 Live from San Francisco, it's theCUBE. Covering Red Hat Summit 2018. Brought to you by Red Hat. Okay, welcome back everyone, this is theCUBE's live coverage here in San Francisco, California at Musconi West. I'm John Furrier, co-host of theCUBE with my analyst this week, co-host John Troyer, co-founder of Tech Reckoning. Our next guest is Dave Abrams, executive general manager of data at Insurance Australia Group. Welcome to theCUBE. Thanks for having me, guys. Talking off camera before we came on about the challenges of data, it's cloud scale. You guys have been around for many, many years. You're dealing with a lot of legacy. You've got cloud right in the front step. What's going on? Take a minute to explain what you guys do, your role, and your environment. Absolutely, no, so we're a large insurer in Australia and we've got offices in New Zealand and across Southeast Asia, so we're kind of expanding at it in our reach. But we've been around for a hundred odd years and we've really grown a lot through merger and acquisition over time. And so what that's meant, just as a bit of a byproduct of those kind of merger and acquisition process, is that data's been siloed and fragmented in different brands and different products. And so it's been hard to get, for example, just a holistic view of a customer. What does a customer have, all the products they hold? You know, are they personal customers as well as a business customer? All that sort of stuff doesn't kind of line up. So we've had that big challenge and we've been working over the last couple of years to even just kind of consolidate all that, unify that data into one platform so that we can see across the group from a holistic perspective and build that single view of customer. And that's now helped us sort of understand what our customers are doing and what's important to them and how we can better support them and offer better services. And what are you doing here at Red Hat this week? What's the objective? What are you doing? And speaking, are you talking to the folks? What's the solution with Red Hat? Well, so yeah, we're primarily here as a result of the Innovation Award. So we were nominated and we're successful in our award for that category in our region, which is wonderful and we're really honored with that. So we're here because of that, we're here sharing our customer story with the rest of the Red Hat team and the rest of the open source community around really what it's meant for us to use open source within a big corporate that's kind of traditionally been based on a lot of vendor technology, right? We've been driven predominantly by the big tech vendors that have come in and sort of helped us build big solutions and platforms, which were great and wonderful in the fact that they were there and they lasted like 10 years plus. And that was all good, but now because things are changing so fast, we need to be more adaptable and unfortunately, those platforms become so entrenched into the organization and sort of lock you in that to adjust and to be adaptable, you can't take it out very easily. It doesn't even stack up sometimes from a business case to say, why would we take that technology out? We'll just have to dig deeper and we'll just have to spend more, right? So we're trying to re-reverse engineer some of that. And the role open source for you guys have been part of new systems, recruiting, Calent, everything, what's been a benefit the impact of open source? Absolutely, it's huge and you're right. I think one of the biggest benefits for us that really plays out is in the talent side, right? For our people to say, not only are we transitioning our organization as a whole and the way we operate, but we're really transitioning our people. We're transitioning from kind of the workforce that we had and that got us to where we are today, but we're now setting ourselves up for the workforce of the future. And it is a different skillset. It is a different way of approaching problems. So, bring this new technology to the table and allowing people to experiment, to learn to update their skills and capabilities exactly what we need for our company. So we're pushing that hard. Yeah, that's great, it's not a real cultural shift. Can we maybe transfer over a little bit to the actual tech problem you had, right? So you multiple countries, multiple data warehouses, multiple systems, so what were you looking at and then what was the solution that you kind of figured out and then went next to you? Yeah, so when I first started the role a couple of years back, we had something like 23 different separate individual data warehouses. They were all sort of interconnected and dependent on each other and had copies of each other in each other and it was just, it was a little bit of a mess. So the first challenge was to really sort of rationalize and clean up a lot of that. So that's what we spent a fair bit of time up front doing which was basically re-acquiring the organization's data from a massive amount of core source systems. In the vicinity of, I think we take data from roughly about 150 to 200 core systems and we want to take that data essentially in as close to real time as we possibly can and pump that into a new clean unified data lake, right? Just to make that data all line up. So that was the big challenge in the first instance and then the second instance