 Hey guys, thanks for taking the time out to join us for our session today on our BlueMix private cloud with Red Hat solution. I know you guys are nice and energized after your lunch So let's let's keep that energy going in the session We're gonna use this time to talk about like I said our new offering a joint offering between IBM and Red Hat a Managed OpenStack service powered by our Red Hat OpenStack platform So take you through the agenda really quickly and as we go through we have a couple of presenters And we'll let them do introductions as they come up to to speak first. We're going to have our IBM counterparts give an overview on IBM BlueMix private cloud solution in general Which will bleed into Overview of our new joint solution IBM private cloud with Red Hat So that is as I said earlier a managed OpenStack platform powered by Red Hat's OpenStack Then we'll kind of have a look at the architecture of solutions He was going on under the covers how it's similar to maybe What some of you guys are running in your own data centers how that can bridge the gap between a managed service and a private cloud service? Then I'll take take the reins over and we'll look at some of the Open-source enterprise solutions that we see our customers running on Red Hat OpenStack Whether that be managed or unmanaged in a private data center And then we're going to take a look at some work that's going on on the BlueMix private cloud Solution in and some of the kind of internal IBM teams to see how OpenStack is being utilized by some of our IBM counterparts So I'll go ahead and hand the mic over to Andre and let him introduce himself and to give an overview on the IBM solution Good afternoon everyone My name is Andre Baerfield. I am the senior product owner of IBM BlueMix private cloud with Red Hat at IBM So I'm gonna briefly talk about The offering BlueMix private cloud generally Jumping to some details about the Red Hat offering Really high-level review of the architecture before we really talk about the thing that I think is the most important piece Which is what do you do after you have this cloud deployed? Pardon me on these first couple slides. They're a little bit word-heavy But I just want to get some definitions out of the way before we jump into some of the other details So private cloud is a type of cloud computing that delivers similar advantages to public cloud That includes scalability and self-service But unlike public clouds a private cloud is it is dedicated to a single organization and provides a greater degree of control Predictability and security So IBM BlueMix private cloud Actually extends on a private cloud model offering private cloud as a service With private cloud as a service customers completely avoid the burden of buying hardware of building cloud operations expertise Of selecting a cloud technology of managing cloud technology lifecycle We think that this creates the simplest manner for customers who need privacy who need that predictability But want to maintain scalability to get into private cloud Because customers aren't responsible for the capex in many cases customers aren't don't have to figure out what the platform is Customers don't have to understand the underpinnings of OpenStack in order to consume it Okay So this slide is intended to compare on the left Public cloud on the right private cloud and then in the center where IBM BlueMix private cloud Believes key value is so I'm just going to walk through this pretty quickly here So everybody knows public cloud is on-demand consumption of compute storage resources. It's fast. It's elastic It's consumption-based building There's no hardware management involved there It's awesome But some cause customers experience a regular performance because you're sharing machines with other folks You have no idea who they are There's vendor lock-in there There's also many customers have security concerns about who else is using their machines where their data is Where the data is actually What data centers they might be in So the private cloud is intended to solve that problem so private cloud offers predictable performance, you know exactly who's using the machines You know it's only your folks. It's only your technology. It's only your team using the devices You have cost control because you bought the hardware and there's no You don't have to wonder what the bill will be at the end of the month because you've invested you built this thing yourself You know exactly what's going to happen with your bill at the end of the month Security controls all in the space that you own. It's your data center. It's your network engineers your team Integrate into existing it you have your data center. You have your existing database technology So you feel secure, you know the space But it's challenging to scale private cloud Right, you have to every time you run out of space. You have to buy more machines You have to figure out how to extend the cloud technology onto new machines You have to manage the life cycle of the cloud on your own You have to figure out how to get from Newton to Pike It's capital intensive. You have to buy all the machines you pay out of pocket Sometimes millions of dollars to get the machines installed. You have to build a staff who has understanding and expertise in the area and obviously managing hardware So these there's a bunch of benefits here, but there's a bunch of challenges here So we think that with private cloud as a service You gain speed of public. It's very elastic. It's not purely consumption based building. It's monthly building which is still more predictable I think or Less capital intensive at least than building your own in your own data center It's dedicated infrastructure so you gain the predictability of the performance It's monthly billing based on node count So it won't jump that the cost at the end of the month is one that you can predict Dedicated machines without you being responsible for the maintenance of the hardware Their security controls Predictable performance. I think this is the best of both worlds for those who require the security who want it to Be predictable who want to understand what the bill is going to be at the end of the month But don't want to build expertise Because we manage all the way up into the hypervisor all the hardware all the open stack software