 From New York, it's theCUBE. Covering Big Data New York City 2016. Brought to you by headline sponsors, Cisco, IBM, NVIDIA, and our ecosystem sponsors. Now, here are your hosts, Dave Vellante and Jeff Frick. We're back, this is theCUBE, this is Big Data NYC. We are the worldwide leader in live tech coverage. Intamar and Corian is here as the CMO of Attunity. Welcome to theCUBE, good to see you. Thank you very much, good afternoon. It's great to be here. So, Attunity has been part of what we call Big Data Week since the beginning. We've been doing this, gosh, seven years now. Seven years. And I think Attunity's been there most of the years. We've been tracking you guys. So, give us the update, what's new with you guys? Well, this has been a very exciting time for us, as you mentioned. They've passed few years growing and innovating with the Big Data market as it's been evolving and we're trying to stay ahead and with the customers, with the adoption, as new technologies come to market, as new type of implementations and use cases, customers trying to adopt new platforms. So, it's been very exciting for us. I think especially the last year, year and a half, as we've seen more customers start to move to production and start to implement more large scale type of deployments, more real time deployments. So, it's been an exciting time. So, what do you see as the mega trends that are driving that growth? You mentioned real time. Obviously, big data slash a dupe getting into the enterprise. Remember, the trend was make it enterprise ready, make it hard, and we're kind of through that, right? What are the big kale wins? That's a great question. And we're seeing a few of them today. I think if you go back down to the show floor, you'll see that there's a lot of conversation about real time, new technologies like Spark Streaming, Kafka, in-memory processing. So, a lot of options in which to use real time to innovate and create new types of analytic processing, that requires feeding data in real time, which again, for us, fit very well into what we've been focusing on, which is to enable customers to capture data in real time as it changes across all the transactional and database systems, which for many customers, is most important, most common data sources for the business operations. You're being able to turn those databases into live feeds that stream directly into your analytic process is something we've been focused on. So again, the real-time thing is a big trend. You see it in new technologies coming to market. You see it in just the hype and the discussions, both of customer interest is all about. And again, from our perspective, we have focused on integrating our products, optimizing them for delivering data in real time into a diverse variety of options, depending on the type of analytic platforms where customers may want to process data in real time. So previously this year, we've added support for example for Kafka. So being able to take data that we capture as databases change in real time and being able to feed that into Kafka in a format that is appropriate, is aligned with the type of processing that customers want to do. So that is definitely one of those trends, the real-time trend. Where are you seeing the most success? Is that you've seen in particular industries or use cases or where is the traction? So we're seeing an activity across multiple different type of industries. We're seeing a lot in financial services, banking, insurance, and these are industries that have a lot of data built into the business model. Retail, manufacturing, so healthcare. So we're definitely seeing the activity across a lot of different industries. And what's interesting, we're seeing two types of projects that drive it. One is a very specific analytic application. So our customers are very specific analytic problem they're trying to solve. They know that very well. It's really stemming from a very specific business question. And they're building an infrastructure to support that specific application. On the other hand, we're seeing customers that are building infrastructure. They recognize that data is gonna change their business. They want to become more data-oriented companies. And they're building data lakes as a foundation. So they may not have a specific application or a specific use case yet in mind. They have a few that they're thinking about, but they wanna make sure they're building the right infrastructure to accommodate it and enable different data scientists to build applications. So we're seeing these two different types of scenarios. It's not limited to one of the others. What we have seen is that on the data lake side, to your question there, we're seeing that they're starting to move more in production and they're trying to grow in scale. So if I mentioned earlier real-time being one of the key trends, I think the other thing we're seeing is scale. So if when we talk about real-time technologies like CDC changed data capture, which is part of our core technology to capture changes and make them available in a very efficient real-time manner, when we look at scale, what we're seeing is that customers are starting to scale as they move to production from having a smaller initial set of data feeds that they're ingesting into the lake to having a much larger number of data feeds they're ingesting. So now we're talking about many dozens going to hundreds with plans to go to thousands of data feeds from databases all over the world going into the data lake, where they're merging that data with sometimes IoT or other data, social data that together