 We are here at the Red Hat Summit and we are going to talk about CoreS and Red Hat. Now these two companies I love very much so I'm kind of interested in you know to learning you know how this integration work. So we have with us Brand. Brand can you tell us a bit about yourself before we kind of get into CoreS and Red Hat. Sure, thanks for having me. My name is Brian Graceley. I'm director of product management, product strategy for the OpenShift team. Been at Red Hat for about two and a half years and just we're finishing up a great Red Hat Summit right now. Right, so there are like if you look at CoreS there are like a couple of products there. There was Container Lakes Linux and they have Tectonic which is Kubernetes and then they have Rocket which is I think Container Run Time or something like that and and they have... Quay, the registry. Yes. So how are these products going to get integrated with the Red Hat or will they remain independent? Right, so so the acquisition happened at the end of January. The engineering teams have basically lived together for the last three plus months and we've kind of kept it quiet just because we knew Summit was coming up. We wanted to give them time to put things together. In essence when we looked at these these two product sets we said you know there's an element of having a Kubernetes platform that you want to be simple to run and maintain. People are trying to make something that seems just like the public cloud but for anywhere and then there's a part of a platform that you realize that the more customers want to put more applications on there it's got to have a certain amount of robustness to be able to deal with lots of different use cases right. So the things we've done is we basically our mindset has been let's take the bolt the best of what both of those two platforms did so the OpenShift platform the Tectonic platform and put those together in a very thoughtful way so that we get the best of operations so that it's a simpler platform to run simpler to deploy applications but also the robustness and experience that Red Hat's had in terms of you know working with lots of different customer use cases that the CoreOS team never got exposed to. So you highlighted three or four products that let's stick with this OpenShift and so what is going to be the final product? Will it still be OpenShift and you will just you know merge the Tectonic technologies or there'll be a new product line? Yeah so we're gonna have from a Kubernetes perspective we will have one brand the brand will be OpenShift. Okay. So let's start with the host level so we'll start with CoreOS. So what we did is we said you know what became Container Linux originally CoreOS Container Linux what that did incredibly well was all of the automated operations over-the-year updates. Right. The one thing that it didn't necessarily have was it was built on a kernel like a Gen2 kernel which didn't necessarily have the breadth of applications certified against it that REL has had in the past. So what we did was we are taking the work that the previous Atomic team had done around you know miniaturizing the REL kernel for a immutable container bringing over the automated updates and over the air work and that repackaging will be called Red Hat CoreOS. Oh interesting. So the the Atomic brand will sunset over time Red Hat CoreOS will be our container centric immutable. So that will replace the project sorry Red Hat Atomic host. Yes. Okay. Yes. And that will be mostly is it rebranded Container Linux or Container Linux plus you know what you explain you know bringing all those capabilities to that. So really the way we look at it is there's two really important pieces we want to have the best of the operations right over the air and then we want to make sure that every ISV that's ever certified against the REL kernel can continue to just work right so we get the best of certifications customers don't have to go through any process but they get you know brand new operational model if that's what they choose to have. So that's the first step we're going to you know our offerings will be still available on REL if customers want that but also now on Red Hat CoreOS. Excellent. Okay. So that's the host level. Okay. The second thing is at the platform level the Kubernetes platform level and what we really found was when we talked to tectonic customers what tectonic delivered was a really great operation centric Kubernetes cluster centric viewpoint right what it didn't do was anything that was a really developer centric application centric DevOps centric so no pipelines no middleware no service catalogs and so forth. So what we're going to do is we're going to bring over the tectonic console so so a richer set of of operation level viewpoints we're going to to bring that into OpenShift so the brand that will continue there will be OpenShift. OpenShift's had a much richer enterprise footprint for true deployments for Kubernetes and so what what we'll end up having is if you are an application centric person logging into the platform you will see what you used to see with OpenShift very application centric very robust in terms of building applications. If you log in as an administrator you're going to see what's going to look and feel like the tectonic console all the Prometheus work so our belief is that we're going to now have a richer we're going to have as rich an experience for the application developer we're going to have a richer experience for the operator and then the third immediate piece that we'll do is we're going to take the tectonic platform operator and make it infinitely simpler to to manage and update that platform because we're getting Kubernetes updates every quarter. That's interesting because I don't know how I put it but depending on who you talk to sometimes OpenShift is seen as a not for but distribution of you know Kubernetes you know quite you know kind of far from upstream communities but tectonic was also always seen as you know kind of always upstream so is that correct? So not tech not not 100% so tectonic was always just upstream right that was the claim was just upstream and then obviously they integrated Prometheus they integrate flannel and so just taking upstream tectonic or Kubernetes never gives you enough to do everything you still have to have network storage monitor right so that was tectonic. OpenShift was always is still is everything from upstream Kubernetes and then there are some additional things that we have done that we will get back into mainstream Kubernetes but that we had done ahead of the community the community just didn't have the ability to accept them so there's some more advanced routing capabilities we have in there one or two little things but from purely a Kubernetes perspective any customer who wants to grab any Kubernetes tool from the community kubectl kube anything everything works exactly as you would expect in in OpenShift and always has. So okay so that how challenging is this integration because you know as you said you know operators will see different tectonics if I'm not wrong and then we will see you