 Welcome to our first episode on the Learn Live Azure Hybrid Cloud Study Hall on Learn TV. Welcome everyone who tuned in this morning, depending on where you are, maybe already in the evening or in the middle of the night. My name is Thomas Maurer. I work as a Senior Program Manager for Azure Hybrid at Microsoft and I'm joined by Lisa Clark. Hi, Lisa. Hey, Tom. I'm so excited to be here. Hey, everyone. Thank you for joining. Let us know where you're joining from. My name is Lisa Clark and I work at Dell Technologies. I'm also a Microsoft MVP and at Dell, I'm focused on all things Microsoft Hybrid Cloud, specifically the Azure Stack portfolio of products and increasingly more and more Azure Arc. I'm super excited to be here. This is one of my favorite topics to talk about. No, that's awesome and I'm super happy to have you here. We definitely have the right person here to talk about today's topic which is going to be our first Learn Module of this series, and it's basically an introduction into Azure Stack, and especially as you just mentioned in the Azure Stack portfolio. You can obviously follow us and do this Learn Module by yourself. If you're new to Microsoft Learn, no worries, I will explain this to you in just a bit. But if you're familiar already, we have a QR code here, we also have a link here which you can go, and this will take you to the Learn Module we will go through today. Please join us here. Tom, you can have an interactive experience, but you can obviously also watch this episode here with us, and we will basically go through these topics, and you can also do it later at your own time. Just be happy about that. What are we going to talk about in this specific Learn Module today? What are the objectives, Lisa? As I mentioned in my introduction, the Azure Stack family is in fact a portfolio of products. We're going to learn about each of the products within the portfolio and be able to describe those. We're also going to look at describing the basic architecture of each of the products, with a focus on the primary use cases for each of the products, and talk about some of the core capabilities. So for each of the products, as you see there, Azure Stack Hub, the OG, Azure Stack Edge, and Azure Stack HCI, we're going to basically be able to describe the basic architecture of the core capabilities and focus quite a bit on the use cases, because it is a portfolio of products, and so sometimes you don't can get a little bit confusing when I use which solution or what combinations I use, so we're going to spend some time on that for each of the products. Okay, that is awesome. And again, if you're not familiar with Azure Stack right now, don't worry, that is why we are here, and that is why we are doing this Learn Module today. Also, if you have questions and comments, and you just want to say hi, feel free to use the comment functions. We have our producer, Laurel, in the background who will help us find you, and then obviously answer your question if we have enough time. So with that, I want to directly move to Microsoft Learn, and for that, we have already put up our Learn Module for today, and so I just want to take a brief moment to actually explain what Microsoft Learn is, and so Microsoft Learn really is our free learning platform for everyone. It's not just about Azure Hybrid Technologies, by the way, it's also for other Azure and non-Azure technologies like Microsoft 365 and so on, also some open source technologies as well. So if you're interested in learning about different products and services and even metallurgies as well, you can actually go and check that out. So we have the concept of these different learning paths, and so we have one which we're gonna start with today on the Azure Stack HCI, what we call the Azure Stack HCI Foundations, but we also have one on Azure Arc, which we will have later today in the day available as well, and we will mix these up, and by the way, those sessions are recorded, so even if you cannot join all of them live, feel free to also check out the recordings. Obviously with the live experience, you can also get your questions answered. And so each learning path has different modules, and so if you're here in the first module of this series, introduction into Azure Stack, and you can see here we went through the learning objectives. There are also some prerequisites. I will not focus too much on too strongly on the prerequisites. It's really, if you're not, you don't need to be an expert on these topics, but you should have heard probably these words before, and that's good. And then on the bottom- We've heard the words before is enough, because the prerequisites are great. I love Microsoft Learn. It's an amazing tool to learn about, especially for beginners, just learning about the cloud basics. I love it. But yeah, on the prerequisites, they are just a guide. You do not need to have sort of hands-on experience with any of these to work your way through the modules and to gain knowledge from the modules. So I can vouch for that. Yeah, awesome. So each of these modules then has what we call units, and basically there's just different sections in this module. And again, as Lisa explained, these are the ones we're going through today, and you can also obviously go on your own pace as well. We will go through these, and we will talk about the different services and products and the Azure Stack portfolio. We will also, by the way, have knowledge checks. So in this case, we have multiple knowledge checks in that module, meaning we have a couple of questions after a topic, which we will go through and basically try to answer that with you. So you will be able to actually also answer the questions live through a poll, and then we can see what people think is the right answer. And then we obviously will tell you what the right answer is as well. So with that, I think, Lisa, we should really go and start with the introduction of this module. Yes, let's get started. And encourage everyone to follow along with us. That would be great. We've been here for a while to walk through it with you, so definitely join along with us and especially get involved with the knowledge checks as well. Yeah, absolutely. So for the introduction part, obviously there must be a reason why we have our Azure Stack portfolio and why you do hybrid, right? And maybe you can explain a little bit. I mean, you're working with a lot of customers on this and you obviously did a lot of projects and you know a lot on that topic. What are the reasons why a customer actually would do hybrid or why would they actually care about that topic? Yeah, so it's a very interesting topic. I think for many years, there's been a lot of various reasons thrown about as to why you might not use the cloud. I think we've come a long way since then and we've realized that actually there are perfect scenarios and use cases where you would use or have your workloads living within Azure. But then of course there are some workloads and some use cases where that just doesn't work. And what I love about Microsoft and Azure is that they don't want those workloads left out. They want them to be able to benefit from all of the great technology that you get within Azure. So some of the scenarios that I see within my role at Dell is I think, data regulations and privacy are a massive one. So anything in sort of healthcare or public sector, we see the need to keep the data on premises. And sometimes that is a real need and sometimes it's because you know regulatory frameworks et cetera have maybe just not evolved at the pace at which technology has evolved. So yeah, sometimes they are very real and sometimes they are perceived but there is definitely a need to keep data on premises. And then you know, we all know the latency and bandwidth use case is very, very real. There's not so much we can do about that more often than not. And so sometimes for latency intensive applications we just need to keep them on-premise. We need to keep them close to where they are being used. One of the scenarios actually mentioned in here and I love this. I love that Microsoft Learn follows a story. I think that's always great for learning and I know that there's a sort of number of fictional companies that Microsoft Learn follows and they'll give you in a scenario and that's what we can see in this introduction module, right? And one of the scenarios there is around virtual desktop infrastructure. And we're gonna come on to that. We'll talk about that I think when we get to Azure Stack HCI but the need to have that running on-prem and closer proximity to its users. So there's a lot of very real use cases and reasons as to why workloads need to run on-premises, in a data center, at the edge, in a factory, et cetera. And what I love about Azure is the fact that they don't want any of these workloads to be left out. Do you have anything else to add on to that, Thomas? No, that's a great, absolutely great summary on why you want to do that. And again, I see that a lot obviously. And we also acknowledge that, right? I just wanna, I always highlight one thing which Jason Sanders and he was the engineering lead for all the Azure services did during the Microsoft Ignite keynote in 2019 in Orlando when we still had in-person events. He basically said that we believe that hybrid is going to be an end state for our customers and not just the in-between state until everything is moved to the cloud, right? And I think that is where you draw a strong message. It's like that we actually like know that, hey, there are reasons and you pointed all these out, right? I just worked with a company who has, for example, factories all over the place or retail stores. And even though they have a good connection and they could run their applications in the cloud, they don't wanna rely on that internet connection to the cloud because what that means is if that connection stops working, the factory would stand still or the retail store would not like work, right? Imagine if you can't wanna go shopping and you can't because they have no internet anymore. That would be a disaster, right? Like for me at least. And so it's important to also like have these edge locations involved, have that available. And again, you also mentioned things like regulatory compliance and so on, which is absolutely