 Live from Las Vegas, Nevada. It's the Cube at IBM Edge 2014. Brought to you by IBM. Now here are your hosts, John Furrier and Dave Vellante. Okay, welcome back everyone. Live in Las Vegas for IBM Edge. This is the Cube, our flagship program. We go out to the events, we track the signal from the noise. I'm John Furrier, founder of SiliconANGLE, joined by my co-host, Dave Vellante, co-founder of Wikibon.org. Our next guest is in the trenches making it happen. We love talking to customers, Brendan Macaulay, executive director IT, service operation, heartland payment systems. Welcome to the Cube. Thank you very much. So the thing I want to get to right away is we love talking to the customers out there, share with the folks out there just to kick this off. What's going on in IT? I mean like operations are critical. You know, Dave and I always joke about 70% of the money spent to keep the lights on and the rest is for innovation. That's kind of flipping around the other way with the cloud, with DevOps, software-defined data center, with all the refresh going on, certainly with virtualization. There's a reset going on. So talk about in a big picture, what is some of that reset? Yeah, you know it's very interesting. We have in the last five years seen IT budgets be fairly flat, but at the same time explosive growth and really the challenge has been taking infrastructure that was very siloed, very purpose-built for a specific product group or a specific business unit and really get building in some more flexibility into a true multi-tenancy environment with a real I.O. blender and it's become a lot more unpredictable. So you used to have these waterfall development cycles. Now you have Adelskrum and DevOps and you know really iterative approaches, failing faster, all those kinds of cliches if you will. The lean startup, the lean IT department. Absolutely. So you've got really got this kind of, these competing interests, you know, you've got this old, you know, this old stodgy way of doing IT and this new startup kind of thing and, you know, you're competing with lights of Amazon for I've got to ask you two questions. The mindset issue is one. I want to get the cultural question, but a lot of the young talent coming in, they don't want to load Linux patches. They don't want to rack and stack gear. Absolutely. So you have a cultural shift with the new generation coming on board this modern era and also you have a cultural issue on the legacy IT department. Talk about that. What is the collision course look like? What are some of the things that you've seen that work and don't work in that environment? How do people break through that? How do they transform to get there? You know, I think a lot of companies that have been around for a while like Heartland face that same issue. So basically what we've seen and really the wins we've had for merging older applications and, you know, maybe some past thought process with some, some, those newer challenges where people don't want to do those rote manual things, you know, on a day-to-day basis. Automations really helped a lot with that. So building in a whole lot of orchestration and automation and those are self-funding activities at many times. And really what we've, what we found is immersing newer talents with, with employees that have been in IT for a long time and building those wins around automation. It gets everybody out of doing the mundane. So talking about the, the 70% keeping lights on and 30% for innovation, you know, automation I think has really been key for us to basically say, you know, we're always growing. We've, we've our server infrastructure for it, for example, is grown by a factor of 20 in the last five years. Our staff hasn't grown by a factor of 20 in the last five years. So there's more and more stuff to do and once you get everybody in the mindset of, you know, let's automate the mundane so we have more time to be innovative and do the more exciting, exciting tasks and add new business value to Heartland, you know, it sends the message that it keeps IT relevant. It keeps IT part of the solution and it really, it really also helps make, make administrators lives easier. You know, they're not up at all hours of the night patching, like you mentioned, or early in the morning and it also improves quality too. So, you know, you start to eliminate a lot of the human error and you get, instead of administrators, you get engineers, you get automators. Are you coming at 70, 30 needle? Oh, absolutely. So we are for sure, we're, you know, we're not there yet. So we're not, we're not at 70, 30 innovation by any means, but we're getting closer and closer to 50, 50, which I think is a heck of a lot of progress. You know, when I, when I was at Heartland, when I started at Heartland about five years ago, really the number was about 80, 20, you know, 80 percent keeping lights on 20 percent innovation and we've really moved the needle quite a bit in that area. Talk a little bit more about Heartland. Maybe talk about your business and what's driving things. Yeah, so Heartland payment systems is the fifth largest credit card processor in the United States and that's always been the leading statement for Heartland. I mean, that is our legacy bread and butter application. So five years ago, that was about 90 percent of our revenue and today it's about 60 percent of our revenue. So that's really interesting. We are, we are big believers in penetrating new markets and acquiring companies that are creative to our operating margins or have complementary product offerings and the complementary markets. We are in the, like I said, we're in the credit card authorization space but we do debit card processing, gift and loyalty payroll, school solutions like the ID card where you can load money onto it, student loan processing and those, those applications make up another 40 percent of our revenue and the real challenge there has been, it's been