know OpenShift so how long will it take to integrate it? So so we've already got some some baseline integration done so three months we've got that done we showed a bunch of live demos this week the next release of OpenShift which will be 3.10 which aligns to Kubernetes 1.10 will be out sort of the end of June okay that will be the first time customers will see sort of tech preview of these early integrations so you know roughly four months window pretty pretty good for two for two teams and then we'll see further things in in 3.11 and 312 so for the second half of the year you're going to start to see a lot of these things come out okay the other really big thing that we're doing as part of this is that operator framework that CoreOS had initiated probably 18 months ago they had delivered a Prometheus operator an etcd operator and a vault operator last week at kubecon we we open sourced that to the entire Kubernetes community which was great we saw a lot of interest in it we've actually and then this week we talked about it as something that will be you know fully integrated fully supported by by the OpenShift platform and we got 60 or 70 ISVs who signed up and said we're fully behind this we actually saw five or six database vendors in the last week who have built working functional operators so our belief is that's going to be the new way that that ISVs are going to package their software they're consistently telling us Kubernetes is going to be sort of the new unit of compute that they're going to build against and and people love this idea that I'm going to use an operator to basically build a you know an on-demand service that'll run anywhere sort of self-driving it's got lifecycle management built in it's got an SDK to allow you to have all the knowledge about your application it'll do some basic metering so you know what's going on so that that's for us really really exciting now there are two more pieces left yep so so quay the registry from quay will will continue in the form that it was before so previously it was a standalone registry as well as an online SaaS service those will continue they'll simply be rebranded red hat quay and red hat quay dot IO over time we will probably see the the software version of quay get integrated into the platform but for now the ecosystem is still we like the fact that there's lots of choices for customers so quay continues we doubling down on quake customers love everything we've seen about quay so far tons of customer testimonials about it and last piece is arcade year rocket yeah so rocket rocket is not gone but rocket sort of served its purpose probably two or three years ago right when you know in essence core OS and really the whole community said you know what we don't we know we learned our lesson in the virtualization days where one vendor dictated the format we don't want that to happen again so you know rocket essentially became the OC the the foundation of the OCI the opening initiative there were a lot of push from chorus folks and other people you know to a lot of get Docker on the yeah so I mean core OS spearheaded that we made a ton of contribution so you know it's been a little bit now but but OCI 1.0 is out so in essence that we now have a standard around around containers you know you can still use Docker you can use OCI we're seeing the Kubernetes community adopt something called CRIO you know CRI for the run times so we're seeing now a lot of innovation around are there alternative run times that you know are more secure and so forth we announced some things this week called build on pod man and some other stuff but so rocket for the most part has been deprecated there's still a few engineers in the community that work on it but we've kind of got everybody's gotten behind sort of OCI compliant run times so that does kind of the whole history of integration I'll go back to the container OS part once more because I talked to CentOS folks and Fedora folks also and since Fedora and CentOS is kind of upstream to REL right so what so what is who is going to be upstream for Red Hat core OS so the the original container Linux right before we made any integration that community that code base will still remain open source there are still non core OS maintainers that we're working on that so we don't want to disrupt that community we'll still see core OS engineers you know keeping that going because there will be innovation that continues to happen around immutable right we'll take those you know we'll work in that community we'll take those learnings we'll put those into Red Hat core OS the the Red Hat core OS since it will now be based off the Fedora kernel we'll see you know work kind of core OS work going on in that community as well and bringing those learnings overall to the Fedora community right so you know over time you know we always talk about the importance of the operating system but there as we get into more fast-moving environments like you know people want better ways to operate them you know they want to deal with them it's mutable they want to deal with updates and so forth so I think we're going to see a lot of cross learning between these couple of communities just there's learning in the immutable community that is going to get exposed to the Fedora community that we think is going to be really good we're going to you know sort of new ideas refreshing of ideas but this doesn't look very structured the way Red Hat has things are still like you know sent to us is there Fedora is there so well there's a very structured way between you know Fedora is sort of the I mean Fedora has sponsors communities fully community driven but it's sponsored and funded by Red Hat right so what will happen to the Kuru is the Continuous Linux Community right now you're seeing a lot of non-course developers are there yeah so we will still support that community and endorse that just like we have Fedora you know if it continues to grow that's great and and we will continue to support and endorse the Fedora one as well so you know for us this is an opportunity there's a realization that managing Linux host is evolving the entire market's not going to go in one direction or the other just lots of things so we want to make sure that fundamentally we know how to work in in open-source communities we have no plans to change that we we still believe that the operating system is very important it's evolving and changing so we want to make sure we're actively involved with that and we've learned through a number of acquisitions and some good and some bad like you you don't want to do harm to an existing community and so you know same way that we're not going to leave any existence on it yet we're not going to leave we're not leaving any existing tectonic customer behind right we're not going to leave a community behind either right yeah because sometime if you do touch community it is perceived yeah that you are kind of taking over yeah so you have to be very careful that's true yeah awesome thank you so much for talking to me today and we'll catch with you in the next open source event absolutely thank you very much