 Live from Boston, Massachusetts, it's theCUBE, covering Red Hat Summit 2019, brought to you by Red Hat. Welcome back live here on theCUBE as we continue our coverage here of Red Hat Summit along with Stu Miniman. I'm John Walls, it's great to have you here in one of America's great cities. We're in Boston, Massachusetts for day one of the three day conference. And we're now joined with Denise Dumas who is with Red Hat and working on the REL8 release that just became, I guess, available today, right? Huge news. I have to first off compliment you on that rockin' these red hat, red earrings. And then I look down below, you have this, you've got the red hat sneakers on too. So you are company branded up and down from head, literally from head to toe. I'm very proud of the earrings because some of the support guys made them up on their 3D printer back at the office. How cool is that? I love it. Now we had Stephanie chair us on a little bit earlier when we were talking about REL8 and all that came with that. And we talked about the deeper dive we're going to take with you a little bit later on now. We're at that moment, just first off in general. I mean, how do you feel when something like this finally gets out of the beta stage, gets moved into a much more active space and now it's available to the marketplace? It's like. Thrilled. Fresh air, right? Oh, thrilled. Well, you know, and in a way it's almost an anti-climax because we're working on 8.1 already and we're talking about REL9, but this is just such an opportunity to take a moment, especially for so many of the REL engineering and QE team who are wandering around the summit and for us all to just kind of say, oh, it's out. It's out. Let's see if they like it. I hope they do. But you know, we've been working with so many of the customers and partners through the High Touch Beta program. 40,000 downloads of the beta and it has been tremendous feedback. We've been really pleased to see how many people are willing to pick it up and experiment with it and tell us what they like and what they don't like. So Denise, it's always great to hear the customers, but take a second and celebrate that internal work because you know, so much code, so many engineers, you know, years worth of planning and coding that go into this. So you know, give us a little bit of the look, you know, behind the curtain if you would. Well, you know, so much community as well, right? Because like everything else that Red Hat does, it's totally open source. And so many communities feed into Fedora and Fedora feeds into REL. So we took Fedora 28 and pulled it in and then did a lot more work on it to try to move it into, this year, we've done the district differently. There's a core, kernel, bin noodles, you know, and then there are the application streams. So we've done a lot of work to separate out the two types of package that make up REL so that we can spin the application streams faster. That's where things like developer tools and language runtimes, databases, the things that are more aimed at developers that where a 10-year life cycle is not a natural for those, right? And yet the core of REL, the kernel, right? That you rely on that, we're going to support it for 10 years, but you need your application streams to keep the developers happy. So we tried to make the admin side happy and the developer side happy. All right, so you're as vice president of software engineering, your team had, certainly its focus is along this way. In dealing with, I guess, the complexities that you were, was there maybe a point in the process where you had an uh-oh moment or I'm just curious because it's not always smooth sailing, right? I mean, you run into speed bumps and sometimes they're bearings and not just bumps, but in terms of what you were trying to enable and what your vision was to get there, talk about that journey from the engineering side of the equation and maybe some of the, maybe the hiccups you had to deal with along the way. So, REL 8 has been interesting because in the course of putting the product together, the REL organization went through our own digital transformation. So just like our customers have been moving to become more agile, the REL engineering team and our partners in QE and our partners in support have worked together to deliver the operating system in a much more agile way. I mean, did you ever think you would hear agile and operating system in the same breath, right? It's like, whoa. So that has been an interesting process and a real set of challenges because it's meant that people have had to change work habits that have served them well for many, many years on, it's a different world. So we've been very fortunate to take people through a lot of changes. They've been very flexible, but there have been some times when it's just been too much too fast. And so it's like everybody take a deep breath, okay, we'll do, you know, a couple of weeks, we'll consolidate. It's been a really interesting process. Clearly the kernel, you know, when we take, so we've got the 418 kernel and the kernel comes in and we have to understand what the kernel configuration is gonna be. And that can be a lengthy process because it means you have to understand when you pull a kernel out of the upstream, some of the features are pretty solid, some are, you know, maybe less solid, we have to make an educated call about what's ready to go and what's not. So figuring out the kernel configuration can take a while. We do that with our friends in the performance team. And so every inch of the way we build it, we see how the performance looks, maybe we do some tweaking, you know, change that lock. Everything we do goes back upstream to make the upstream kernel better. And so that as well has been an interesting process because there's a lot of change. I mean, we're really proud of the performance in RHEL 8 we think that it's a significant improvement in many different areas. We've got the Shaq and Larry show tomorrow, we'll talk all the way through performance, but that's been a big differentiator, I think. So Denise, security absolutely is top of mind, always some updates in RHEL 8, maybe if you walk us through kind of security and some of the policy changes. Yeah, yeah, so it's security, we bake security in, right? I mean, we have a secure supply chain and you talk about difficult things for RHEL 8, right? Every package that comes in, that is we totally refresh everything from upstream, but when they come in, we have to inspect all