 Live from Barcelona, Spain, it's theCUBE, covering KubeCon CloudNativeCon Europe 2019, brought to you by Red Hat, the CloudNative Computing Foundation, and ecosystem partners. Welcome back to theCUBE here in Barcelona, Spain. It's KubeCon CloudNativeCon 2019. I'm Stu Minimai, co-host for two days wall-to-wall coverage is Corey Quinn, and we're always thrilled when we get to speak to a user, and not just any user, but Fernando Alvarez, who is a cloud architect at Xpy Orange. Fernando, muchas gracias for joining us, sir. And so, Orange, we are familiar, many people are. Xpy Orange, though, maybe you can explain to our audience a little bit what this group is inside of a large global brand. Xpy Orange is a subsidiary from Orange, Spain, and from an orange telecom group in France. And what we try to do is to reinvent the way that telco companies operate, going more in a software way than the traditional way. So it's more or less what we're trying to do, and we started operations in September of last year, just with a different proposal to see if it could make it viable for the ecosystem, smaller, medium businesses in Spain. So digital transformation, many people talk about it, but I've had some really good conversations with customers in the last year or so. Data is so important to businesses these days, being data-driven and being software at the core of what you do. So it's sometimes overstated that, well, every company will be a software company sometime in the future, but you have done these transformations before, and that's what brought you into Xpy Orange. So tell us a little bit, your role as a cloud architect, what's your mission and what's your role in the org? Well, my mission is to make all the different pieces inside the whole IT stack to work together, especially in a cloud environment. So from the designing, from the whole ecosystem that supports the platform, and at the same time, supports the whole company as a telco operator or multiple telco operators. What my role is to make sure that everything fits together. We're trying to accomplish it, and we're very happy to have it in a cloud environment, in public cloud and using Kubernetes as our container orchestration engine. Yeah, so can you lay out in one public cloud, many public clouds, data centers, and what is your... We are now in one public cloud in AWS, and but having this cloud orchestration layer allows us to move to or to go multi-cloud or hybrid cloud as soon as we want to do it. But I think that we have to keep it simple from the beginning, having a type schedule to start operation is key to have the, our value proposal into the market and to do so, we have to do it in a simple way. So going first in one public cloud, going public cloud first, because it's not a logical movement in a big company, even though we are on a spin out, but normally big enterprises want to do in their own way in a private data center. So what we wanted is to be very fast and to do so, the election is clearly logical to go public cloud and to have an orchestration or engine like Uber data studio or everything. Do you find that making decisions that enable portability in the future, if you want to move to alternate clouds or go hybrid, is in any way constraining what you're able to do or the speed you're able to innovate with? Yeah, but I think benefits are way better than the drawbacks of that. Normally every single decision you have to make about the architecture of any piece, one of the key aspects is to see if it involves vendor locking for any of the components on the stack, for example in the public cloud. But I think it was the effort because most things that you can design as an engineer or as an architect can be solved not only using a specific solution from the specific cloud provider but using a more generic way. In this way, then you can assure that you can move more or less easier to other cloud or to other infrastructure. All right, so I guess it begs the question. You said it's AWS today and Kubernetes, it's OpenShift, yes, that is the Kubernetes platform. How did you come about choosing that and obviously Red Hat, one of their strengths is working in lots of different environments, so as you go to that hybrid and multi-cloud, was that the driver for them or were you a Red Hat customer? How did you end up with OpenShift? Yeah, that was one of the drivers. The other one was the support for the platform. We were in a really tight schedule and we knew Kubernetes well enough but we weren't sure if our knowledge were enough to be in operations in only nine months. So for that, we get Red Hat on board to have all their knowledge in terms of support and the professional services to help us to define how to do things with their platform on OpenShift and because OpenShift is a Kubernetes distribution, we were sure enough that we share the Kubernetes way of doing things. So that for us was a logical selection. What was it that drove your move to the public cloud in the first place? And I guess your entire digital transformation by extension. Did you say what, sorry? Yeah, what drove your entire decision to first go to the public cloud and secondly to go, I guess as part of your larger digital transformation. Yeah, the main reason probably was the speed. At the beginning, the whole company was started thinking that we were going to build our platform on a private cloud but once we made the numbers and see that needed one more year to start operations with zero value to the customer, the decision was pretty easy. Let's go public cloud and let's think about this if it's really at value in the future. All right, so Fernanda, if I heard you right, you said nine months from when you went to deployment. Big companies aren't necessarily known for their speed of change. Talk a little bit about the organizational dynamics, how much internal ramp up there was versus relying on your partners and your vendors to be able to help you meet those schedules. The good news is that we had the full orange support to start a new company and we started as a separate company recently because we wanted to be very