was really a scale problem, so getting the right technology that would help us scale into, because we've predominantly been using our own data centers and keeping a lot of stuff in that sort of on-prem mode but we really wanted to be able to give ourselves scale to not only be able to take advantage of cloud infrastructure just to give us that extra compute and that extra storage and processing but really also to be able to leverage the commoditization that's happening in cloud, right? Because all cloud companies around the world commoditizing technology like machine learning and artificial intelligence so that it's available to lots of organizations and the way we see it is really that we're not going to be able to compete or out-engineer those companies. So we need to make it accessible and available for our people to be able to use and leverage that innovation on our work as well as do some smart stuff ourselves. So are you using infrastructure as a service open stack or what's your solution, what are you guys doing? Solution is yet to use open stack as our first sort of real step into infrastructure as a service. So that's really helped us set up like I was sharing this morning, set up the capability for us to now scale in a really cost efficient way. And we've ported a lot of our traditional, dedicated applications on infrastructure that was like appliance-based and things like that onto open stack now so that it gives us a lot more portability and we can move that around and put that in the place where we think gets us the best value, so that's really helped. I'm kind of curious, you worked with Red Hat Consulting and how was, I was curious about that process. Did you, was that the result of a kind of a bake-off or were you already Red Hat customers and said, oh hey by the way, can you give us some advice? Yeah, it really came about, I mean we've been working with Red Hat for many years you know and it started back just sort of in the support area of Linux and rail and using that kind of capability and Red Hat's been there for us for quite a long time now and I think we've sort of done some exploratory type exercise with them around open shift and the container world but what really started the stick was just getting their expertise in from an open stack perspective and when you know that was a key platform that we really wanted to dive into and enable and so having them there as our partner and helping us provide that extra consulting knowledge and expertise was what we really needed and helped us deliver on that project and we delivered in amazingly tight time frame so it was a great outcome. It was a fast delivery. Fast delivery, exactly. What about the business impact? A lot of people look at open stack and some of these new technologies and suddenly there's a legacy stuff going on you have you've got all these things everywhere. What was the actual business benefits? Can you highlight like did you get like faster time to market? Was it like a claims issue? I mean what were the key things that you look back and saying, wow we kicked ass and we did these three things? Yeah, yeah, yeah. I mean really what it boils down to is faster time to market, right? And just the ability to move quicker. So to give you an example, the way we used to work is that it would take you say probably weeks maybe even longer to provision and get infrastructure stood up and ready to go for different projects. So I meant that there was all this lead time that projects needed to go through even before they could start to write code and even start to add value to customers. So we wanted to sort of take that away and that was a big hindrance to be able to experiment and to be able to play with things. So again we wanted to take that out of the picture and really free people up to sort of say well the infrastructure is done and it spins up in a matter of seconds now on OpenStack and you can get on with the job of trying something out, experimenting and actually delivering and writing code that will produce an outcome for our customers. You guys launched new applications. What was specific outcome that came from standing up, putting that OpenStack together. Obviously you're experimenting. So what was the result app? Yeah, not only in the app space but more so the biggest sort of benefit we've got is really in the data space where we've now been able to essentially stand up our entire data stack using open source technology and we've never been able to do that before. And this is the environment that's allowed us to do that by just allowing for us to do that, test and trial and say is Kafka going to be the right tool for us? Are we going to use Postgres? Whatever that is, it's allowed us to sort of really do that in a rapid way and then figure that thing out and start to move forward, so. You get a lot of scar tissue, you guys have done a lot of work out there. Good work, so I got to answer the question. With Kubernetes containers now part of the discussion as a real viable way to handle legacy but also new software development projects. How do you look at that? What's your reaction to that as a practitioner? Are you guys excited? Yeah, yeah. Things in motion, what's your color on that? Absolutely, it's in fact it's been something that we've kind of had on the radar for quite a while because we've been working with containers, so