>> Narrator: Live from San Francisco, it's theCUBE. Covering Red Hat Summit 2018. Brought to you by Red Hat. >> And welcome back once again to theCUBE. We're here live and open in this open-source conference, Red Hat Summit 2018 here in Moscone Center in San Francisco. My name's John Troyer. We are coming close to the end of the day three of Red Hat Summit. Been here, catching all the live coverage on theCUBE.net. Great to have with us: our two guests here from UPS, Innovation Award winners here at the show, gentlemen, welcome. >> Thank you. >> Thank you. >> Mark Falto and Dan Savarese. So, welcome folks. So, we're going to talk about your journey to using OpenShift, how you guys picked it, what you guys stood up, and as we were just kind of, I saw the on-stage story, and as I was just talking to you now before we went live, I'm just so impressed by the time to market, time to value, that you guys were able to achieve. You and your teams. Which, if you think about it is really, the fact that this is a real story, and it's not just a marketing example is really great. We're living, and sometimes I wake up and I say we're living in wonderful times in 2018. So, Dan, kind of set the stage for us. You are a principal infrastructure architect, you're one of the folks that helped bring the system in, you were already Red Hat Linux ... >> Yes we were. >> ... users, but what were you looking at as you were trying to make this decision, and what were some of the drivers to bring OpenShift in-house? >> Well, we knew we wanted to go cloud, but we weren't sure whether it was public or private, so we felt that in order to start the transformation to cloud we should really focus on private. These boundaries to get that up and running, and a way to modernize our applications to be cloud ready. So that was the goal when we set this up. We had a very tight timeline, we had applications that wanted to go cloud, so we made the decision... >> Mark Falto: (mumbles) was knocking at the door. >> Everybody was knocking at the door. Right, so it was a matter of just what did we want to do? Like anything, we viewed a number of different private cloud solutions, and we really liked OpenShift because of it's flexibility, it's open-source capabilities, and the fact that it was... docket containers, which was our container strategy going forward, which we wanted to use. >> Was this the beginning of your container strategy? Had you been using them before? >> No, we hadn't been using them before. We just made the decision, prior to making the cloud decision, that we wanted to go containers and docket with the containers that we wanted to use. >> So, you some sort of evaluation, you say you know, this seems like something worth happening. In the olden days, we'd go off and you'd do some sort of POC, and you'd spend a couple months doing that, and then you'd look at it, and what are the toy projects? >> Dan Savarese: Right, yeah. >> You guys wanted to actions, so can you talk a little bit about that and the timeline there? >> Right, so we made the decision in late fall 2016 to do this. My team runs all the infrastructure, architecture, so we work with the applications to design new architectures for them, and basically we started working with Red Hat on success criteria that we established for the product, and then once we got through that we started having sessions with Red Hat and using collaborative dev ops approach with everybody in your organization who'd be affected by this private cloud we're putting in. So Mark, our info sec folks, our networking folks. We were on a very tight timeline, we had an application wanting to go quickly as possible, and they wanted to be up and running in like the late spring, early summer timeframe, so they didn't give us much time. So, a lot of work and effort into figuring out how we wanted our architects OpenShift for, not only to be operationally successful for us, but from an application perspective. So, it was important we do this in a collaborative manner, and get everybody's input in doing that. >> Yeah, that's something that's going to be interesting to dip into, right, because you can't just turn on speed like that. As I've been kind of jokingly referring to, right, you have to turn into kind of a dev ops in an agile organization. Even at the infrastructure layer. >> [Dan Savarese} Yep. >> So, Mark, within a few months you've got OpenShift up and running. Now you've got to put some apps on it. These are not new apps. You have all your existing portfolios still having to run, but so, yeah, what were you looking at putting up there, and how did you approach that in terms of native practice. >> Our strategy was to take new applications. We're trying to find app teams that we thought had at least a sophisticated enough process that they could take on the automation that we really wanted to drive with the platform as well, right? It was not just containerization, but the transformation in dev process that came with that. So it's get a pipeline in place. Understand how to use Jenkins, how to use the plugins that are necessary to make that happen. You need the right app teams that are ready to take that on. We had an example application as part of our edge initiative called Sipe, which that team we thought was ready to take this on, it was the only way we were really going to meet their timelines, too. So we worked with them for a number of weeks, and not just us, but we