can create new types of analytics. However, as they're scaling it, what we're seeing is customers are running into interesting challenges they need to address. One, being the cost of development. So if every feed requires a development effort and there's a lot of them, that becomes challenging. The formula for time to market for cost becomes challenging. So what we've done in the Trinity is we focused on making the process of creating a feed of putting one in place easy. So it's configurable, it's a configuration environment with an application that's designed to handle data ingest in an optimized manner rather than a development custom scripting type of activity. That takes away the issue of the development being a bottleneck for scale. The other is around agents because as you scale, what happens is that you want to support a lot of databases. Some of the traditional technologies use agents. They deploy agents on each database source or close to it to be able to identify changes, harvest them and enable you to bring those into your lake. However, as you're scaling to support hundreds and maybe more, starting to install, maintain, configure agents across a wide network of environments that doesn't scale well. So again, that's where we focus on creating agentless technology. We refer to it as zero footprint environment and that again allows to scale easily without a big impact on the IT environment. You said there's just so many vectors of growth. You've got how many data sources, you've got how much data is coming down that sources. I would imagine it's how much you open up the pipe and allow the percentage of data from that source. All those things are going up really, really rapidly and then there's this pesky thing that I like to talk about maintenance and keeping up with these things. So how are people addressing that challenge? Do they just open up one pipe? Do they do a little bit and then open and move to more sources? How do you manage this increasing complexity and scale while you're still keeping the airplane in the air? That's a great question, especially maybe with my ELFOS background. Oh, there you go. Keeping the airplane in the air. But no doubt that, first of all, addressing the diversity is very important and what we focused on was creating an heterogeneous platform that enables to decouple between sources and targets so you can scale the number of sources and the Atunity Platform, Atunity Replicate, it is again an enterprise platform, a unified platform for data replication and ingest. It costs a wide variety of systems. So it spans databases, data warehouses, of course Hadoop, Kafka, on-premise in the cloud. So it gives you a platform that enables you to get data into a lot of different places. And many of our customers start with one project they may want to bring the data into the lake or maybe they started by bringing it into a data warehouse but then they find that they also want to bring it into Kafka. So it goes to Spark or it goes to other environments like Cassandra or they want to bring it into the cloud. So yes, it starts to evolve. And we're actually seeing more customers refer to this as a unified universal type of infrastructure. So they recognize they need to put in place type of systems that will give them the agility, give them the flexibility to get data to all these different places. So both supporting a lot of sources, a lot of targets and having the optimizations that are required for each one of them. That's what we're focusing on in the Trinity Replicant. And you've made announcements recently to support these areas or can you take us through that? Yes, we have. Last week we had big announcement again, expanding the footprints of the endpoints of the systems that we're able to ingest and bring data from. And last week we had an announcement around supporting SAP. So SAP is one of the largest, the largest application market with tens of thousands of customers. And of course some of the world's largest companies running SAP. So their core business data is in SAP. The challenge is that getting data from SAP and merging it into other data in the data lake, for example, is challenging because the inherent structure, it's a big application, it's complex. Knowing what data you may need to be out of SAP, being able to do that in real time is challenging. So we have very unique IP related to SAP integration. We've merged it into a Trinity Replicate. And now we can completely facilitate the process of getting SAP data out of SAP in real time very efficiently with integrity, unpacking complex structures of SAP so that once the data lands in Hadoop or may develop it through Kafka or may develop it through Spark or in a database, it's now laid out in a format that is accessible and easily available for developers and data scientists and other users. So that was SAP. And yesterday we made another announcement related to the growth of data lakes. So what we've seen is that as data lakes grow and as more feeds come to it, the issue of operationalizing that environment and managing that at scale becomes an issue. So it's not just the ease, it's not just the efficiency, but we now have customers that are growing to hundreds of feeds. If you have hundreds of feeds across multiple servers that get the job done, how do you manage it? How do you get control over that entire environment? Because if you're going to run it in production, if the operation allows it, you need to have control. So now we have a whole system that supports and extends the Atunity Replicate, it's called the Atunity Enterprise Manager. And that system basically