like I see that a lot too. So with that, you also have a little bit of a overview to share how these things actually like our services, our hybrid services come together in that sense. Yeah, absolutely. So the sort of Azure Hybrid Overview. So let me, let us put your- Oh, yeah, let's get this light up. That's okay, don't worry. Perfect. So yeah, the Azure Hybrid Overview pitch is something that I have been doing almost every week for a year now in terms of pitching this to Dell's customers, Dell's partners, internal, Dell sales, Dell pre-sales, basically anybody who wants to hear it, I am there and I am talking about how Azure is hybrid by design. And so I wanted to share just a little bit about that here because I think it really sets the scene for this module today on Azure Stack, but also the next module later today on Azure Arc. And so I thought we would just go ahead and kind of share a little bit about that. So what I love, like I said previously about Microsoft is that they understand the real need for workloads to live outside an Azure data sensor. And they are, in fact, they are hybrid by design. And I think this comes from the fact that Microsoft is a little bit different from other cloud providers, shall we say, in that Microsoft have heritage and experience in the on-premises world, right? So they understand, like we do at Dell, the complexities of the on-premises world. And so I think that is what's driven Azure to be hybrid by design. And how do they do this? They do it via two mechanisms. So the Azure Stack portfolio, which is what we're gonna talk about today, and then Azure Arc, which is gonna be in the next session. And so today we're gonna look at the Azure Stack portfolio and that consists of three products. So we've got Azure Stack Hub, we've got Azure Stack Edge, and we've got the newest product in the portfolio, which is Azure Stack HCI. And we're gonna dive deep into these today in terms of their use cases, their similarities, but also their differences. And you'll see I've got a line there that leads to Azure Stack HCI and we'll come on to that. And that's because Azure Arc is integrated into Azure Stack HCI. But then we've got Azure Arc. And what is Azure Arc is the ability to extend Azure management and monitoring capabilities to workloads that live with outside Azure, and also the ability to consume Azure services on-prem. So Azure Arc decouples Azure services and allows you to run them on-prem, which will be covered in depth in the next episodes. But what does that actually mean? Because they say Azure anywhere with Azure Arc, that could be in a workload running in your data center as we spoke, it could be a workload running at the edge, but it can actually also be a workload, or a virtual machine running another public cloud. And I love this. I love this approach from Microsoft and Azure. They really are looking to ensure that any workload can benefit from Azure innovation and capability. It doesn't need to be an Azure workload within the Microsoft data center, although sometimes that's still the best location, obviously for workloads. But I absolutely just love this approach. I think it just shows that Microsoft and Azure really understand their customers and their requirements. Yeah, absolutely. I think you hit some absolutely great points here. Obviously, we got to talk about the Azure Stack portfolio and I would put it next minutes anyway. But also with Azure Arc again, I see like being able to connect infrastructure in terms of servers, Kubernetes clusters, vSphere environments, Azure Stack and all that to the Azure control plane and manage everything from top from Azure, that's a great benefit. And then on the other hand, and you also mentioned that is like bringing Azure services to the places the customer need them, right? So we have for example, like one example is I spoke to a customer very recently and he said basically we love Azure SQL. Like we love that it's a service. I don't need to take care of updating. I can scale it. I can do all these great things with it. However, I have scenarios like locations where our internet connectivity is just not reliable enough or he has also one or two locations where data sovereignty kicks in where we don't have an Azure region, for example, close by in the same country. And so he needs to still store them within that country. And so what we do is if a customer cannot use the Azure service in Azure, we bring the Azure service to the customer. And I think that that's pretty great, absolutely. So let's talk about more about Azure Stack and let's switch back to the learn module. So if we go here back to my screen, if you're in the learn module and you're on the introduction page, we would now go to the next unit and actually explain what Azure Stack is. And so really what is Azure Stack? And I think if you are familiar, like for those who are familiar with Azure Stack in the past, Azure Stack was one product, right? But Lisa, that has changed now. Yes, it has. So when Azure Stack first came around, it was one product and that was Azure Stack Hub, as I like to call it, the OG, the original product within the portfolio. And it was a way to consume, again, Azure services, IS and PAS services on-prem. And we'll come on to a little bit more detail about Azure Stack Hub. Next up was Azure Stack Edge. And then like I say, the newest product is Azure Stack HCI. So yeah, there are family of products, infrastructure-based, which allow the extension of Azure to date centers and edge locations. Like I say, there's three of them in the family currently with Azure Stack HCI being the newest member to the family. And they do have some similarities, but they also have some differences, right Tom? And do we want to talk a little bit about some of the characteristics that the products share? Yes, absolutely. So you're absolutely right. And we have these, and obviously there's a reason why we have different products there because they have very specific purposes. But one thing where we come to the similarities and the characteristics, you can see here also in the learn module itself, that they all are, these are all for on-premises and edge locations, right? Meaning you cannot, this is not a multi-cloud play in that sense. You cannot put an Azure Stack, it's not a cloud provider. But for your on-prem locations, where you need some sort of infrastructure, again, for different purposes, and we will dive into these in just a bit, we will definitely have that as a great solution. They're obviously all closely integrated with Azure. And I think I would even say they are part of Azure. It's not running in the Azure data center, but they're actually part of Azure. They are delivered as an Azure service. So when you look at things like billing, for example, that is done through Azure, right? So for example, all of these services, they're usually deployed as a service as well. So you have some similarities there as well. And I think also they can benefit very much from Azure services. As you pointed out before, Azure services really, and I like that term of saying, hey, Azure really helps to also make your on-premises environment better, right? And I think that is one thing. And they also help very much to modernize your current infrastructure, right? If you currently have your virtualization environment and you're actually doing that, that's all great. But with moving to a cloud approach, you can take advantage really of some new innovation. We are driving with Azure and bring that to the on-premises world as well. They all come, as I already mentioned, I've already mentioned billing, but they come all on a paper use basis model. There are also different options. If you don't like that, there are options for certain products as Azure Stack Hub to have a different way there. But basically what our customers really like is that paper use model to actually just pay for what they use, even if it's on-prem. So that's fantastic. And then again, as I mentioned, those are really purpose-built solutions. So we have Azure Stack Edge, which has a very specific purpose when it comes to, for example, machine learning and AI at the edge. We have Azure Stack Hub, which is a great solution, especially if you are in a disconnected environment and you want to take benefit of the Azure control plane, the resource manager and no worries for all everyone watching here. We can go into that a little bit deeper. And then Azure Stack HCI, which allows you to do a couple of awesome things from very small deployments, for example, in your retail stores or branch offices, but then also very powerful deployments, for example, for your SQL server environment in your data center. And Lisa, you also mentioned obviously Arc before. I often get the question, and I know you answered that a little bit before, but I often get the question. So Thomas, what do I need? Do I need Azure Stack or do we need Azure Arc or is Azure Arc replacing Azure Stack? Or what is the deal with that? Yeah, and I think if we just think back to my slide, and now I understand Thomas, why you suggested we put the slide at this point of the discussion. I think the way to think about it is, is the Azure Stack portfolio products is infrastructure products from Microsoft, right? And they're partners. So very quickly, just to clarify, Azure Stack HCI and Azure Stack Hub, you would need to get from your chosen OEM. I remember where I work, but you might have a different chosen OEM, right? So you would get Azure Stack HCI or Azure Stack Hub from them. And, but you actually consume Azure Stack HCI and we'll come on to this from Microsoft, right? But all three of those are infrastructure. And I would say that if you are all in on Azure, you are familiar with Hyper-V, you use all the tools that you've been used to with Hyper-V and Windows Admin Center, et cetera, then maybe going with an Azure Stack product would be for you. I mentioned at the beginning that Azure Arc is actually inbuilt into Azure Stack HCI, so it just enhances it even further, giving it that really deep integration with Azure. Whereas Azure Arc, right, is there's no infrastructure kind of that comes with it. You extend Azure Arc, we mean Azure Arc agent to infrastructure in which you are already running on. So that might be something entirely different from Hyper-V. I'm