Explain business capacity planning

- Before we deploy to the cloud, it's important to understand what we need and what are some of those requirements. Join us, as that discussion is starting right now. - [Announcer] You're watching "ITProTV." (light uptempo music) (logo whooshing) - Hello, ladies and gentlemen. Thank you for tuning in to more of the CompTIA Cloud . That's right, in this episode, we're going to be really diving into some of the things that we need to be aware of when we decide, "Hey, we want to move to the cloud. But we need to know what is it actually that we need in the cloud." And that's where capacity planning comes in, and just really understanding what the business requirements are. Now, Ronnie, I know there's a lot of things that we have to take into consideration here, but what are some of the things that if a viewer is maybe seeing this for the first time doesn't know and has never deployed to the cloud, that they need to be aware of? - So, and like everything else that we do in IT, there's always a cost associated with things that we actually bring up. That includes not only on-premises but in the cloud as well. And it's important that we have that understanding, because it's easy to actually think because it's not actually right here, that it really doesn't cost anything, and that it may actually have an impactful cost on your business. So what we want to do is make sure we at least understand some of the different components and the requirements that we're talking about as we start to move more things into the cloud. So as we take a look here on screen, this will kind of help us out. We're going to actually kind of be talking about the idea here of making sure we understand the ideas of what we tend to have to do in the cloud, excuse me, on-premises, and then we're going to actually compare it to what ends up happening in the cloud as well. We need to talk about requirements and templates and licenses. All these are actually going to be included in this episode as we get started here. So, Wes, let's go ahead and start by taking a look at the idea here of on-premises business capacity planning. Now, long words, not exactly super big words here, but the idea is fairly simple. Is that when we start to build out a network, okay, there's always going to be the idea of, "Do I have enough," okay, "To actually do all the things that I want to do? Or do I not have enough to do that?" So an on-premises type of planning here. When we start talking about capacity, capacity itself is costly. The reason why is that you don't tend to just build for the need of the moment, you tend to actually build for the need of, let's say, a year from now or a little bit longer, and now you're buying a ton of equipment to be able to do that. And when you do, Wes, there's a term I want them to key in on, and I want you to key in on as well, which is the term CapEX, okay. Now, Wes, let's talk about this idea of CapEX - [Wes] Sure. - Which is an abbreviation. What does that stand for? - That's capital expenditure. And there's a couple different models that we need to be aware of, right? Are we going to buy stuff up front, or are we going to pay over time? - Yep. So for us here, a lot of times it's a mix of things, right? You may have some capital where you're going, Wes says, "This is the server that we need." Well, you might be able to purchase that over a period of time, okay? And, you know, be able to, you know, go ahead and put it on credit or something like that. And that's still capital expenditure because they're really charging you for it, but you're still paying over time here. But it's upfront cost is what we're going to talk about. - That's right, so I have to procure it, I have to go out and I have to get that server. We have to put that in our closet, and we have to basically kind of define why we need that and the money upfront. - Yeah, yeah. So it doesn't matter if, you know, that in three years you're going to end up paying for it, it means that money's gone, right? - [Wes] Okay, sure. - Pretty much is what we have. So there's a lot of capital expenditures that go out in terms of on-premises planning. Now here's the big one, and all of us know this, if we've ever pulled up to a buffet, Wes, okay. It is easy to actually pay the money that we need to, sit down and just continue to eat and eat and eat. In terms of networking, it's easy to actually grow that particular network, okay? When you do, you're buying pieces of equipment and you're putting them there. But, Wes, here's the problem. What if your network never actually gets to that capacity that you actually thought I built all this stuff out here. It becomes harder to remove. So just like a used car, Wes, sometimes what ends up happening here is that you'll never get the value out of that piece of equipment that you just bought. If you don't need it anymore, it either sits there or you end up selling it on secondary market and you're not going to ever get back, that same amount of value to it. - Sure, so let's talk about an experience you and I've actually gone through. In fact, you actually deployed the server, right? You know, in a former training institute that