We deliver to the customer APIs in a dashboard so they can consume private cloud Like public cloud Introducing IBM Blue Mix private cloud with red hat. So March the 31st This year we release the red hat option Which subplants the existing community-based release with red hat. So this option as you can see here. Oh Let me jump back for a second. So IBM Blue Mix private cloud generally is available in two models So there's the dedicated model which means hosted. So that's in IBM global data center And there's also a local model which means on-premises in the customer's data center IBM Blue Mix private cloud with red hat is currently only available in the dedicated model Which means it is available in IBM cloud data centers and more than 25 data centers around the world Customers in this scenario since it's hosted. Do not pay for any of the hardware. It's purely consumption on a monthly basis Totally dedicated to the customer now with red hat open stack platform powering the cloud so IBM Blue Mix private cloud with red hat is certified and managed deployment of red hat and We think that this is powerful because it enables us to remove some of the critical key pieces that challenge customers From running a private cloud enabling them to have that privacy that they deserve and need So why red hat? We continue we've continued to find That customers with existing relationship with red hat want enterprise support For their enterprise workloads running on red hat and in order for us to support that customer need We needed to extend our platform options Red Hat has a significant enterprise Linux market share. I think that number is 65 percent of the Linux market And so that means 65 percent there's a potential that 65 percent of the Linux market share Every one of those customers might want to get into cloud Some subset of them want to get into private cloud And we want to make sure that those who don't want to own all the infrastructure and don't want to own The expertise and open stacks still have an opportunity to have that privacy In their clouds red hat also has a robust partner ecosystem and supports enterprise applications Joe is going to talk a little bit more about those enterprise applications next As I mentioned earlier, there are there's both the dedicated and local model of blue mix private cloud Which is hosted and on-premises respectively And I've been blue mix cloud with red hat is currently only available in the hosted model Although we are working on Implementing the local model on premises in the next quarter. So few technical highlights here I'm gonna just brush through them really quickly. So open stack release. It's on red hat OSP 10, which is Newton API's and horizon console are available configured in HA Several networking features Scale out compute scale out block storage based on red hat sef An object storage options based on open stack Swift Customers can bring their own images customers can bring their own licenses to the cloud There's an option to bring your own IP in case you want to use private IP address range for your VMs Softlayer or IBM cloud data centers offer a direct link and private network in case you want your VMs to talk to Infrastructure in your own data center and this is naturally backed by 24 by 7 IBM global support For the hardware to the open stack software up through the hypervisor. I Mentioned that their IBM cloud. This is available in IBM clouds around the world. There are more than 25 IBM cloud data centers available This is here for your reference Okay, I think I'm running out of time. So I'm going to Very quickly talk about the architecture at a very high level here As IBM private cloud with red hat is only available and dedicated. We're just going to talk about the dedicated model So each to get started What we're trying to do here is provide Customers an opportunity to get started at a very small scale, right? We don't want customers to have to buy into a hundred node cluster to get started with private cloud And so we've figured out a way to provide a three node compute cluster as a minimum footprint But in order to deliver that there's a couple of required features. So each IBM bluemix private cloud with red hat starts with a pair of dedicated firewalls on the front end and Then a pair of dedicated open-stack controllers where the service is run in HA and then a bare minimum of a three node Compute cluster and there are three different types of the compute no cluster But I think that's beyond the scope of this discussion to get into that detail. So there's 10 gigabit per second connectivity Compute nodes can be added on up to 100 compute nodes per cloud You can add on SSD block storage, which is powered by red hat Ceph You can add on object storage, which is powered by open-stack Swift. So this is a different view of the of How the cloud can scale out it only actually takes a look at the compute perspective. There's No concept of storage here, but very quickly you see the controller nodes on the left side of the screen So there's HA proxy on the front end that Vip is managed by u-carp to ensure if there's an issue with one of the controller nodes that the IP on the front end has switched over to the other node The open-stack services that are supported are running on both of those nodes rabbit MQ My SQL and sender also running there So the minimum three no compute node and minimum three nodes of compute can be scaled out to a hundred nodes per stack The services that are supported are Keystone, Nova sender Neutron plus Elbas Glance heat Cylometer Swift and horizon Versions are available there for your reference and the technology that we use To support the controller nodes is also lifted listed here for your reference Open-stack database is my SQL and Pricona MongoDB for a Cylometer Ursula, which is our deployment technology that's home homegrown It's also available publicly on GitHub in case you want to check it out as Ansible based configuration technology Networking Linux bridge Neutron and VXLan as mentioned object storage is managed by open-stack Swift Block storage is managed by Red Hat Ceph The hypervisor is KVM from rail and the network security is via to firewall with VPN If you VPN is optional bring your arm IP is optional as well This is another view of what I've described here So it talks about the controller nodes