not sure, we should have checked at the beginning what I can say and what I can't say, but for instance, it could be another Hypervisor provider, right, like VMware. You could have some VMware infrastructure running in your data center, but you want to take advantage of some Azure capabilities and some Azure services, then Azure Arc is for you. So there are lots of different options here, very in keeping with Microsoft, they like to provide lots of options, which is great, but it can sometimes get confusing, and I think that's what's gonna be really beneficial about this episode and the next episode is explaining a little bit more about how they complement each other, because I think that's your question, Tom, they don't replace each other, they complement each other, and then it really depends on your requirements, which route you go down or if you have a mix of both. That is true, absolutely. I also quickly wanna take, by the way, some time for the people who are commenting on the different streams on YouTube, LearnTV, Twitter, Twitch, and so on. So it's great to see, we have, by the way, viewers right now from Switzerland with Nuno, we have people from Australia joining in, we have George is here, I can see that, Patrick from the Netherlands, we have Scott from Scotland, that's good. I did not make this up, that's true. And then we also have people from South Africa and India with CFA, so, and Yutui Rimui. I hope I pronounced, I'm sure I did not pronounce it correctly, and I apologize for that, I'm sorry, but that's awesome. So thank you everyone for joining, it's great to have you here. So with your explanation now, that absolutely is helpful. So we have a knowledge check, I promised you that we actually go into this knowledge check and actually see and go through that. So with that, we want to include you in that. And so we go to the first question and we wanna give you some time as well to actually go and answer that question with us. So again, you can scan the QR code and answer it or you go to aka.ms slash polls and you can actually fill that out and actually vote for what you think is the right answer. The question is, which of the following feature is common across the free Azure Stack products? So A, it's purpose-built, pre-configured and Microsoft certified and validated hardware. B, it has built-in support for disaster recovery, or C, all of these support up to 16 nodes. So this is absolutely interesting question and we can obviously go into the different details of this answer, Lisa. There's definitely some stuff which is true for some of the products, but not for all of them. So it's really need to be very careful here. It's like really common across all the products. So let's speak maybe about the solution C. I know that like Azure Stack HCI, for example, supports and up to 16 nodes. But if you look at the others, Azure Stack Hub and Azure Stack Edge, they're different, right? Although, maybe not, but... Yeah, I think it's Azure Stack Edge that would be the one that doesn't fit this number C and we'll come on to why that is when we cover Azure Stack Edge, not to give away any spoilers, but it's, yeah. I think whereas Hub, Hub and HCI support for up to 16 nodes and then more in a cluster set. How many in a cluster set, Tom? How many in a cluster? That's interesting question. That's interesting question. How many cluster sets can you have? I don't know. I think it's, so I want to say for HCI, it's up to 16 nodes in a cluster and up to 64 in a cluster set. Okay, okay. So that is, by the way, that's interesting. So that is something we get a lot is like questions. How, like, what is the biggest size of a cluster, right? A cluster can, like as you mentioned, for Hub and HCI, we can go up to 16 nodes. But then the cluster was saying, well, but we need bigger environments. And we said, yeah, of course. You can have multiple clusters and you put them in a cluster set for easier management and put these together. So that's definitely a good thing. And then obviously another one we have is built-in support for disaster recovery. So I know for a fact here, that Hub and HCI support disaster recovery or like high availability in that sense, if you will have multiple nodes. But again, I would say, and that's my opinion, Edge doesn't really, like the Edge appliance, and we're gonna talk, as you mentioned, more about that. I don't think really that that's an option right now there that you can do. It's also not designed to do that. Like the use case is not doing that. Yeah. So this leaves us basically with A. Yes. A purpose-built pre-configured and Microsoft certified and validated hardware solution. And I think that is absolutely true across all these different Azure Stack services. So let's see if we are right here. And we can also see, by the way, there's a lot of votes here for answer A, that is good. And we see a couple of votes also in the other topics because they're not, don't get, like don't think you're wrong now. It's not, sometimes they're not easy questions again. Like two out of three usually support these. So it's, and if you're not familiar with Edge then, which you will be later on. So it's A, we were right, absolutely. So let's go to the next one. This one is a great one. What is the common pricing model supported by the Azure Stack products? And we have A, reserved instances, which is, by the way, a great way to save money in Azure. We have B, spot pricing, which is also an interesting piece, which you can do in Azure. And then we have C, pay as you use. I think that is like a common cloud, you're a billing model obviously. You would expect that one, wouldn't you? Sorry? You would expect pay as you use. That's like, yeah. Yeah, and I think that is to come back and we will have a look if that's the true answer. Pay as you use really is, as I mentioned in the cloud, a real thing, right? It's most of the time you use pay as you use. And since Azure Stack really is part of Azure, they are pay as you use. And so let's see if I was right. And again, people are voting. So there are a lot of votes. Again, as I can see on answer C, and I will see if we are all correct. Yes, so it is answer C, it's pay-by-use. Reserved instances and spot pricing are really cool things if you are running, for example, virtual machines in Azure or even other Azure services. And you commit, for example, like in the reserved instance case, you commit to using that, like let's say virtual machine or service for the next year or three years. And then we get you a discount because you committed to using it, right? So that is definitely some interesting stuff there. So with that, let's switch back to the learn module. And again, in my case, we already checked the answers. So let's go to Azure Stack Hub, Lisa. And I know you do a lot of work on Azure Stack Hub. Like in the very early days as well, you were very early involved around Azure Stack Hub. Yeah, so I won't lie, Azure Stack Hub has a special place in my heart. It was learning about Azure Stack Hub just before its launch that got me super excited and involved in all things Azure and in all things Azure Hybrid. Because I think when Azure Stack Hub was launched, it was really quite ahead of its time. So let's talk a little bit about what Azure Stack Hub is. And so the way that I explain Azure Stack Hub is it is effectively a mini Azure region within your data center. So it runs on infrastructure, which is co-engineered by Microsoft and OEM. In my case, that would be Dell Technologies. And so the co-engineer and appliance, that appliance is delivered as what we like to call a black box in that you don't really get to tinker around with the nuts and bolts and the infrastructure inside. It's really not a virtualization platform, it really is an appliance. And it runs an Azure region within your data center. Now, what do I mean when I say that? It means that you run the Azure portal, the Azure marketplace. You run a version of the Azure Resource Manager. Really everything that Azure is is packaged up into Hub and allows you to run that on premises in your data center or at the edge. It starts at four nodes. So four nodes to begin with, but when you think about that engineering accomplishment, that's quite impressive because you'll have literally Azure is starting in four nodes. That said, it only has a subset of IaaS and PaaS services, the most popular ones. And also a key thing to keep in mind with Hub is if you are going to run an Azure Stack Hub, you really do need to be familiar with the sort of cloud operator skills necessary to do that. And what do I mean by that? You will be responsible for managing the Azure marketplace. You will be responsible for managing the offers and the plans and the capacity. Basically everything that Microsoft do for Azure behind the scene, you will be responsible for. But it's really, really cool. It's for very specific use cases. So very much the use cases where connectivity is an issue or data regulatory restraints apply. And so we see a lot of it in the defense industry and also in industries where they have to run in kind of hostile environments and they really cannot be connected to the internet because with Azure Stack Hub and it's the only product here, you can be entirely air gapped, entirely disconnected from both the internet and from Azure. So it's a really, really cool product in case you didn't tell, I kind of love it. But yeah, that's kind of how I would high level describe Azure Stack Hub. What do you think, Tom? Yeah, no, yeah, I mean, I know you're right. So I don't feel gonna mess with that. So yeah, you're absolutely right. It's an integrated system, right? You get it as an appliance solution if you will. And you have obviously the hardware packaged up nicely by the vendor to support this. And then you get this possibility to run basically really Azure in a disconnected environment, right? But again, I also wanna manage expectations here. There's obviously some IaaS and some PaaS running on this, but you can't imagine, you mentioned the engineering effort which went from like, hey, going from this super large Azure data center where we have thousands of servers to a four node version is pretty big. So obviously there are some limitations with that as well. I always need to tell customers to like, hey, don't expect like everything you get