dock in particular, and one of the things that you come across is just management of containers and just the ongoing maintenance of those kind of things where they start to get a little bit un-wielding, a little bit out of control. So, we started off trying to build our own solution to that as a lot of corporates do and quickly found out that's a huge engineering challenge. So, things like Kubernetes that have now come along and the investment that's been put into that platform will really open up that avenue for us and even seeing just the new innovation that's been put into our open shift here that sort of takes a lot of that management and service administration out of the equation for you is wonderful for a company like us because at the end of the day we're an insurance company, right? We're not a technology engineering company while although we have some capability, it's never going to be our strength, right? We're really here to service our customers and to help them in the times when they need our help. But you guys are a data company, data's critical for you, I mean data-driven. Yeah. How will you become more data-driven as a result of all this? Yeah, so now that we've got our data all in one place and we're able to get those single views of customers, we're able to put that data now into the hands of people that can really add value to it. So, for example, into our analytics teams and get them to look for optimization in price or in service, claims processing, all those kind of good things that are helping our customers reduce the time frames that they would normally go through in that part of that experience. And I think one of the other things is not only that, but also enrich our digital capability, right? And enrich that digital channel. So make it more convenient for customers. You know, it used to be that customers would come along and it's literally like coming to the organization for the first time, every time, you know? I want to say fill in that form again from blank. You're like, we don't know anything about you, but now we're able to enrich that form. Exactly, it's very painful to ask you your name and you want to ensure your house, tell us all about that house, you know? What is it made of? You know, what type of roof material? What's the wall? We know all that. We've probably seen that house 10 times already. So why wouldn't we just be able to pre-populate that kind of information and make it more convenient for customers? Personalization becomes critical. Absolutely, absolutely. That's great. I like the way you underscored and told the story. Just like with cloud, you just can't take your broken old IT apps and just throw them up in the cloud. You had to do a data exercise and you had to do a consolidation and a cleaning. And sure, that involved open source, but you didn't get the tech stack first. First, you have to fix your data app. Absolutely. And that was a key part here. Yeah, absolutely. That took a lot of effort. And that's one of the things that I think we really invested in it was, because a lot of the time what we've seen is organizations have sort of attacked the low hanging fruit, like the kind of the external, the digital data that they might be able to get, not that offline data that's been, one and generated by the branch and the call centers and all those kind of areas. And we dug in deep and invested in that space and got that right first, which really helped us a lot to accelerate. And now, I think we're in a better position and we can definitely take advantage of that. Yeah, thanks for chairing your insights here in theCUBE. I got to ask you a final question. For the folks watching that are looking at you and say, wow, this guy, he got down and dirty, fixed some things, he's going forward, innovative. What advice would you give someone watching as another practitioner? What have you learned? What's the learnings that have been magnified out of this process for you and your view going forward? Yeah, yeah, there's a lot of learnings we can share, but I think some of the key ones is, I think there's sometimes a bit of a sort of attempt to try and solve everything yourself, right? And we definitely did that, right? Try and build it all yourself and do everything right, but it's a challenge and use partners and look for things that are kind of going to help you accelerate and give you some of the foundational work that you don't have to build yourself, right? You don't have to build everything yourself. And I think that acknowledgement is really key. So that was one of the big things for us. The other thing is, just investing early and getting things right up front, like pulling your data and consolidating it into a single platform, even though that takes a lot of time and it's quite challenging to sort of go back and redo things. That's actually a huge investment, a big win to really help you accelerate at the end. That investment up front does pay off. So, congratulations on your innovation award. Thank you. On stage. Dave Averson, General Manager at IAG Insurance, Australia Group here inside theCUBE. Sharing the best practices, it's a world, you've got to do the homework up front. Open Source is the way, it's an operating model for innovation, theCUBE bringing you all the action here. On day two of coverage, stay with us for more live coverage after this short break.