we had, we were going to deploy this server. Thank goodness they actually loaned it to us first. (Ronnie laughs) But I can imagine if we had to buy that server, right? This server did everything, a lot of what we needed at the time, but it was super, super expensive, alright? And had we acquired that, had we paid the upfront cost, now we have a certain amount of years that we have to get our return on investment. And, Ronnie, we learned very, very soon that, yeah, that was a powerful server, right? It had all of the hardware we needed in it, but we learned that we were not going to get the full utilization. A good percentage of that server and its resources would just remain unutilized. - Right. And in that sense, now, like we said, you just have to either live with it or you have to find someplace to store it. - [Wes] Eat the cost. - Yeah. And, Wes, even right here at "ITProTV," at the beginning we actually did have a lot of capital expenditure because there were some things that we built online, or right here on-premises I should say. - Sure, yep. - But as we started to upgrade, now we had this bunch of used equipment that we've really still got stored in a room somewhere that we pull out if we need to, but it's really kind of sitting there, and that's something that we have to deal with. - And that, you know, the resell, you might say, "Well why don't you just resell it?" But the depreciation on that stuff, I mean, especially, ladies and gentlemen, those of you that have joined us probably know a little bit about computing technologies. In six months, is it still relevant? Well, it might be, but a lot of times, even in six months, right, we can't really justify the cost if we're not fully utilizing it. - Right, so there's a lot of that. Now the other things about the idea here of on-premises is capacity planning. Good thing is it doesn't tend to actually cost us anything in terms of planning, but we find out something, once we actually buy all this stuff and implement it, is that errors can be very costly. They hurt right away when you realize, "Wes, we bought the wrong thing." (laughs) Now you have to either cajole or you have to actually be able to see if you can talk to somebody and say, "Here's what we really needed, not this thing that we actually ended up getting." So errors can actually hurt us if we're not very careful. Also, when we do the planning itself and when we actually finally get what we want, we tend to separate and categorize very easily. So systems become very neatly, you know, labeled and identified as, "Here's what is for this, here's what is for that." So that actually is kind of a neat part of it. And, of course, there's monitoring and there's also maintenance to all this stuff, okay. So an on-premises type of business capacity planning, we have to actually think about all these things right away as we do this. But the great thing is in planning, it doesn't cost us a cent. But when we go to cloud business capacity planning though, here's the great thing, Wes. The idea of us that we actually have to actually build up capacity, the great thing is we only pay-as-we-go, okay. And that means it's not so costly. We can actually build up a server in the cloud, virtually, the way that we actually need to, and we're really paying for that utilization from the time we build it to the time we end up tearing it down a lot of times too, okay? So that's kind of a key. - So with OpEX, or the operational expenditure, right, versus capital expenditure, Ronnie, and to tie it back to some concepts we've already talked about, right? One of the things you mentioned was metering, right? We had that metered service, pay-as-we-go. We're only paying for those services that we're utilizing. And let's say that we happen to make those errors where we didn't do appropriate capacity planning. We're not necessarily eating that cost for two or three years later trying to get that return on an investment. - Right, yeah, we simply can actually say operationally, we don't need any more, deprovision it, and we are ready to go. Clouds, of course, in terms of capacity planning, we can actually easily grow them. And that means on-demand. If now we realize that we're actually at capacity, we can go ahead and spin up more, and we're only paying for that utilization. Deprovisioning is as simple as us turning the thing off a lot of times, and we really end up stop paying for the utilization of it, then we can get rid of that service. We still have to pay whatever the prorated idea is that we've actually been using it. But overall though, very quick to actually not have to worry too much about. Recovery can be almost instantaneous. In other words, errors that we might set up, or errors that we might have, we can kind of recover from them very easily. And here's kind of the drawback though. This is messy type of capacity planning, and that means we don't know exactly where everything is, right? When we start using a cloud, it could be right next door to us, but it could be a thousand miles away and we don't know. It be on-premises, it could not be on-premises for all that we know. Monitoring and maintenance is still the same. We have to actually