in Azure, you get in Azure Stack Hub, right? It's really for certain services, you need to check out if that really is a good fit for you. But again, then it provides you with a disconnected portal. It provides you with our Azure Resource Manager so you can use basically ARM templates which you can use in Azure. You can use them also in a disconnected environment. I think that is really, really powerful as well. So you mentioned a little bit about the use cases and I think that is interesting. Again, we spoke about like, hey, that's gonna be a disconnected scenario and scenario where we don't have connectivity or direct connectivity to Azure but you still wanna take advantage of this. And I absolutely agree. I love the Azure Stack Hub solution on this as well. But there are also some things we maybe need to talk about what is like, and you mentioned already a couple of these what Azure Stack Hub is not, right? I think if you just wanna replace your virtualization environment Azure Stack Hub is probably not necessarily the right solution for you, right? There is definitely with, for example, Azure Stack HCI probably better solutions out there which you can help with that. Or if you wanna customize things really like on the different hardware pieces or you wanna do some special settings and stuff like that. That's because it's delivered as a appliance or as an integrated system, there's not really possible. And that's also a good thing, right? It's like a good thing that people cannot temper with the environment. And that is like sometimes really want that. But like if you want it, probably not the right choice. No, I think if you, and I've seen this, right? Especially in the early days. And I've also seen this, I'll just say from some other vendors, when you start to try and customize Azure Stack Hub, you're gonna run into problems. You probably don't really understand its use case very well and you will very quickly run into problems. The reason that it is co-engineered by Microsoft and OEMs, the reason that it's delivered as an appliance is because you really need the infrastructure to be consistent and predictable for lifecycle management and also for bringing Azure services to it. It's effectively what Microsoft do with their infrastructure in their data centers, right? Is they have to be quite prescriptive about what it looks like and how it runs so that they know when they deploy new services to it. It's all going to work nicely. Yeah. And you're hitting an absolute good point because that's also one thing listed. Obviously we, this is packaged up in an integrated system and we make it easy to apply updates to it. But you're still, the hardware is still in your data center. You're still responsible for actually, like someone needs to manage that hardware still, right? It's different from what you need to do, but if something breaks, the Microsoft person coming into your data center and pulling the disk out and putting another one in, I know that vendors in some cases and some managed service providers offer these things. But again, it's not managed directly by Microsoft. It's not something where Microsoft is actually going on you to your systems and does all of it. That's also something customers in some cases don't want, right? They want to really have that disconnected scenario and they don't want to have Microsoft being able to connect to your solution, right? And so that is definitely a good case here. There's both infrastructure and software updates required. So they're definitely not eliminated. There's the Microsoft updates and then there's also the hardware updates for your OEM. You need to make sure that they are carried out. You need to make sure those updates, those updates are happening like you say you need to make sure that you are responsible for that underlying hardware. And I think as Azure Stack Hub has now been around for a while, we've got really clear on what it's not and what it is and that I'm seeing that in terms of the use cases when someone comes and they're like, I think this is an Azure Stack Hub use case. More often than not, it is. And for those where it's maybe not the quite fit, we obviously have another two products that we'll come on to in a little bit. Absolutely. I also got a question in the chat here. Someone is asking, can I say Azure Stack Hub is a cloud extension? Now, I would say, well, depending on what you mean by extension, but I would say all of the Azure Stack products are cloud extensions, right? It's like all of them bring some sort of Azure into your data center. And so for that, I would say, yes, you can say that, but you need obviously again, people can understand different things from what an extension is. So you need to be clear there, okay, what it is and what it not is. So we talked about the use cases already, but we have a couple again, like also the learn module here, as you can see highlights a couple of these use cases, latency and bandwidth, sensitive workloads, where you actually wanna have like these services running on-prem where you probably cannot deal with it, like completely disconnected environments or as you mentioned, compliance regulatory restrictions, which are interesting as well. So next up. Before we go on to the knowledge check, because I have scrolled ahead and I want to make sure that we're giving everyone the chance to get the questions right. Will we just quickly mention what some of the past services that are available on Azure Stack Hub? So like we said at the beginning, there is a subset of IaaS and PaaS services available on Hub, and it is a subset for reasons that we've described. It does start as little as four nodes, but these include virtual machines. So that is Azure virtual machines by the way, which is pretty cool. You've also got app service web apps, API apps, you've got Azure functions, you've got SQL, MySQL, you've got Event Hubs, Key Vault, IoT Hubs, and a few others there as well. So service fabric clusters and Kubernetes clusters. So just to give you an idea of some of the past services that you can run on Hub, which may or may not be coming up any question. Awesome. No, thank you for doing that. That's definitely gonna be helpful for our audience to answer this question. So let's go into the next knowledge check and I'm gonna share again, going back to the deck here. So we have a question around which of the following platform as a service services, so past services, does Azure Stack Hub support? So does it support A, Key Vault, B, Cosmos DB, and C, Traffic Manager? So very important now, I hope you listened to Lisa in the last couple of minutes because she actually went through it. And so there's definitely some interesting stuff here. And I think what we need to consider here, and I think that makes the answer at one point, if you understand these services and you understand the concept I mentioned with like going from these large Azure data center to these even four node systems, right? Four to 16 node systems. There's obviously a lot of work going on. And if I look at these services, I think C and B. So C is definitely like, if you're familiar with Traffic Manager, it's really a global service on Azure, right? To actually have traffic distributed across different Azure regions. So that is pretty big already, and obviously globally distributed. If I look at Cosmos DB, similar thing, basically it's also service which scales very well and also can be like replicating across different region. And it's pretty like, if you look at it, how it works, I would say if you're a little bit thinking how it works, it's a pretty big service, right? So, and then we have Key Vault. And for those who don't know what Key Vault is, it actually is a secure place where you can store your secrets and certificates in a secure way. We have Azure Key Vault obviously in Azure, but is it also available on Azure Stackup? That's the question. So we've got some votes coming in, I can see on the poll. And it's leaning heavily towards a Tom. Okay, let's see. Again, you can still vote, but we will already go ahead and see if they're right. And yes, it's correct. It's Azure Key Vault. Again, if you listen to Lisa, then you would have known that. So make sure you're aware of that. Now, does that mean, by the way, I always get this question, I cannot use this. Now, I cannot use Cosmos DB. It's obviously not running on Azure Stack Hub, right? But I can still build hybrid applications, meaning I can still take advantage of these services where some stuff is running on Azure Stack Hub and other stuff is running in Azure. And then I can use, for the part that I'm running in Azure, I can use Cosmos DB, for example. So that's definitely something. So let's jump to the next question here, which is, which of the following is the best, is the benefit of Azure Stack Hub? So A, no need for infrastructure updates. B, a highly customizable infrastructure. And C, support for the Azure Resource Manager deployment model. Now, Lisa, I think we talked about the Azure Resource Manager a little bit, but maybe you can also shed some lights why this is actually a great tool, why you would use that, and why, what can it actually do, like? Yeah, so the Azure Resource Manager, or ARM as it's known, I think it's been referred to quite often as the secret sauce of Azure, hasn't it? And it actually goes back to what you just mentioned. So on the previous question, you said sometimes if something isn't listed as available on Azure Stack Hub, does that mean that you can't use it? And again, it depends on how you're using Azure Stack Hub. So if you're using it as a completely disconnected scenario from Azure, but if you're using it as part of a hybrid scenario, like Thomas said, you could build hybrid applications, right? And the key to being able to build hybrid applications and having them consistently run across Hub and in Azure is Azure Resource Manager, right? And I've seen some really cool stories from software vendors, so ISVs who have built applications in Azure, use Azure Resource Manager to build those, and then they've got customers who want to consume their applications, but they're not near an Azure data center or region, or again, they have data regulatory and privacy issues. And what they can basically do is because they've used Azure and they've used ARM to create this