and how things are connected, but this is primarily here for your reference I think it's also be on the scope of this Discussion I think what we want to do here is focus on Now that you have this private cloud that you consume like public cloud What cool things can you do on type of the VMs? So I think at this point we're going to pivot towards a discussion about Red Hat technologies that add value to Bluemix private cloud Thanks Andre, my name is Joe man I'm a senior solution architect that covers the certified cloud and service provider program at Red Hat So I've worked with Andre and his peers a lot over the last couple months on this solution And I think the first thing I want to talk about before we get into the rest of the slides is that we've been hearing from especially our joint customers that have been running rail workloads and looking for kind of landing spaces for those that This idea of a certified hypervisor in a certified environment to run their rail workloads And then the extended technology on top of that is very important So what we're going to talk about a lot in the next couple slides is really our emerging technology portfolio including open shift container platform Ansible ansible tower cloud forms, but really at the basis of all this And I'd be remiss if I didn't mention it is the need and the desire to have a supported platform for those rail workloads to run So that was a big motivation for for both sides of the audience are both sides of the table to work on this solution So like I just mentioned a lot of what we're going to talk about here is our emerging technology portfolio and and Red Hat focuses as much on solving Using our technology to solve technical problems and technical issues as has helping our customers and partners solve kind of process issues as well So, you know, this is a I really love this slide. It kind of describes different kind of Iterations of the of the growth our technology is seen in general over the last couple years From the development process from the waterfall and agile styles to the dev ops mentality Which is a big part of what open shift container platform helps enable as well as the idea of breaking up these monolithic and interior applications into to microservices so having these service microservices combined to provide the Service that wants a monolithic application previously provided also we see on the right side on the infrastructure side Kind of this movement from physical servers to virtualizations now to even containers So from physical servers of virtualization, I think a lot of the enterprises out there have squeezed the blood out of that turn Up so they have finally kind of virtualized their environments and now they're looking for that next level Of containerization so to speak to get a little bit more juice out of the out of the compute power that they have And then lastly pertinent to this conversation Particularly a lot of our customers and partners are moving out of the data center business They don't want to run these workloads and manage and pay for the power and pipe in their own data centers So managed and hosted solutions are becoming more and more popular with the move to move move to the cloud So as I mentioned earlier this this dev ops mentality microservices container orchestration Changing how the way our customers develop their applications OpenShift is our Container enterprise container platform that utilizes Kubernetes as a scheduler And leverages Docker image container format to deploy these microservices. So as you can see OpenShift itself is really a tool to kind of separate the development team and the infrastructure team and solve those problems separately So in conjunction with a managed infrastructure service like a BlueMix private cloud with red hat it kind of takes the infrastructure headache out Of the of the customer's hands so they can surely focus on these large development environments So what this looks like in practice is a large open stack deployment With OpenShift and OpenShift nodes deployed on top of that So you can massively scale these development environments and even scale up and scale down as the as the developer need increases, so the developers themselves have the ability to Access the OpenShift environment in a self-service manner There are multiple languages and containers that are shipped in our registry to allow them to build The applications that they need to in their environment It also allows them to automate and scale these workloads as the demand increases and decreases Obviously it's open source and enterprise grade So it kind of stands up to the test of security as well as open standards that red hat finds very important So it's built for both traditional and cloud native applications the source to image Functionality we see a lot of our customers using to take some of the more monolithic old-school applications to try to kind of put them into the The microservices mentality break them apart and put them back together to make them a little more resilient And it's obviously great for net new applications that kind of embody that microservices workflow It's truly integrated with our hybrid cloud story This this this platform runs wherever red hat enterprise Linux runs So if you have customers or partners that are running OpenShift in their private cloud data center in public clouds Deploying this this Platform in blue mix private cloud with red hat the functionality is exactly the same the developers Interact with OpenShift the same way they would no matter what the underlying infrastructure is so at red hat You'll hear us talk about this hybrid cloud mentality a lot and we want to extend the flexibility and functionality of our products No matter what the underlying infrastructure is so that's a big part of red hat story This is just a little quick diagram of the solution. So let's see if I'm a little laser here So there are really two kinds of servers in the OpenShift deployment We have our nodes here, which are actually housing these pods which are essentially containers Docker image format containers here Our schedule or over here the master nodes essentially take care of all the connections and routings that go on in the nodes So if we have a pod here That is connected with a service layer to another pod on another node Kubernetes in the