keep up with those things when it comes down to these types of capacity planning here, okay? Now, once we actually get that in mind, right, Wes, we can start to take a look at the idea here of requirements. Now, I actually have private cloud in parentheses, and that's because the exam itself does not really tend to call out public cloud requirements a lot, right? So they tend to actually be focusing in on private cloud requirements. First and foremost, you need to make sure that you actually do some type of business-need analysis. You really need to actually have this or not. That's actually kind of a key when we do that. So you can actually do some type of analysis to actually help you out. Now, the other one, of course, in the private cloud is you will come out with some of the capital expenditures for the servers that you're going to need to be able to virtualize and build the cloud the way that you want to. So you also have to consider that idea of the hardware. Do I actually just need one server? Do I need two? Whatever else it might be. So you actually have to think about hardware, the types of software that you're going to use. Again, makes sense here that you might actually need to figure out whether or not one user's actually using that software, or is it multiple users? What type of licensings do we actually need here? We'll actually talk about that a little bit more in detail later on as well. And then how many people are actually on your network at any one time? These are kind of some very generic requirements that Wes is going to help us understand a little bit here. But overall though, basic requirements that we always have to consider when we start to actually do some capacity planning here for us. And all this actually relates back to the business itself so far. So this is in contrast a lot of times, and also, of course, even overlapping with what we call performance capacity planning. But just kind of realize that a lot of this is actually upfront. Does our business really need all this stuff that we're actually about to take a look at and do as well? And of course, lastly, Wes- - [Wes] We need money, that's right. - Yeah, when it comes down to it, you need some form of money, and here it is. If we go, "Just buy whatever you need," well, there's going to come a point when somebody's going to look at the bottom line and go, "Wes, you broke the piggy bank (Wes laughs) and you robbed your sister's piggy bank too to pay for this." - [Wes] That's right. - It might not be appreciated. - Yeah, and, you know, things can always go well if you've got buckets of money to throw at it, but eventually those buckets are going to run (laughs) dry. So business analysis and understanding the full requirements of what the business needs helps you to automate, or, not automate, excuse me, optimize is the word I was looking for. Optimize what your deployment need is and make sure that it's within the constraints of the budget that you have. - Yeah, and all that's key, right? So the idea here is for us to have a nice scope of what it is that we really do need. Now, there's also going to be the issue here of different templates that we can use that are out there in the cloud to help us to actually begin the planning for it, to actually go, "Alright, I might need this. I might need this here." You build out your network, you use these types of templates to really help guide that capacity planning. It can help speed along the process. It's not going to help you do everything, but it can at least help you speed along in terms of capacity planning as well. Now, Wes, in terms of everything else that we're talking about right here, okay, in this, we can dive a bit deeper on the performance side on a lot of this here. But there is one business requirement we wanted to actually do a little bit of a deeper dive on, which is going to be the idea of licensing that we actually want to do here. So, Wes, when it comes down to it, there are actually different models of licensing that we can actually make sure that we kind of pay attention to here, okay? Per-user, exactly like it sounds. At least I think so. Wes, am I wrong on that? - [Wes] No, I think you got it. - Alright, so I'm not going to actually, you know, continue to actually extend that out any further than we need to. - Sure, that's right. - Now socket-based, this one's a little bit more interesting, Wes, okay? This one actually says on your motherboard how many sockets are you actually currently using, and will charge you by that number of sockets. Now, even if you have four sockets on that motherboard, or eight, I don't even know how many that you could have, but it's really the active ones that matter, and they'll actually license it based on that. Volume-based, of course, is the same thing that you think of. It might be what you also think of as site licensing if you need it. So how many are actually using it? And then this one's a little bit trickier, Wes, core-based. And this is when you do have that CPU but you have multiple cores instead, they can charge you per-core if they need to. And ultimately it will actually come in the form of some type of subscription as well, okay? So