application and architect it in that way, they can pretty much take that and run it on Azure Stack Hub, which is just very cool. So I would say that C is looking good for me and definitely leaning towards C on the votes coming through that I can see. Yep. So this is absolutely, quickly also, like I like the magic behind Azure, and like Azure Resource Manager really is the magic, because if you're interacting with Azure even through the portal or CLI or PowerShell or APIs, you're actually interacting with the Resource Manager, and actually that's responsible for doing all the subscription management for all the resource grouping, tagging, role-based access control, and so on, and also deployments. So for example, as mentioned, for example, like infrastructure as code deployments as well, but everything actually which is deployed, I can go through that. And though that's super handy if you want to have these consistent deployments. Now, one thing I really want to highlight again, I mean, there's definitely a need for infrastructure updates, right? There's definitely, you need to update your infrastructure. However, we make it easy by packaging these updates. So I don't think it's A, because everything needs to be updated. Yes. B, it's highly customizable. And I think that is also important. It's an integrated system. So it's definitely not something you want to go out and customize or can customize. We have, again, with Azure Stack HCI and all the solution there. So I agree with you. I think I would also lean towards C. And guess what, it's definitely the right answer. Excellent. Perfect. So let's switch back to the learn module and let's go on to the next section. That is Azure Stack HCI and I'm super excited about this one. Me too. So by the way, Lisa, what does the HCI stand for in Azure Stack HCI? It stands for Hyper Converged Infrastructure. Yeah, yeah. So for everyone who doesn't like it, it's like when you heard the term the first time, if you come from classic virtualization scenarios where you probably had a send storage, right? And you had your compute notes and all that. You probably haven't necessarily heard about HCI, but like Hyper Converged Infrastructure really brings like storage compute and networking all together in this solution as well. Do you know the other day I learned, so I thought, so if you go from sort of the three separate compute storage networking and then you move to Converged Infrastructure, I always thought in my head that that was the compute and storage converged, but actually it was the compute and networking that was converged. And then Hyper Converged Infrastructure is all three. And I only learned that the other day. I was, one of my colleagues is very kindly taking me through, yeah, the sort of history of hardware and infrastructure. You can see that I need to get out more. But I've been enjoying that by learning that the other day. So yeah, it's bringing compute storage and networking all together and converging that into one infrastructure, which has lots of benefits. It's not always the right thing to do, depending again on your workloads and your use cases. But if you're trying to, if your workloads are okay to run on a Hyper Converged Infrastructure and you're wanting to reduce your footprint, reduce your attack vector, for instance, HCI can be the way to go. Okay, so that's awesome. Thank you for that. I mean, it's always good to have someone from a OEM and hardware vendor helping us with these kinds of things. So if you look on Azure Stack HCI is, what would you, how would you explain it? Yes, okay, so Azure Stack HCI is a new purpose-built operating system for Hyper Converged Infrastructure. And it's delivered as an Azure service, which is pretty cool in itself because with that comes a whole bunch of benefits, right? For instance, you can wave goodbye to being out of date because as it's a cloud service, as you would imagine and expect from a cloud service, it's always up to date. So you can wave goodbye to having to run out of support of your operating system and having to upgrade it. Doesn't mean there aren't updates. Everything needs an update, everything still needs an update. But that's what Azure Stack HCI, it's a new purpose-built operating system for Hyper Converged Infrastructure, like a stripped-down version of Windows Server really. And then again, it runs on Windows-certified or validated hardware from an OEM. And there are two routes that OEMs could go down with Azure Stack HCI, and that was to provide a validated node and or an integrated system. So at Dell, we went with an integrated system because that means that we can provide full solution level support, we can apply our life cycle management guidance, et cetera. And I think the integrated system really allows customers to focus on actually the Azure innovation on-prem, you know, and less on keeping the hardware kind of up to date and they can focus on all the exciting stuff, shall we say. But that's what Azure Stack HCI is. Of course it has Azure Arc integration, and I'm sure we'll come on to a bit more about that as we go through. Yeah, absolutely. And I will even, we will even show you a little bit of like how that actually looks like in a very quick demo as well. But I also interestingly find it like interesting, okay, there is, as you mentioned, there's validated servers and it can do like different than from Azure Stack Hub, it can already start at very small two node clusters, right? Up to like 16 nodes. So that is pretty cool. So even if I have a very small branch office, I can take some. I also saw some vendors like Dell and other OEMs having these very small solutions like bundled up, which work really great. We also see rocketized solutions out there for people in the field as well. So there's some pretty cool stuff going on. I think a very important part is, as you highlighted, it's a limited Azure service and that obviously has some benefits and it's tightly integrated to Azure as well. So if we go down a little bit, we have here a graphic which shows a little bit how that actually looks like, right? And I think that is kind of like what you also explained very, very well here that on the bottom you can see we have the validated hardware. By the way, so the question now probably comes up like what is the validated hardware? How do I find out? How do I get this hardware, right? And I have existing hardware maybe. So how do I, can I leverage that as well? And so we have the Azure Stack HCI catalog where you can find the different hardware solutions from OEMs and the hardware vendors and our partners there. And then we have right now, I think over 400 solutions out there, like 400, right? So it's not a very, very specialized thing. So the chances are high that your standard servers are actually in that catalog. So it could be that if you already own servers that they could probably already work with this, but then obviously you can also go out and check out this catalog to buy new hardware. And the great thing also, by the way, next thing is even if you try this out, if you then want to install Windows Server on it, later on, if you're like tried it out and you think, well, it's not the right approach, you can still go out and use it for something different, right? It's not like a closed system in that sense. And then as Lisa said, Azure Stack HCI really is the operating system powering this hyperconverged infrastructure with technologies like Hyper-V, Storage Spaces Direct and software defined networking. And we take obviously the things we have from Azure, all our learnings from Azure software, from the Azure hypervisor and we're putting that in and taking that into Azure Stack HCI OS. Now there's a little bit of a difference there obviously because in Azure we exactly know how the hardware exactly looks like and we only support a certain type and we have a really purpose-built systems there. But obviously in the Azure Stack HCI site with these 400 solutions, we need to open it up a little bit, we need to change the management a little bit so we give you a little bit more on that sense. And then on top of this, and that becomes really like the interesting part, you can obviously run high virtualization workloads, Windows and Linux VM, super high performance if you need to. But then you can also run, for example, the Azure Kubernetes service on top of it, you can see that. So you can obviously create a Kubernetes cluster like of any vendor, you can have OpenShift and others as well and actually run that like because you can create VMs, you can also install things like that. But we even provide you with the Azure Kubernetes service on Azure Stack HCI. So you get a service, which you can run a Kubernetes cluster as a service, if you will, in your own data center to deploy, first of all, your containerized workloads, which you created by yourself. But then also, and we'll talk about that later in the other modules about arch-enabled services. So these services from Azure, which we talked about, which can come into your data center, what they need is a Kubernetes cluster and we support different ones, but if you wanna have a full stack Microsoft solution, you can go with Azure Stack HCI and AKS, like the Azure Kubernetes service on Azure Stack HCI and then deploy Azure Arch-enabled services on top of it. For the management side. I think this is the coolest part, right? Azure Stack HCI, the fact that you can have your traditional virtual machines running alongside Azure PaaS services, right? And I love the fact that the AKS Azure Kubernetes service is, you can only get it on Azure Stack HCI, right? And I just think that, I think that this is very, very cool. So it's more customizable, like Tom said, than Hub, therefore it can start smaller than Hub, so it starts at two rather than four and therefore addresses a bit of a wider, and we'll come onto this, a wider range of use cases. But I think this is just Azure Stack HCI really is the perfect blend of hybrid, really taking care of those on-prem work codes but allowing you to enhance them and make them better with Azure capability, but then also allowing you that deep integration into Azure and the ability to run Azure services. And I'll let Tom carry on because he's going to talk now about the management options, which again, builds on that. Yep, absolutely, thank