mass in the master here make sure that all these services are connected and then they serve out a Service through this routing layer that end users can access without having to really understand the inner workings of OpenShift So it's a fairly It's a fairly complex Kind of configuration under the hood, but at the end of the day it allows our developers to easily deploy these applications at scale and with the master nodes in the environment we can automatically kind of Regenerate these pods if pod A goes down that's connected to pod B spin them up on different nodes Define where the pods are deployed with With zones and rules and have a lot of control over our application platform So as I said earlier the service layer and down here on the infrastructure layer we can run anywhere We're a red hat enterprise Linux run. So very very flexible application platform It can it can integrate with it would get existing CI CD pipeline as well as Red Hat Enterprise Linux at the base of this can be integrated easily with the existing Operations teams tooling especially if they're rel shops already. So that's You know a huge part of the tenability of this solution Another layer above the OpenShift solution We've also released our full middleware platform At in our can our container registry. So J-Ball shops can now deploy All of these services feed Henry with a caveat feed Henry's coming soon All of the rest of these services can be called and deployed in OpenShift With with doc doc or pool commands and managed and deployed in pods So that's huge for our customers that are middle big middleware shops and J-Ball shops already because they can they can continue to deploy EAP applications fuse AMQ BPM processes in OpenShift without having to relearn kind of the infrastructure and the technology another big part of our workload deployment on OpenStack in general and especially with BlueMix private cloud with Red Hat is is our Ansible tooling Using Ansible to deploy workloads in OpenStack no matter where that OpenStack lives so to speak So I'm sure most of you in the room are familiar with Ansible, but Ansible is an automation language that Perfectly describe an IT organization's environment with kind of simple human readable what we call playbooks So you don't have to have any kind of special coding Expertise to be able to track through an Ansible playbook to understand what that playbook is responsible for so it's a really accessible Automation language we acquired Ansible. It's been about two years now So Ansible itself is completely open source available to anyone out there in the world and then Ansible Tower is our enterprise kind of Ansible management web UI tool that we offer so Tower provides role-based access control for your organization to allow certain users access to certain playbooks to deploy Applications that they should be able to without having them metal with infrastructure that maybe they shouldn't have access to so this Ansible and Ansible Tower solution I think is going to play a big part in helping our customers deploy workloads in BlueMix private cloud with Red Hat that they Previously have deployed in their own private data center So the playbooks don't really change much except for the infrastructure piece So you now direct the Ansible playbook to deploy this same solution in BlueMix private cloud with Red Hat with a very very minimal Re-architecture of the playbook itself So that's going to be a big part of kind of what we could you know migrating workloads in this in this hybrid cloud model between Different providers in different environments So what's coming up? Down the pipe for the Red Hat our IBM BlueMix private cloud with Red Hat solution So the first one of the first couple things we want to look at is integrating cloud forms with the solution Obviously, it's open stack Red Hat open stack platform under the hood. So the technology work is not going to be Much heavy lifting we just it's just a process So we want our customers that are using cloud forms to manage their current on-prem Private cloud to be able to have visibility into the BlueMix private cloud as well. So that's on the roadmap We're also working with SAP to certify rail for SAP HANA on the BlueMix private cloud platform So we'll have that we hope to have that so our customers joint customers can deploy SAP workloads Obviously, we're we have the IBM has a strategy for rolling upgrades of the environment So as new versions of open stack Red Hat open stack are released customers can enjoy in place upgrades take advantage of new features And finally as Andre alluded to earlier in his presentation BlueMix private cloud with Red Hat the local offering so this same solution in customers data centers as opposed to in IBM BlueMix data centers So with that I'm going to hand hand the mic over to Paul who's going to talk a little bit about what they're doing at IBM with With BlueMix private cloud Hi, my name is Paul Tchaikovsky. I am the technical lead for this current release And I wanted to talk a little bit about some of the things we actually run on top of open stack ourselves and also Talking about like the composability that it brings to the table and also how by Utilizing the open stack inside of software data center. We also get a lot of interesting hybrid capabilities So we we offer open stack in our data center or in yours And if it's in ours even if it's in yours we we get access to all of the All of the tools and features we have inside of soft layer and all of the BlueMix services So as well as running open stack VMs you get access to like the BlueMix container service You get access to running physical machines in soft layer and whatnot and they can Talk to each other fairly easily because they can be in the same data center And we do that at a blue box So we basically On on composable infrastructure, I like to think of it in two steps We have what we define the infrastructure as and we were usually using open stack heat for that But we also use the the soft layer APIs to get machines out of soft layer And then we use vagrant and and and other tools to do like dev environments and stuff And we combine that with Ansible which does all of our application Installation and configuration and that separation of duties works really well for a number of reasons Which we haven't got a lot of time to get