also had Red Hat partners helping us too, to freely build out the automation for the pipelines, get an example running for their complete automated pipeline. >> Right, so can you describe a little bit what it does? It's a business line app to help managers do some decision making and planning. >> Yeah, it's a decision assistance application for supervisors at hub facilities where, you know, where you move packages. >> Okay, so real business impact. Before, they had, I don't know what they were... Either papers, or... >> Most of it was manual processes, and it wasn't like the speed to market. The information wasn't real-time, so Sipe was all about driving real-time decisions in the field. >> Right, right, so it must have had an immediate impact, then. So, it sounds like you were up, and then also within a few months of it. >> Yes. Yeah, yeah. I mean, we're able to get at least a pipeline going within weeks for them, and that demonstrates the capacity to get yourself to production, right? And then, they're in production within a number of months after that. A couple of months after that. >> Yeah, that's great, and I'm assuming you'd then be able to revise it and kind of improve the functionality since. That and some other apps. >> Mark Falto: Yep, yep, yeah. >> Was that a big shift for you and your developers, to kind of get to this stance, of frequent releases and a pipeline? >> Yeah, it's a huge process shift, it's a cultural shift for app teams, they have more capability from the infrastructure than they have ever had before. So, they now have tools to deliver much faster than they're used to, so they change their team structures to help us facilitate that. They bring in Red Hat or other consultants to help them backfill their skillsets, so it's big transformation, yeah. >> Nice, nice. Now, I wanted to explore a little bit. One word you hear a lot here: hybrid cloud, multi cloud, we've been hearing that a lot this week. This particular app runs on Prem, right? >> Dan Savarese: Runs on Prem, yes. >> Let's make sure we it right, on PremiSys. >> Dan Savarese: On PremiSys, right. It runs actually, between today at the centers. >> Yeah, I'll get extra bonus points for that. But, as a portfolio that you have to manage, your IT department and infrastructure, UPS does take advantage of different clouds and different code running and different places, as well as sas and everything else, I'm sure. Can you talk a little bit about to portfolio? >> Sure, so we have a wide variety, as companies as large as UPS, you can imagine we have a lot of IT solutions. We leverage all of them to our benefits, so we have past solutions in the cloud, a lot of analytic stuff that we'll do out there, and we also have a lot of sas solutions that we use to do different various works across our organization. So, we are using cloud in multiple ways. Out next journey is hybrid cloud. How do we take OpenShift now, 'cause what we have is a situation where we need a lot of processing power for what we call out peak season, Christmas of course, right, but when we size today on OnPrem we size for that peak season. So, the challenge is for us now is how can we use OnPrem for, say, nine months out of the year, then expand into the cloud for peak season, reduce cost, and then drop that down after we're done with peak season? To really drive us really efficiently from across the spectrum. >> Right, right, right, and obviously OpenShift is going to be a probably key element to that. >> Dan Savarese: It is a key element. It is a key element. >> Yeah. You know, this only took a, this journey was, if you just look at the timeline we just talked about, really only a few months, and it seems like this was kind of a corner-turn for your engineering organization. I mean, is that inaccurate? >> I would say so, both from an infrastructure perspective, the biggest thing when you run a common environment, which we do in OpenShift, in other common environments you have no control over how the applications affect one another. For us, what we like about OpenShift, it gives us that capability, so application 'A' doesn't step on application 'B'. And, I think, for us, it just made our lives a lot easier from an operational perspective for that reason. >> Right, it wasn't about the tool, but the tool helped enable the processes, and then that yielded the time to market and time to value. >> Yeah, and from an app side, building these new architectures really requires containerization. It requires the automations, so we can't attack the proper microservice patterns and practices really without OpenShift as a platform underneath it. It's foundational. >> And I'd like to stress the fact that it really was all of us knew we had to get something done. We all came together. There was no... The silos were merely broken down. We knew we had a mission to get through, we knew had something to get done in a short period of time, and we just came together in such a strong, collaborative way of driving the solution. >> Wow, that's great, and congratulations. Innovation Award here at the show, it's been a great week here at Red Hat Summit. So, Mark Falto, Dan Savarese from UPS. Congrats, and thanks for being here on theCube. >> Well, thank you John. >> All right. Well we are here live in San Francisco. We are finishing up day three. We'll be back after a short break, and all of our live coverage of Red Hat Summit on theCube. (light percussive music)