you. No, again, I agree with you completely on this. So yeah, we talk about management and I also want to take time to answer a couple of questions in the chat, which are super interesting. Thank you, keep them coming. We will, I will go through them really quickly in just a bit. So obviously if you have a solution on-prem in that hybrid mode, you can use Windows Admin Center to manage that and do all of your management tasks and do to set it up and all that. However, you can also take advantage of Azure. As we said, it's an Azure Connected Service. So you can also manage certain things through Azure as well. And I will go, I'm going to show you that in just a bit just so you get a little bit of a look and feel. So let's have a look at the chat here. One of the questions is so, and that I find very interesting, is there any lab or development kit for Azure Stack HCI available? Now, if you're familiar with Azure Stack Hub, you probably know that there is a development kit for Azure Stack Hub, but you can download in some one note to try these out. On the Azure Stack HCI side, you can download the Azure Stack HCI operating system and use that and there's an evaluation period. I think it's 30 to 60 days, Lisa, you correct me, which you can actually use it for free and you can install it and you can manage it through that and try it out. So there's no special version as soon as, as long as you don't activate it in that sense, you can start using it and try it out the full OS and actually install it. In fact, I downloaded, I have like two old servers which I can basically install it on and then try it out and clustering it and try it out how it works. I also got an interesting question. So, on the labs front, really useful resource that my teammate, Jeremy, or Casper has created is he's created some labs for Azure Stack HCI. So for testing out some various scenarios, so helping you with deploying Azure Stack HCI, testing some of the functionality, et cetera. So I've just popped the link in there that Laurent is going to share and there's an MS Labs, you'll find it on GitHub. But then also my team has gone ahead and added some of the Dell functionality into some labs. And so I'll pop the link in for that as well, just if you wanted to test some additional labs that the team has gone ahead and created. But I would say that the MS Lab on GitHub resource is as great that Jeremy has created. So I just wanted to give a shout out to him there. Awesome. So yeah, then the other question I have here and that's a very good one is, so what are the limits for Azure Stack HCI when you are disconnected, right? So there is definitely, so Azure Stack HCI, you can run it in a way disconnected. However, you need to like every 30 days, you basically need to connect to like basically send out billing information because it's again, built through Azure. As I mentioned, it's like an Azure service. So we need to know how much you consume. So it needs to connect every 30 days. And yes, so if you're using like, so if that's okay for you, that's great. So we have a lot of customers using it that way. The limitations basically is if you don't connect after a while, we will not shut down anything, but we will like prevent you from creating new services on top of it if you are not connected, right? If you don't do that in 30 days, right? But if you connect like every 30 days once, and it's again, very small data which goes out there, you basically can use it even, I would say in a semi disconnected environment. But again, it's not designed to be like in the air kept environment as of today. So that's for sure. I get this question a lot is like, okay, what if I'm not connected, it's just not reliable. That's fine. I mean, if you can fix the internet somewhere in like 30 days, then you're absolutely fine. And again, we will not shut down your services if they're running. And just to build on the consumption piece. So we've said that it's delivered as an Azure service. We've said that you need to connect once 30 days to send back billion information. How is it built? It's built per core, right? Yeah. So it's built per physical core of your machines. And then that basically of your physical, obviously this physical course would be physical machines. But that is what you pay for how billing is done. And again, you have a evaluation period. I think it's 30 or 60 days, we need to check on that. And so you can actually use it without having any charge on it. So if you just want to try it out, it's super easy to do so. So before we actually go into, and there's a lot more in the learn module, especially I wanna go to the use cases, but before I actually do that, I wanna quickly show you how this actually looks like if you look at Windows Admin Center, for example. So here I am in a demo environment I created and this is Windows Admin Center. This is basically running in a web browser. I installed that on a management machine. And you can see here, this is my overview page about my Azure Stack HCI cluster. So you can see here that what it's doing that everything is green, luckily. I had that a happy like how much storage did I use of that, of that volume, what is going on? So you can see this is a pretty fresh one of that. We also have on the left side here is the menu. So let me quickly look at this. So you have, for example, here the possibility to manage virtual machines, your physical servers. So if I click on this, you can see here, I get a little bit of a nice overview of what's happening on that machine. I can also get the inventory. So you can see here, these are my Azure Stack notes. And then you can obviously manage all the rest of the hardware like volumes, drives and so on. You even have storage replicas or you can even have stretched clusters from two locations as well with Azure Stack HCI. We have the software defined networking built in and other networking stuff updating. And even you can connect Azure Monitor. And I will show you that in just a bit where you can actually go out and like have like the monitoring piece in Azure for this, right? You're still here have of course the diagnostic locks and so on performance monitor where you can go out and check the performance of your system. And I also want to quickly highlight this. This one is really cool. So if you have, if you want to run now your Kubernetes cluster, you can really take advantage of AKS like the Azure Kubernetes service on Azure. And Windows Admin Center will help you to set this up basically as a wizard when you go actually through and you deploy it on that. So you have your, you can really quickly set up your Kubernetes cluster. Now, let's see what we can actually do when we said it's an Azure service. So if we go to Azure and we go to our Arc page here we can actually see here what you can do manage with Arc and again, we will speak about that in other modules. But if I go down here to Azure Stack HCI you can see here my Azure Stack HCI clusters showing up. So I have one cluster here and one this is the cluster I showed you. It shows connected. So if I click on this, you can see here it now looks like an Azure resource, right? It's part of a resource group. It's part of a subscription. You can also see what I'm running here. So you have different operating system builds. On the bottom you can see the nodes. You have the monitoring piece here. So you can see here that everything is healthy and so on. So I can like, even if that, if I don't have direct connection to that Azure Stack HCI cluster if that for example would be like another data center where I don't have network access with a VPN or anything I could now go to Azure and actually have to look at that system. And then I can also like enable log analytics and much, much more I can take advantage of role-based access control. And as Lisa mentioned, what we can do now here is also we can, this is a preview feature which we just announced. You can now manage the virtual machines on top of Azure Stack HCI directly from the Azure portal. So you can go out and create VMs. You can manage them. You can create new virtual networks. You can manage the disks, the VM images. Again, this is all in preview right now. So this is done by the Azure Arc Resource Bridge. Super cool solution out there which you can actually use and manage. Also on the configuration, I also wanna show that there's some stuff you can do. So first of all, the Windows Server subscription add-on which basically allows you to get guest licenses for Windows Server on your Azure Stack HCI system. Before that you would need to buy Windows Server licenses and license your physical environment. With that you can get it from there. You also can see here on the bottom Azure Arc-enabled benefits. So meaning, sorry, Azure benefits. So things like host accession. So you get, for example, that if you move a VM, for example, to Azure Stack HCI, it will be treated as an Azure VM. So for example, you get a range of extended security, updates and so on. So if you wanna learn more about that, you can turn this on here and you can learn more directly on our docs pages as well. So that is pretty cool to me. Now think about it. Like if you have one cluster, well you'd say probably Windows Admin Center is a great place to manage. But if you now have like, like let's say 50 clusters or even three clusters and they're all placed around the world and you're sitting in a coffee shop somewhere and you need to create a new VM, you basically just log in, securely log in to the Azure portal and you can go out and create your VMs on your Azure Stack HCI running in your own data center directly from the Azure portal. How cool is that? That is very cool. I think infrastructure operators will be now thinking, how can you sit in, I want to sit in a coffee shop and be able to manage all my clusters around the world. That's super cool. And one thing that you mentioned that actually, we didn't highlight at the beginning is the stretch clustering capability. So Azure Stack HCI was launched with the stretch clustering capability. And that is specific to Azure Stack HCI. And another thing to sort of keep in mind with this, we said it's delivered as an Azure service, as a Cloud service. And so it will be updated with new features at a more frequent pace than maybe previously operating systems