fast. So instead of having all the processes from the big company to do something that maybe it could fail or maybe could affect the brand, we decided to start a new company from scratch with orange in its name because we have all the well-known, all the brands is well-known in the world. But at the same time, we wanted to start from scratch. That's why we started with little people most of them were coming from the software industry instead of the telco industry and we started to build from scratch the whole company and that we were 20 in February 2018. Now we are more than 200 and we started operations in nine months from January 2018, so I think it was really completely success in terms of speed. If you were going to do it all again starting over, what would you do differently? That's a really good question. Probably I will put more effort, even more effort in transmitting the right culture because when you grow a lot, you have to be very carefully and transmitting the right culture to the new commerce because it's very easy to let's dissipate the culture that you create at the beginning when you are only 10 or 20 people and it's very difficult to maintain it when you are 200 and then if you are 200 with a wrong culture, you are transforming yourself in a big company with a small revenue, so that's something that needs to be taken into account. Okay, so what's the roadmap from here? Does the 200 then help infuse into the rest of the company or how do things work going forward? What are we doing now is to, we build up a completely new IT stack that was from the beginning multi-tenant to host multiple telco operators and now we are hosting our second telco operator, that's Orange Spain, Brands for Small and Medium Enterprises that is now coming to our stack. So this is in our roadmap for this year. What we are doing is integrating all the stack from Orange Spain to the new one and at the same time trying to complete our portfolio with new products and these new products could be managed and commercialized by Exbiorems as a telco provider and also by Orange Spain as another telco provider. All right, when people look at the show, there's so many projects going on and so many different pieces. We sometimes hear there's a lot of choices and how do I make them? How did Exbiorems, how did you figure out what pieces of this stack was red hat, mostly prescriptive as to how you do or were you choosing the service mesh and all the other various pieces and what can you tell us about your stack? Well, what I can tell you is that we put a lot of effort on designing this stack by ourselves, not having any turnkey solutions because we think that this is key for the success of the company because normally telco operators put a lot of effort in their core network but don't put so much effort in the technology, in the software technology but now things are changing a lot and we really think that the software layer is as much as important as it was the network and here is the real, the perceived value from the customer now resides in the software part so we design in each part individually and we selected the right partners for starting the development on each part and then make all together to work instead of going of a full stack provided by a unique company. Perfect, as you've gone down this path have you started to look down the serverless environment at all or are you strictly in a, I guess in a more container based approach? Let me broaden that a bit, are you looking into functions as a service and other I guess serverless technologies or are you mostly keeping it to I guess more commonplace things that are like a half a step back? Well, in telco industry, what is traditionally the vendor, the traditional vendor for the telco industry are the network vendors that are more in their way of virtualization instead of the contentization not even to mention the deployment serverless. So we are putting a lot of effort on making them to understand and some of them are understanding it really well that it's key to have their products be able to make an extreme automation, sorry. So it's a pity that we don't have enough time to use technologies like serverless. We use them for little operations in our internal stack, but we are not at the point of using it in products that we have because what we are doing is trying to, for example, to move the management part of the network services to the containers and now our efforts are in that place. And to be very clear, that's absolutely the right answer. You have to meet your customers where they are with things that are appropriate fits for the problems that they have and advocating for a technology stack because oh, it seemed like the right answer when I pulled a bunch of people on Stack Overflow or something is never the right answer to solve those problems unless how do I make people on Stack Overflow happy is the question. Spoiler, you can't. Yeah, that's completely true. So Fernando last question I had for you is, here at a big show, what are you looking to get out of the show? What excited you to bring you to the event and any other things around your experience so far what you're hoping to do that you could share? I think that the most important thing when we are talking about digital transformation for any site company is the people and the mentality of the people. So I can never say enough times that we really need to invest time with people to embrace the change, to embrace the kind of culture that is behind the cloud native mentality because if not, if we don't do so, what we are doing is just porting our old stack to a new technology without changing anything. So putting that there for talk with people make that this change happen together with people that is working already in big companies is key for the structure of any story. All right, well, Fernando Alvarez, really appreciate you staring at your story. Congratulations on the progress so far. Thank you very much. Best of luck in the future. Thank you. All right, for Corey Quinn, I'm Stu Miniman. We'll be back with lots more and thank you for watching theCUBE.