just doing the consulting piece and say here's that advice and good luck to you, but rather we also carry it through and we also do not only do the architecting, we also implement it to make sure the customer is successful. Gotcha. So there's so much new information out there that you give a little hand holding a long way. You help them manage the entire process rather than, like you said, just handing it off. Yeah, absolutely. So there's, for example, one of the things we do for customers who may not have a good understanding of the overall ecosystem is we do a one to five-day session to help them understand what's out there, what's available and what works best for what situation and it depends on what the customer is looking at. We can look at it, okay, let's look at your environment. How does this work for your situation specifically? We have a one-day, three-day, four-day, five-day depends on what they're looking to do, really to get them started and then hopefully by then we'll win their business to carry that on as well. I like that. Kathy, you said you start with a problem. Let's talk to some folks that may not even realize they have a problem. What are some common problems that you see that you're able to help with? I want to trigger their, I want to inspire them and help them realize that, wow, didn't even think of it as a problem, but maybe we can address it. What have you seen? Sure. I mean a lot of times it comes down to two main areas. How do we make more money? How do we sell, well, from a customer perspective? How do we sell more of what we offer to the marketplace? And second is how do we save money? How do we save costs in our environment? So I'll give you a quick example of one which is mentioned yesterday when our customer news star and impetus had a joint talk at the session yesterday. What they were, one of the things that news stars Mike Peters brought up was that they were looking to really reduce costs for their data center. So just from consolidating the licenses and being able to meet their data needs, they were able to save a significant amount of money to fund additional projects to put some of the newer technologies in places to address their data quality needs, data governance needs, et cetera. Kathy, I love hanging out with you. You just have so much information. I have a favor to ask you. What's that? Is there any way we could get a demo for some of what you do? Can we go check out kind of a demo of what you have? Let me think about it. Sure. We have a demo out there, and let me take you there. Fantastic. Let's go check it out. This is Kathy Chang again, and I have here Vivek Ganesan, who is full architect. He's going to show you a little bit of what we have been showing at the Hadoop Summit here. Thank you, Kathy. Thank you. Here we have our demo booth at the Hadoop Summit here. We actually, this demonstrates our real-time analytics capability that we could engineer on top of Hadoop. If I may just point to the demo showcase here, this is a real-time social sentiment analysis engine. This is built on top of our real-time analytics engine called Vajra, which we are just announcing at the Hadoop Summit. This does high-velocity ingest of various data streams, routing and workflow for analytics and data pipelines, as well as the ability to run complex algorithms in real-time, and also the ability to persist the data into Hadoop for batch analysis. As a showcase, we've actually built a sample demonstration application that looks at tweets, so we are streaming tweets, and in real-time, we are running sentiment analysis on the tweets, and you can see the sentiment being shown here as the streams are scrolling by. We kind of slowed down the tweet stream a little bit so that it's visible, but you can see negative sentiment in red, neutral in yellow, and the topic that we are talking about is gun control, actually, which generates quite a lot of tweets, as you could imagine. So positive would show up as green, that is pro-gun control, and negative as red, and neutral being yellow, and we've mapped this to locations as well, which is projected on the map here. So every time a tweet appears, we automatically in real-time do the sentiment analysis and project that on the map. This is just a demonstration of what this engine is capable of. You can imagine a lot more complex applications where you're running analytics and deep learning on rapid data streams that are coming in. Here is another example in a different vertical. This is for the financial services, and what this is, it's talk analysis, and you can see the ... It does several stuff here, for instance, we are looking at Google. It captures the data coming in and runs analytics on that data in real-time, and we also built a learning model, which generates signals by holder cell signals for a set of stocks that you could be tracking. Again, both of these demo applications use our project, which is a real-time analytics engine called Vajra, and we have a lot more use cases to offer, and we will be releasing white papers and webinars about this pretty soon. Thank you.