eye-opening for IT because when you build IT for a certain kind of product line and then you start getting into a diversified portfolio, it introduces a lot of new challenges about mixed workload and unpredictability and then you've got things like DevOps like I said, like I talked earlier where you're learning to fail faster and you're doing, taking this iterative approach to not only development but also like application and product offerings. So if you experiment, that really puts new challenges in your infrastructure. So you've got acquisitions going on and you've got new products that you're supporting. Have you been able to, I wonder if you could talk about your infrastructure a little bit, you know with Tom Rosamilia, infrastructure matters especially when it breaks. I have to talk a little bit about, have you been able to, let's talk about historically IT and not even historically today, it's mostly in silos, you came out of I think the storage world right and then now you're managing the entire infrastructure operations right. So you have your storage silos, your compute silos, your networking silos and generally applications of purpose, you know, have infrastructure, this purpose built to support them. Have you been able to build a more sort of horizontal platform, infrastructure platform to support these emerging applications, to support acquisitions and the like? I wonder if you could talk about that a little bit. Yeah, absolutely. So it's funny you say platform infrastructure, that's actually the name of our infrastructure team now. It used to be Windows administration and AIX administration and it is as siloed as you could imagine. So the storage administration, VMware administration, network administration, firewall administration, all those teams are gone and now we have a platform infrastructure team and we have infrastructure engineers that are much more integrators now instead of subject matter experts and rocket scientists on a certain kind of silo, that siloed view really kind of breaks down when you start to get to rapid infrastructure delivery and rapid application delivery and really our infrastructure team, I'm really proud of them, they've really changed their mindset. They understand that the name of the game isn't knowing some kind of piece of technology as well as possible, it's really it's really delivering value to the business and bringing transparency, good monitoring, rapid infrastructure delivery to the market, that's really what's important, that's what our customers are asking for. So how did you achieve that? Maybe you could talk about that and I mean you don't just wake up one day and say okay we're going to change the whole organization and we're going to change the infrastructure, maybe talk us through the journey that you took there. Yeah the journey there was really started with virtualization back in 2009, I mean we really invested in VMware heavily, that really helped us on the the x86 side you know Windows and Linux infrastructure delivery so we got some wins there on hey we can be a little bit more responsive. And it contributed to your server growth I presume. It sure did, it sure did and it created some new challenges, that actually created a massive storage challenge and massive network challenges and so starting in about I think it was about 2011 we booted out our incumbent purpose-built storage provider, I won't name names but we and we changed to IBM XIV platform and really the what allured me there was the approachability of administration so basically that technology is all about delivering value to the business quickly so when you're doing things like you know provisioning out loans to VMware or physical systems that you're not it's not a road exercise in just babysitting technology you're really getting what you need to get done accomplish fast and really letting the technology you know take take a process that takes minutes or you know hours down to just minutes or seconds and also really providing a lot of transparency on monitoring so the XIV is all about simplicity so I look at it as the consumerization of IT and infrastructure so it's basically I like to refer to an XIV as kind of like an iPod of sands you know it's you know what you're getting there are no hotspots there are no surprises and they're it's really easy to to manage and that so we started with with the windows and red hat virtualization and then we went to storage and said now we've got server server delivery fast with virtualization we've got storage delivery fast with the XIV and that started giving us some clouds in the company say you know what we are we are getting more nimble and that really progressed down to the network environment our firewall environment and it's still we're still working on it but all phases of our infrastructure delivery now are are accelerating and are getting much simpler and also more predictable from a user experience perspective friend I got to ask you the DevOps because I'm fascinated by the innovation going from 80 percent keeping the lights on 20 percent innovation at 50 50 is a huge task obviously the comments on Twitter already coming in on that fantastic stuff DevOps really is the culture you guys you mentioned that earlier but I got to ask you about automation orchestration these are operational things that create reduce things mundane tasks allow a free up to work on some creative stuff so I want to ask you about automation and orchestration at the ops level and then the pressure you're getting if any or the drivers from the app developers because at the end of the day people just don't wake up and say hey let's do some DevOps there's some other forces at work there is it the applications and how is that working yeah so DevOps is it's interesting yeah more and more what we're seeing is developers going back to the consumerization of it you're they're expecting the same kind of experience that they get from amazon but also