those are kind of the ones that we want to at least make sure that you go over and that you have a basic understanding of. Licensing is going to be one of the very important things. But having said that though, enough of me talking, (Wes laughs) and that means what we want to do is actually kind of walk you through some of these different requirements that we're talking about here. And Wes has a nice demonstration for us. - Definitely. And we are using the Azure platform in this. So pick whatever platform you're going to use, I would definitely say, and stick with it. So for us, what we're going to do is kind of bring back some of these concepts that we've talked about, and you can see how it applies to something as simple as just creating a virtual machine. So that's what I'm going to do here. I'm going to create a virtual machine, and in order to do this, I'm logged into the Azure portal and I'm just going to click Virtual machine. And in this, and I'll tell you what, ladies and gentlemen, too, we've put these step by steps in the notes so that you can gain access to 'em if you choose to do this as well too. One of the things I'm going to do is just in the upper left-hand corner, I'm going to choose Create and then a Virtual machine. Now, in doing this, we have to pay attention to some things. One of the things that Ronnie mentioned was subscription, right? These are subscription-based, and you'll see that we have the opportunity to have various subscriptions. Now, I've got a specific pay-as-you-go subscription for me. The other thing I'd like you to keep in mind too is in part of your business planning, one of the things you have to be aware of is that in certain instances when you create or use a resource as part of that business requirement, it might require you to have additional resources that maybe, if you're not aware of, you didn't plan for. So, for instance, I'm going to create a virtual machine, but in order to do this, I have to create a new resource group that is a part, that bundles all of this, all of the things that make this virtual machine run. So for us, we're going to go ahead and we're going to just choose, I'll say CloudPlus, and we'll call this VMResourceGroup because I am, you know, that creative. (laughs) And we'll go ahead and choose Okay. Now, the next thing I have to do is I'm going to name the VM, and for us, we're going to just keep this, you know, the naming convention simple. We'll call it CloudPlusTestVM, alright? And now that so far, other than keeping in mind the resource group and the subscription that you are using, that is part of that business requirements in the capacity planning. So far we're pretty good. The region, I'm going to leave the defaults, we'll talk about those a little bit more. And same thing with the availability options. Now, I'll say that one of the big things that I want you to keep in mind is that when it comes to the resource planning and business requirements, the size is very, very important what you choose because you're going to be given a set of objectives, and those objectives say that you need to support this type of deployment. How would you, you know, handle that, right? And we can see that the size, Ronnie mentioned per-socket, right? That is a vendor-neutral way. Each platform will call it something a little bit different. For instance, here in Azure, they call it the VCPU. That's your socket-base. You want more computational power, you can have more than one virtual CPU, you're going to increase the amount of processing power. We'll talk about performance coming up in another episode. But for me, what I'm going to do is I'm just going to keep the standard B1 because, well, I'm cheap and it's easy to demonstrate. But one thing I kind of did skip over here, and I want to make sure that you're aware of too, part of the business requirements that also you have to keep in mind is what operating system in this instance are you going to use? And it might say, "Well, Wes, what does it really matter," right? Well, it does matter, because what software is needed in your environment? Do you need a LAMP Stack? Do you need to have Linux? Do you need to have Apache, MySQL, PHP, if you will, installed? Because if that's going to be the case, chances are you're not going to be using a Windows environment, you're going to be using a Linux environment, right? And that's part of understanding those business requirements. The other thing that we need to understand is think about the support model, right? Are you going to go proprietary? Are you going to go open-source, right? And there is no right or wrong, but you got to think about what the support model is in the end. So I'm going to go ahead and I'm going to do an Ubuntu Server here. I'm going to use the basic size. Keep in mind that that's all part of your capacity planning. Now, for me, I'm going to choose a password and I'm going to just type the username, and this'll be CloudPlusAdmin. Again, my creative nature here. And I'm going to go ahead and I'm going to put a super secret password, make it complex here so that it's, you know, a little bit more secure. Alright, and we'll go ahead and we will move forward. Inbound port