were, right? Yeah, yeah. So by the way, I also want to quickly go to another question from the chat, there are a ton of them. And I apologize if we cannot answer all of them. But one, I see, can I extend storage by adding additional storage spaces, direct nodes? That's the technology behind Azure Stack HCI and Azure Stack Hub when it comes to storage by extending these. And the answer is yes. So what you can do is actually you can like, if you have a two-node or three-node cluster on Azure Stack HCI, for example, four-node cluster, you can add more nodes as soon as you need more resources, right? If you need more compute resources, you can add it. If you need more storage resources, you can add it. In some cases, depending on your setup, you can just also, if you just need more storage, you can add more disks depending on how your hardware setup is and how the support is for that. But yes, you can absolutely extend this. It's absolutely not that you buy it once, one size and then you're just stuck with that. You could really start small and then extend it if you need to. Just a quick one. When it comes to Azure Stack HCI, if you are going to need to expand in the future, start with three nodes. Don't start with two nodes. If you think you're going to be going to any text band storage, that's an important one. So what there also is, there's some common use cases, obviously. Now, this is interesting. Okay, what do we actually need Azure Stack HCI for? Now we talked about how awesome, like some of the features are. Again, we can run virtual machines on it. We can run Kubernetes on it. We can run Azure Arc enabled services on it. So here are some of them. So again, like one of them I like very much and I see that a lot is like branch office and edge locations. Because we can start from a very small deployment, we can actually go out and deploy this in these locations, right? So you can have a very small two node cluster if you just need some print service, some application server at the location. Again, I also spoke about retail stores and factories. You can do that. And the great thing is also like clusters, usually they need a witness option, right? So they need to like decide, especially if you're a two node cluster or even like just an even amount of nodes. So one thing we added is also the possibility of a cloud witness or a USB drive as a actual witness, right? So that is pretty cool. So again, to support these small deployments to make sure that this is. Now, Lisa, do you want to talk a little bit about VDI? So VDI is a pretty cool use case, especially since you can now get Azure Virtual Desktop on Azure Stack HCI. And we've actually seen a lot of this come up for customers. So some customers who are actually using Azure Virtual Desktop, and maybe it's okay for some of their workers, but actually in some locations, they need that to be running on-prem. And they can do that now because you can now get Azure Virtual Desktop on Azure Stack HCI. So I think the two sort of past services right now that are specific to Azure Stack HCI are the Azure Kubernetes Service and the Azure Virtual Desktop. So yeah, that's definitely one that we are seeing more and more of. Yeah, no, that's fantastic. Again, VDI is definitely a big one, especially because people really like the Azure Virtual Desktop solution in Azure. But then they obviously have scenarios where, again, they need to have it on-prem, right? Because of latency reasons or connectivity reasons. And with that, you can really have these pools of these VDI machines also running that. And the great thing, that obviously supports Windows 10 and Windows 11 multi-session nodes that is only supported in Azure or on Azure Stack HCI. So that is pretty cool. The other thing is obviously, as I mentioned, we spoke about small deployments, but we also have like, if you have these SQL servers, we have a couple of customers which really don't really large, really big SQL server environments, and they need to obviously have the optimum on performance. And Azure Stack HCI can really deliver on that as well. We also have things when it comes to security, with virtualization-based security, which we can actually put machines in virtual secure mode to actually protect from attacks in that phase. So that is definitely interesting. And then if you go down more, obviously if you need some sort of disaster recovery for these environments where you actually have a stretched cluster deployment, that Lisa mentioned that also before, we will have that as well. Now, one thing I really want to highlight, which I think is missing on this list, is now if you want to run arch-enabled services on it. And again, if you're not familiar with arch-enabled services, this is basically just the way you can run Azure services on premises, right? So that capability, what you need there to run this is a Kubernetes cluster. So with Azure Stack HCI, AKS and Azure Stack HCI, you can now run, for example, Azure SQL managed instance, Azure Postgres or Azure App Services, which are currently previewed, such as, for example, web apps, logic apps, API management functions in your own premises environment, which I think is pretty awesome to do that because again, there are needs for this and you can now build these hybrid architectures. So if you have an application which wants to run on past services, but the application needs to run in the cloud and Azure, but also on-prem, with Azure arch-enabled services, you can now build these architectures anyway and use that the same architecture and not depending on the location where you're running. So that is pretty cool. I definitely, we need to add this to this list as well. And I think you could bundle quite a lot of these into sort of a category that we're seeing and referring to do is just data center, consolidation and modernization. So consolidating hardware down into that hyper-converged infrastructure for the workloads that need to continue to run there, modernizing it to modern infrastructure, but then also modernizing your applications that run on it. Like you say, taking advantage of Azure data services on-prem. Very cool. Right, knowledge check, because I think we've, we could speak all day, Tom. But if we're Azure Stack HCI and Azure Stack Edge, sorry, and summary. There is definitely so much to talk about about Azure Stack HCI and ARC and all that together. So definitely join out our all next episodes. We will give you the link in just a bit at the end, but in terms of time, let's go through the knowledge check. And then we also have a very, very quick look at Azure Stack Edge. So our fifth question today is, which of the following workloads does Azure Stack HCI support? A, Hyper-V, VMs running Linux? Well, okay. B, Azure Key Vault or C, Azure Virtual Machine Scale Sets. So all interesting options in that sense. I would say, I mean, I mentioned it a couple of times where you can run. But interestingly, so C and B, those are more services which are running in Azure, which are in that sense, more built into the Azure Hub environment, right? Especially the Key Vault piece. We had this question before. So if you wanna run Azure Key Vault today on-prem, we would need to have Azure Stack Hub. So let's vote. Let's see what the votes are. The votes are coming in and I see already that goes into the right direction. I think, I know the answer. Do you, Lisa? I hope I know the answer. All right, I need to double check. But no, of course. So for everyone, again, happy voting. Just vote and go on. But let's have a look at it. And obviously it's answer A. So you can obviously run Linux VMs on top of Azure Stack HCI for sure. You can also run Windows VMs obviously as well. So the next one is, which of the following technologies is mandatory in Azure Stack HCI? Now, we quickly went through this. I quickly talked about this one specifically. So is it A, the network controller, which does software-defined networking? Is it B, Azure IoT Hub? Which frankly, this is kind of like a trick answer, maybe, and then C, storage spaces direct. So I think I know which one it is. I think, so A, you're saying network controller for software-defined networking. Let's touch on that for a second. I believe that there is an option to have software-defined networking running in Azure Stack HCI, but it's not mandatory. So if you, if this is taking me back to my exam days, you know when you had to very specific, you had to read the question properly. So software-defined networking in Azure Stack HCI is optional, which means the controller would be optional. So I'm going to eliminate A based on that. Azure IoT Hub, I have in my head as it's more of a service. So not necessarily a technology which would be mandatory, but I get your point on the trick question. So I think it's C. Yes, I think so too. Let's see if we are both correct. And also the votes also say that very clearly for the people who have voted. People said storage space directly. Yes, that's absolutely. That's the technology which built this hyperconverged infrastructure solution in terms of the storage side that actually helps us to replicate data between those nodes. So that's fantastic. Good, let's jump back to the module and the next up, and we go very, very quickly to this one, is Azure Stack Edge. So what is Azure Stack Edge? And again, this is probably interesting, but and it's a very, as we mentioned, there's a very specific use cases for Azure Stack Edge, right? And the way Azure Stack Edge is delivered, it's a first-party appliance. You can actually go today to the Azure portal and order one of these and they will be shipped to you. Again, depending where you are in the world and will directly ship to you and you can get this and it's basically one server unit. And it's designed to run machine learning and artificial intelligence workloads at the edge. So if you need, we have some capabilities to run machine learning at the edge. And so that is really where it comes in. And we have a couple of interesting customers there which help us with this because it also has, for example, GPUs and FPGAs integrated, that's our field programmable gateway arrays, which really help with these machine learning workloads, right? That is not something you usually get in a normal server. It's really something where what you get with these. And then again, if