the crypto. We have to run them through security scans, vulnerability scanners. We've got three different vulnerability scanners that we're using. We run them through penetration testing. So there's a huge amount of work that just comes just to inherit all that from the upstream, but in addition to that, we've put a lot of work into making sure that, well, our crypto has to be FIPS certified, right? Which means you've got to meet standards. We also have work that's gone in to make sure that you can enable a security policy consistently across the system so that no application that you load on can violate your security policy. We've got NF tables in there, new firewalling, network-bound disk encryption. That actually, it kind of ties in with a lot of the system management work that we've done. So a thing that I think differentiates RHEL 8 is we've put a lot of focus on making it easy to use on day one and easy to manage day two. It's always been interesting. Our customers have been very, very technical. They understand how to build their golden images. They understand how to find, tweak everything, but it's becoming harder and harder to find that level of Linux expertise. I'll have much for that. And also, once you have those guys, you don't want to waste their time on things that could be automated. And so we've done a lot of work with the management tooling to make sure that the daily tasks are much easier that were integrated better with satellite. We've got Ansible system roles. So if you use Ansible system roles, we wanted to make it easy. We wanted to make the operating system easy to configure. So the same work that we do for RHEL 8 itself also goes into Red Hat Enterprise Linux Core OS, right? Which will be shipping with OpenShift. And so it's a subset of the package set, same kernel. But there it's a very, very focused workload that they're going to run. So we've been able to do a really opinionated build for RHEL Core OS. But for RHEL 8 itself, it's got to be much more general purpose. We've focused on some of our traditional workloads, things like SAP, SAP HANA, SQL server. So we've done a lot to make sure that those deploy really easily. We've got tuning profiles that help you make sure that you've got your system set up to get the right kind of performance. But at the same time, there are lots of other applications out there and we have to do a really good general purpose operating system. We can be opinionated to some extent, but we have to support much, much wider range. Yeah, I mean, Denise, I think back, it's been five years since the last major release. And in the last five years, Red Hat lived a lot of places, but oh, the diversity of location in today's multi-cloud world with containerization and everything happening there. And from an application standpoint, the machine learning and new modern apps, there's such breadth and depth. I mean, seems like in order of magnitude, more effort must be needed to kind of support the ecosystem today than it was five years ago. Well, it's interesting that you say ecosystem, right? Because you don't play in those places without a tight network of partnerships. So we have lots of course, hardware partnerships. That's the thing you think about when you think about the operating system, but we also have lots of partnerships with the software vendors. We've done a lot of work this year with NVIDIA. So they've been, you know, we've supported their one and two systems, right? And we've done a lot to make sure that the workloads are happy, but increasingly, as ISVs move to containerize their applications, you know, when you containerize, you need a user space that you bring along with you. You need your libraries. You need your container runtime. So we've taken a lot of the real user space content and put it into something that we're calling the universal base image. So you can rely on that layer of real content when you build your container, put your application into a container. You can rely on that. You can get a stream of updates associated with that so you can maintain your security. And when you deploy it on top of RHEL, we're with OpenShift, we can actually support it well for you. Walk me through the kind of the migration process a little bit if I'm running seven and I'm shifting over and I'm going to make the move. How does that work? I mean, how many you... Carefully. Yeah, sure. Because I've got my own concerns, right? Of course. Sure, I've got to maintain daily operation or moment-to-moment operation. I can't afford to have downtime. I've got to make sure that it's done in a secure way. I've got to make sure that files aren't corrupted and things aren't lost and so that in itself is kind of a deep gnashing moment, I would think, a bit. How do you make that easier for me? Yeah, well, especially when you've got 10,000 servers that you need to manage, right? And you want to start migrating them. You absolutely have to come to tomorrow morning's demo. We're going to do... It's live. It's always tricky, right? Live is always, yeah. But migration, so we've put a lot of effort into migration. We've looked at, we're looking at, it's no good if the applications can't come along, right? Why would you migrate the operating system? You're going to migrate the application. So we've got tooling that examines your environment and tries to automate as much of it as we can. It looks at your existing environment. It looks at what you're going to move through. It'll ask a few questions. It's totally driven by, well, plug-in equivalents. We call them actors. And they understand the various, like one understands how to do network configuration, right? One understands how to replicate your disconfiguration. It's integrated with automated backup and rollback, which is the thing that people have wanted for a long time so that we've got a much tighter level of safety there. We won't be able to migrate everything, I'm sure, right? But as time goes along, we add more and more and more into that utility as we learn more about what matters to customers. So tomorrow morning, live demo. Live demo. Get a good night's sleep tight. Put on your crash helmets. Fingers crossed. Yeah. But thanks for joining us here and talking about the rollout and I wish you well with that. Off to a great start for sure. Thank you so much. Thank you. The rail teams are amazing, right? I love my guys. Great. Thanks for being with us. Thank you so much. We'll continue here, the Red Hat Summit. You're watching theCUBE live from Boston.