rules I would suggest for your demonstration purposes and your lab environment, I'm going to choose none 'cause I do not want anybody connecting from the outside public internet back to this machine. So next thing that we're going to do is we're going to review the disks, and we'll just click Next. Now in here, again, remember knowing what performance of the applications, what you are supporting is going to be important here. We're going to leave all the defaults 'cause we're going to talk about some of these options coming up in later episodes where it's going to be a little bit more important for understanding what the options are here. Alright, now we'll go ahead and we're going to choose Networking. Now, here's another thing. Remember one of the things that I said as part of business capacity planning you have to keep in mind that one of the licensing models was a per-network adapter. Here in Azure, and even things in like VMware and stuff, they call 'em VNets, virtual network adapters, right? I need a virtual network adapter, and that's what the requirements are. But have I planned for the fact that, well, I also need a virtual network, alright? So, again, no right or wrong answer, but you do have to plan for that, alright? Do I want a public IP? I'm actually going to say no public IP here because, again, for demonstration purposes I don't need anybody to connect to this. Here's another thing, when I attach my network, my VNIC, or my virtual network adapter to a network, I have to have a network security group, or maybe not. Again, you have to bake in security to this as well too. As far as a lot of these other things, I'm just going to go ahead and choose next. That's going to be, again, some of these other settings we'll talk about in later episodes. But at this point I have created our VM. And remember one of the things that Ronnie mentioned was monitoring, right? We can implement monitoring right here. If we need certain levels of monitoring, we have to be aware of that and what the goal of the company is. Now, I'm going to go ahead and choose Next on this. There's some advanced information in here that we could select. For instance, do you have legacy applications that your business requires you to support? Well, you might have to use a Gen 1 versus a Gen 2 if you have older machines that you need to support. And again, the point and the context is business requirements. I'm going to go ahead and choose Next. And then as far as the tags, I don't really have any tags that I want to implement here. And then for the purposes of automation, you'll notice that the validation has to occur. One of the things that you can do is you can download this template. I'm going to go ahead and download the template here, download it just to my local machine. And what it does is it creates this JSON file here that has all of these settings that we just created, or we just selected. And then what I can do is I can do an automated deployment by taking this deployment template and just plug it in and it will automatically create the virtual machine and all the selections minus one thing, which we'll also talk about later. It'll remove the password that I just put in there. That's part of secure scripting. You never want to include the password inside of your scripts. And then at a later time, I can always use this to create this virtual machine. So, good little way to do that. I'm going to go ahead, we've downloaded this. And that's a great way that you can automate your deployments. We'll learn a little bit later about ensuring that when you deploy a virtual machine or any instance with inside of the cloud that it's in the expected state. It's a really good thing to have to make sure that you have consistency across your cloud-based resources. But once that's done, we've got the validation has passed here, we can see what it's going to cost per hour. And again, hopefully that's part of your business capacity planning. Hopefully it's part of your business requirements here. And you can see some of the personal information here, we'll have that blurred out there, but we can just review all of the settings that we have here. And then from there, once we are done, we can just go ahead and click Create. Now, one of the last things that I'll say is just make sure that if you are deploying these machines or these resources, make sure that it goes through the validation process and it is configured correctly. If you go through the validation process and it gives you errors, it's just a simple backup, if you will, and then fix whatever those errors are and revalidate the machine, and then you can click Create, and there you have it. You can deploy a virtual-type resource to the cloud provided you know what those business requirements are. - Alright, Wes, well, that was a great demonstration to help us understand all those different types of requirements that we do tend to see when we are doing business capacity planning. Well, that's the end of this episode, but there's plenty of more Cloud to go, so make sure you stay around for that. We'll see you next time. (light uptempo music) - [Announcer] Thank you for watching "ITProTV." (logo whooshing)