to go through but if this stuff I'm talking about interests you hit me up and we can talk about any of this stuff in depth so with Ansible we have like We have this monolithic repo for what we call site controller, which is like the operations platform that runs all of our open-sat clouds and some of the other Software that we support a blue box and you can see there's a there's a lot of stuff going on there We have around 25,000 lines in our playbooks Plus templates and everything else. So it's it's there's a lot going on We do a lot of access and security stuff bastion with Two-factor orth. We have the elk stack for logging. We have monitoring. We have a lot of miscellaneous stuff Jenkins We have a bunch of stuff for doing Compliance as well as a bunch of stuff for mirroring yum repos and Pipe I and stuff like that so that we don't have to reach out to the Internet and we can make sure that the version that we're asking for is the Version we get and we don't get messed up by someone updating something in pi pi or whatever And it's it's composable. So using the Ansible inventories we can really Describe what we're going to put where this is kind of the site controller portion of Of it, so let me actually go back here. So with the composability of this allows us to say hey, let's Let's do some ci. Let's spin up, you know Jenkins server and a bastion over here and run all of our open stack tests Through this or run of all our site controller tests through this This is site controller itself and so you can see we have we have two of we have two central site controllers And they're all connected to every data center We have a blue box cloud and each of those has It's got like your elk and senzu and whatnot for for that. And so this is all basically composed together with a combination of heat and and the soft layer APIs and then when we're running it in your data center, we have a bunch of Pixi stuff that were operates similar to say cobbler and that sort of does our infrastructure for the Infrastructure itself and then we have ansible to then deploy based on our inventories and the roles and whatever to To get our applications stood up And so we can go from having no no infrastructure in one of our software data centers to having Site controller hooked back to our central with mirror access and all of that and then get open stack installed Within a very short amount of time I think a couple of hours from from nothing to complete if there aren't any hiccups along the way Just from the power of kind of taking that two-step Method for doing our Automation And that was I was all I'm gonna talk about because I wanted to leave plenty of time for Q&A All right So I wanted to go over this slide really quickly before we take a couple of questions and also really quickly I think I don't know if any of us mentioned this But I think one of the huge takeaways for this and generally for our customers and partners is that? And you know Andre talked about this a little bit But we feel that this is a great way for our customers who maybe don't have the expertise or the workforce to deploy manage and run their red hat open stack environment to be able to You know kind of dip their toe in explore open stack as a solution or a fit for their for their IT landscape With kind of a minimum investment for that for that starting footprint So, you know the consumption the monthly consumption model is is all the rage today, and we know that so we're trying To make this solution as accessible as possible to our customers Especially those that maybe are out of the business of maintaining their own internal infrastructure But still want in need the technology that open stack provides So, you know next steps for those of you that have connections into IBM or red hat if this interests you reach Out to your respective account teams to start that conversation You can also reach out to myself and Andre and as well as as member Mohammed who is Doing another session on the solution on the IBM sponsored track that you'll see at the bottom That's tomorrow at 340 and 208 so That they're gonna talk with a lot with the the first customer on the solution clouds off so The contact from clouds off is gonna talk about how they're using the solution So if you're interested in seeing kind of what's happening in the real world today with the solution as it as it exists That's gonna be a great session. We also have a couple press releases and blogs on the solution That just do a good job of setting the landscape and a little a little bit about what the solution does and what it is so If we have any questions, we have probably about five minutes left I believe so if there are any questions about the solution, you know, feel free to to speak up Yeah, I'll pass the server to Andre to talk a little bit more about the hardware options so there So the answer is no I guess So this the options that are available are those that are available because they're managed by us the best way for us to Effectively manage hardware is to keep a very narrow set of hardware that is is well known to us, right? Yeah, that that's a that's a great question We didn't really touch on that the question for those that didn't hear it is where does IBM's management stop? And where does the customer or the partner that's going on top of that start so IBM essentially manages the open stack Solution the compute the controller up to the hypervisor level the customer and or partner brings their own images Brings their own solutions and own images that are stalled installed on those images And they are responsible for the deployment and whatever happens above the hypervisor layer So this is I think a big value prop for a lot of our customers that don't really want to become open stack Specialists they can simply get access to the API's of the infrastructure and build their solution on top of that That makes sense. Yep And for those of you in the room that have ever deployed a production open stack environment before 72 hours is like Unbelievable so that's a huge huge huge value prop for the solution as well anything else Well again, thank you everyone for for coming Let our contact information is up there if you if you want to reach out to us after the fact And we'll be hanging around here for the next 10 minutes if you have any other questions. Thanks again guys