you don't need them anymore because it's delivered, you just send them back to us basically or basically we pick them up or we let someone pick it up and you just pay for a certain amount of time you're using it. So that is pretty cool. And you manage Azure Stack Edge from the Azure portal, right? Yes, that's a very good point. I'm happy you mentioned that. Yeah, so there's no necessarily, like you have a local setup thing to actually like do the first time setup and connect it through your network. But then after that, you're actually going to manage that cold appliance through the Azure portal and you do all the settings over there. So what are the use cases for this? As we already mentioned, or what are the things here? So again, it's a physical appliance as we said for really for machine learning and AI workloads. And you can run a couple of other things on top of it too, if you need to. But again, while offering these GPUs and FGPAs capabilities that really makes it a good solution for that. And then as you said, manage through the Edge location. So that means it's not necessarily designed to be in a full disconnected mode, right? It's really more connected mode, but why would you then use it? And I think the answer there is pretty simple because think about if you're in a factory and you run some machine learning over your factory floor to see if something is not right. If you do the machine learning stuff in Azure, so you need to send the picture data and everything to Azure, which takes the time depending on your internet connectivity. It will be calculated or computed in Azure. And then you get the information back and then it will take an action. So for example, if you wanna stop something immediately, it goes to Azure and goes back. And that's obviously depending on what you're doing already too much time. So we're running these Edge appliances, you can actually do it directly where you need to and it's super fast and you can take advantage of that and it can detect, like for example, what we have is like picture detection of certain factory floors or whatnot and then deciding like with machine learning, hey, something is going to not the right direction or there's now something different coming in this factory than it was before. So we need to switch a couple of things, right? So that's pretty cool. But again, it's not designed to be fully disconnected. So here we have an example also of how that could be. Again, it's like on the left side, you see these on-premises locations. For example, as I mentioned factories and so on. And then you have Azure like Edge and that can do all the local compute there and get hardware installations using these GPUs and FPGAs. It also has some super fast local storage and then due to all this processing, but then obviously since it's connected, it can also take advantage and send this data to Azure for long-term, right? Or do some additional stuff with it, which is not time critical in that sense. So I think it's a pretty cool solution. What do you think, Lisa? Yes, I agree, pretty cool. Similar to Hub in that it's got very specific use cases, but different in that it is the only one that you buy directly from Microsoft. And so, yeah, I agree. I think we need to jump to the knowledge check now and then move on to the summarizing all of them and see if we can finish on time. If not, we'll send you to the Larry module and you can tweet me and Tom afterwards and let us know if you've got the answers right. Absolutely. So what are the functional and sufficient processing you can provide in Azure Stack Edge? Is it A, Accelerated Artificial Intelligence Interferencing? B, is it administrating monitoring multiple Azure Stack Edge appliances? Is it C, Graphical Interface to an Azure Stack HCI appliance? So again, you can vote here. The question for me, again, is very simple. I mentioned that like the factory floor, right? From the vision processing. And obviously I mentioned the AI and the machine learning we can use at the edge. So in my opinion, it's definitely gonna be A. Yes, yes. Because the question specifically asks about the vision processing unit. So it's not B and it's definitely not C. So yep, A. And then last, the next one on Azure Stack Edge is which technology in Azure Stack Edge is used to implement the IoT Edge modules? So if you pay close attention, you will probably get this right. So we have A, Service Fabric Microservices. B, Azure Container Instances. Or C, a managed Kubernetes environment. Now again, I did not mention that, but you have seen it on the graphic I showed you and was also mentioned a couple of times in the learn module. So what do you think, Lisa? Did you pay attention to the graphics? So I think that it's C, but I am not 100% sure. This is the one I was a bit like, oh, yeah. But I think it's C. Yeah, it could be like, it's a fair question. The answers are also, in theory, all of these are possible, I would say, but it definitely is C. It's the managed Kubernetes environment. So we went through this. Again, let's go through the next page here really quickly and do basically the last pieces to actually compare these solutions and do a quick summary of what we learned today. So again, I wanna quickly highlight very high level that we obviously have different solutions on Azure Stack Hub, Azure Stack HCI and Azure Stack Edge. And they're really purpose built depending on different use cases. And you have heard these different use cases today. We went through this. So really it's a portfolio of products here which you can leverage. Yeah. And we've mentioned some of the similarities and some of the differences that we've gone through. So think about the number of nodes that we've talked about that you can start with for each of them. Think about the use cases. The integrated system approach versus having the flexibility to customize it and have different solutions with Azure Stack HCI. So that's definitely interesting. And also you can see here on the hardware side, that's also interesting. Azure Stack Hub and Azure Stack HCI is really like the hardware by the OEM when you're ordering Azure Stack HCI Edge to be delivered by Microsoft. And you can also see Azure Stack Edge right now supports one node. That's also why at the beginning, some of the questions were, do you have high availability and stuff like that in all of these products? With Azure Stack Edge, you have like one node deployment. Yeah. And then again, there's a huge list of stuff we could go through and you can also see like GPU support. And we had a question earlier about GPU support from someone which we did not have time to answer, but I knew it would come up here. So yes, all of the products support GPUs. So if you're gonna get like hardware, make sure if you need this, check with your hardware render your OEM to make sure that if you need it that you get the right support there for your hardware. Check five up from GPU support, Tom, just to help people out on the final question. Sorry? So one thing we didn't talk about, so just under local Azure Resource Manager control plane, it also mentions multi-tenancy support. Yeah, yeah. And that's something that we didn't mention in HUB. Yes. Let's go up. Absolutely, yeah. Absolutely. I think that's a, you're very conscious of our audience. So let's go back to the actual... I want to set everyone up for success. No, that's awesome though, really. Sometimes I'm a little bit too fast on certain things and you have to be absolutely there to do, to help our audience here. Okay, last question and we're gonna wrap up quickly. Let's go with what is the minimum number of nodes in the Azure Stack HCI? Is it two nodes? Is it one node or is it four nodes? And again, we had different solutions with Azure Stack HUB as well and we talked about different numbers of nodes. So I hope you paid attention when it came to Azure Stack HCI. So I will give some time to vote for everyone. Again, this is the last question we do today. If you want to do the other questions, you can do them by yourself directly on the learn module and you can see, go through these in just a bit. Yeah. So we see some nodes coming in. Yeah, and everyone leaning towards A2, that is correct. Awesome, thank you. So let's quickly go back to the page here and actually have the summary. Again, you get that you can go through but you can also, why I want to show this, you get some follow-up links here as well. So you can actually go, oops, can go out and do these things as well. So be conscious that you go through the learn module. Again, you will learn a lot as usual. And then I want to wrap it up here. I want to show you a couple of things in the summary. We can have a quick look. So Lisa, what did we learn today? So I would say that we definitely are now all able to describe the Azure Stack portfolio and the use cases of each of the products. And we definitely are able to spot the differences and the similarities between each of the three products in the portfolio. No, that's awesome. Again, thank you very much, Lisa, by the way, for joining today. I also want to quickly highlight the Azure Arc Jumpstart project. Super interesting. If you want to try out Azure Arc, we talked a little bit about that. Make sure you check this out. To learn more, you can also go out and check out this link. Again, there's a QR code to discuss these topics of this module. So make sure you check this out. Again, give you some time to actually copy that or take a print screen or take a photo of it to make sure you can go out. Also, make sure to take that learn module. Again, you can go out and take that by yourself and actually go through this and read through all of it we just talked about. And then last but not least, don't miss anything because we have multiple sessions coming up for the introduction to Azure Arc and others. So make sure you check this out. Follow the Learn Life Azure Hybrid Cloud Study Hall. We have, I think, over 14 sessions around Azure Hybrid. So it's absolutely going to be awesome. Make sure you join back. We will be definitely more of us, but also other speakers as well. So please make sure you join us with that. I want to say thank you very much for joining and a special thanks to Lisa for being here today with me and helping me with understanding Azure Stack. Thanks, Tom. Thanks, everyone.