with best of breed infrastructure that they could get from our infrastructure teams at heartland payment systems and the best way to really make an application work well from an operations standpoint is to have the infrastructure and operations teams there at design time for for an application that's where DevOps comes in right so in the past operations and infrastructure is just an afterthought right so give me some servers operations guys and then go away and then when it's time for this thing to go to market you know come in and you guys can monitor it and have a nice day and what really you really see there is that applications fall on their face when that when that's the model right so you're not setting yourself up for success and DevOps it's really taking those runtime considerations of monitoring measuring health scalability recoverability so you know hadr self-repairing self-repairing systems and applications you know it's really taking those considerations into account at design time and uh you know we're we're somewhat neophytes in that still you know it's it's working it's still really though absolutely yeah um but we've got everybody drinking the Kool-Aid we've got some operational uh so we've got a set of 10 commandments from operational values at heartland and those are development guidelines so they're not just operation they're not for the operations teams those are for developers as well so i mean you described that that applications fall down when you do that and it's my understanding and talking to organizations who have made the move or beginning to make the move is that because when you go to deploy it it doesn't work the way it's supposed to or it's not backed up or it doesn't fit the edicts of the dr and ha strategy so the operations guys start hacking it and then they break it you won't maybe admit that then they send it back to the apps guys say your application doesn't work they say that's because you messed it up and get this finger pointing going on so how did you resolve that a lot of guys come at it from john calls it ops dev um but it sounds like you guys are more coming at it from from dev ops it's an application centric skill set that you're driving is that right and how did that affect the operations guys would they was it a training regiment that you had to put them through or yeah i mean i would i would really you know it's interesting i i would call it ops dev i mean in the past that's that's the way we've approached it but but it's it's again building that kind of cloud like we've done with you know the servers and then the storage and then the network to you know the rest of the infrastructure building that cloud of we can build a best of breed scalable infrastructure um my that my uh my career has gone from infrastructure now into operations and i manage the infrastructure teams as well um but it's basically one voice on the ops side of dev ops uh you know one one team in operations including infrastructure helping build helping build best of breed applications with the development teams uh i don't know if i answered your question yeah absolutely do you feel like you're running for mayor when you have to organize and herd the cats and dev ops or do you get to a point where it's kind of a you know it's a cultural thing where there's some norms that are formed take us through some of that and what what did you learn what bumps have you kind of overcome and absolutely resolve that you know i i felt like i was running for mayor for about the first nine months or so but uh it's great uh there's a lot of good leaders behind me and uh in the development environments and in operations and infrastructure it's not a hard sell basically yeah they're having meetings without me you know it's it's you know i'm not you know and and they're really uh the greatest thing is some of our applications that are even the most legacy applications we have we have developers we have leaders of the development teams getting with our operations guys and getting with our infrastructure guys and really figuring out you know let's throw all of our assumptions around the the architecture of this application out the window and how do we how do we repurpose this application with dev ops in mind and how do we you know these operations and development teams used to be really really separated and now they're talking more than ever and they're really they're really saying you know what how do we fix it how do we kind of accept the sins of the past and fix them and and really take this application that's had that's had issues and had had problems growing and really scale it for the future are you using any flash switch to subjects but while i have you yeah so uh we are PO seeing so a couple of great leaders on our team uh heartland and uh on the infrastructure and operations side we've we've we've done a bake off of several flash systems uh all flash over the last six months and uh we are we are doing a try and buy on the flash system a 40 right now and we love it i mean it's i'm not gonna show for IBM but it's it's it's the best thing we've had database app yeah so it's a multi-purpose VMware data stores and database driven apps absolutely okay how about big data is that infiltrated the ops yet in terms of uh is it a dub is it purpose built is it off the shelf yeah right now it's purpose built it's an appliance uh so we we use pure data yeah so that so uh we have a former to tease a pure data system and uh it's it's gone really well we have a sequel based uh data warehouses that aren't going away just because it you know the the transactional yeah exactly transactional and and the migration strategy for the for those are pretty pretty low on the on the low latency more big data cost per gigabyte more throughput do you use a different system for that um not really so uh so flash comes to you know so what we're looking to do actually is uh really virtualize our sand environment