collects information, it can find all the replicate servers, automatically get all the information from them, provide you an easy way to monitor, group, sort, drill down into a wide variety of hundreds of tasks across servers and be able to control them as well as get notifications. So we think that innovation, we're now enabling customers to operationalize and manage the data ingress at scale as they're growing and their data lakes in production. So part of the problem we've heard with the data lakes and everybody, you know, they joke about data swamps, et cetera, but you got this no schema, you put everything in the data lake and then it's like, okay, now what? You just sort of kick the can down the road in terms of actually applying some kind of structure to the data lake. Does Enterprise Manager in a part anyway solve that problem? Is that specifically what you're going after or is it somewhat different? That's a good question, thanks, Dave. So Enterprise Manager, as we've defined today, is really focused on the management and the operations, not on metadata management, where basically the way we see it, we want to integrate with metadata environments, you know, whether it's the Cloudera or whether it's the Apache and Hortonworks, metadata environments that will help companies understand the lineage and all the information around the cataloging or the information that they have within the lake. So these are all foundations of building a data lake from data ingest, which is really what we're focused on, to cataloging and understanding it, to doing the actual data preparation, data analytics, and we try to feed all of that in a way that accommodates all this stack within the data lake. You want to keep the data lake looking like Lake Tahoe, not like the swamp that it can be with all kind of various inputs and things, but really the opportunity, as you say, to more kind of non-traditional data sources to feed what have been kind of a big data process, like the SAP example, really again, kind of a next order of impact in terms of taking advantage of the big data technology. Absolutely, absolutely, because the challenge is how quickly you can get the data and make it available for analytics, how can you make it available in different formats because there's so many different types of analytics customers want to use. And the one other thing I'll mention, because you were talking about different trends, is where people are doing the analytics. And one of the interesting other trends that we have been seeing is beyond the real time, and we talked about the scale is cloud. So where is it that they are deploying? So we're seeing, definitely seeing more demand. We're seeing a trend toward deploying Hadoop and other technologies in the cloud. So then when you think about it, the cloud is kind of a very typical environment you'd expect to run analytics in. It aligns well with analytics because of its natural elastic capabilities. So we see a lot of customers that start in the cloud as a way to run development QA. So they're developing, they're experimenting in the cloud because that's easy. But the production Hadoop may still be on-prem. And then we see more and more that also look to move their data lake or their Hadoop environment to the cloud, running either just deploying their Hadoop cluster in the cloud, whether it's on Amazon or Azure, for example, or using services. So Hadoop is a service like EMR, HDI, Azure and the likes, Azure data lake. So that's what we're seeing. And with the Trinity, we put optimizations that facilitate the movement of data also to those environments in the cloud. So whether you want to do your analytics on-prem, whether you want to do it in the cloud, again, Trinity Replicate provides you a foundation with all the different optimizations and just depending on the configuration you make, you can get the data to where you want it to run. Yeah, the cloud has definitely taken hold. We knew it was coming, right? Everybody sort of saw it coming, but now it's like, wow, it's here. So you got to react to that. I'll give you the last word, we're out of time. But what do you want to leave our audience with in terms of big data week, strata Hadoop, future of Atunity? I think this is really exciting times. I think that we're seeing tremendous growth and opportunities to analyze data, bring new types of data. We're especially excited about customers bring together their core transactional business data with IoT data or other types of data. So what I want to leave customers with is that we're going to continue to innovate around enabling analytics and around enabling efficiencies for managing and integrating big data for analytics. We keep focused on enabling data ingest and doing that in a much easier way, broader portfolio of sources, targets and environments like the cloud. We work very closely with Amazon, with Microsoft, with Google, we're going to keep expanding that. We're going to continue to focus on helping companies optimize their data world, so broader data world environments between their data world of choice over the years and their Hadoop environment. We have software that helps to analyze all of that so they can understand where to run what. So again, we're going to continue to help companies get data and manage it better so they get more value out of it. All right, thank you very much. Thanks for coming to theCUBE. It's good to see you. Pleasure, thank you for the time. You're welcome. All right, keep it right there, everybody. We'll be back with our next guest. This is theCUBE. We're live from Big Data NYC. Right back.