so really at the end of the day big data you're bound by i o right you're bound by latency and uh what we're looking to do is virtualize our sand environment so we we've standardized on xiv for a long time so we used to we used to run it we used to be uh a not non-idm storage environment and we moved to ibm xiv and it was kind of a cookie cutter xiv environment so where we had just multiple xivs and what we found is that there are workloads like big data and um just high high uh high high i o low latency environments like a swim very specialized oltp and and definitely some big analysis environments that needed something faster than xiv and that's where like things like the flash system a40 and svc's and multi-tier can come to mind so what we want to do is we don't want to just uh we don't want to just buy an 840 and put everything on it um we do want to virtualize our data center as much as possible so our strategy was to just expose all of our xivs and flash systems as data stores in vmware and let vmware do that but actually there are there's software out there such as uh the virtual storage center from ibm in the svc that will do that a lot more uh a lot better than we can do as humans right so let the let the computers do that and and really drive up the efficiency and the usage of our of our flash system a40 and deliver better value for us so that sounds like a step towards software defined that automation but you know you hear of obviously everybody's marketing and pushing software defined the challenge is you got a lot of stuff that works today and it's rich data services you talked about xiv you talked about svc um the data services that exist there today are probably much more mature than you're going to get with some software running a bunch of commodity disks so what's your take as a practitioner um around software defined and then we get a wrap yeah absolutely so i think um software defined should be around automation and orchestration and the um basically adding value added services on top of really high performance hardware so i don't think you can just say we're going to take a bunch of commodity disks and and make a software defined uh environment at least i don't think that's it's there yet so um i look and i and i i think that way around servers and i think that'll way around servers not so not as much but definitely around networking like sdn uh you you need a high performance infrastructure and lake intelligent software on top of it in my opinion because nothing's going to beat the speed of basics right so um and in flash right i mean you you you need purpose built flash i mean uh flash is not flash right there are a lot of different flash vendors in the market and some of them are more like around mlc i mean not to get too techy here right you know you got your your commodity ssds and you got some mlc and then you have things that are faster and there's a huge difference in performance and i think uh when you when when you're looking at a multi-purpose multi-tenant environment like you're you know private cloud you know let's use the cliche um if you just if you just make something a cloud you put or automation orchestration on top of a poor infrastructure your cloud washing i like to call it delivering poor infrastructure faster and that's not really the business that we're in and we want to deliver best of breed infrastructure and say you know we can deliver infrastructure nearly as fast as amazon or as fast as amazon but you're not getting 50 millisecond latency you're getting sub millisecond latency you're getting you're getting the infrastructure that you've come to expect you know working with it working inside a heartland payment systems infrastructure and i could check the box on my edix on compliance and security absolutely those things are very very important to us brandy thanks for coming on cube i really want to give you the last word here for the folks out there's been a great performance segment share with the folks out there who are on a journey to try to get 50 50 actually get more innovation what roadmap can you share with them what direction what best practice can you share the folks out there on how to get there yeah i think what you need to do is is when you're optimizing delivery of infrastructure or optimizing operations find the find the largest place that that there's room for improvement and those fun at those activities are definitely self-funding there's challenges in the IT industry with flat budgets or shrinking budgets you know obviously with a flat it budget labor costs keep increasing so you've you've got a limited amount of hardware and software budget but find the biggest opportunities for improvement and find those felt self-funding self-funding activities and the only way for you really to get to that 50 50 op-ex and capex or op-ex and value added innovation kind of kind of mixes to be a ruthless follower of technology innovation i mean you cannot stand still yet the most dangerous assumption you can make is is that everything's fine and i think if you if you assume everything's fine you're falling behind so you need to continuously drive improvement in your infrastructure and i think be it infrastructure or operations practices those you'll get there but you have to make a conscious decision to say we're not going to just sit on some piece of hardware or some piece of technology for five years we're going to ruthlessly ruthlessly improve our our infrastructure mix and our operational mix. Brandon McCauley thanks for coming on the queue we'd like to do another segment of you as a wealth of knowledge we'd love to talk to the folks in the trenches making it happen certainly we'll be here from IBM and what they're putting together in their portfolio but to hear it in action is really what it's all about this is why we have the queue here appreciate your time we'll be right back with our next guest after the short break thank you