What is a Server | What is Virtualization | Emilio Aguero | Skillshare

Playback Speed


  • 0.5x
  • 1x (Normal)
  • 1.25x
  • 1.5x
  • 2x

What is a Server | What is Virtualization

teacher avatar Emilio Aguero, Technology Professional | Educator

Watch this class and thousands more

Get unlimited access to every class
Taught by industry leaders & working professionals
Topics include illustration, design, photography, and more

Watch this class and thousands more

Get unlimited access to every class
Taught by industry leaders & working professionals
Topics include illustration, design, photography, and more

Lessons in This Class

10 Lessons (1h 22m)
    • 1. Introduction to this class

      2:45
    • 2. What is a Server

      7:37
    • 3. Physical vs Virtual Servers

      19:00
    • 4. Servers and the Cloud

      6:48
    • 5. Top Server Functions

      17:42
    • 6. Domains, DC's and AD

      5:54
    • 7. Naming your Servers

      8:36
    • 8. Understanding DNS

      2:49
    • 9. Servers vs Supercomputer vs Mainframe

      9:25
    • 10. Your Turn - Summary

      1:13
  • --
  • Beginner level
  • Intermediate level
  • Advanced level
  • All levels
  • Beg/Int level
  • Int/Adv level

Community Generated

The level is determined by a majority opinion of students who have reviewed this class. The teacher's recommendation is shown until at least 5 student responses are collected.

5

Students

--

Projects

About This Class

My name is Emilio and I am a technology professional having worked with Servers and Server infrastructure across a range of industries and business sizes. Thanks for joining me. 

In this class we'll give you the knowledge and expertise to better understand and use Servers, for use in professional environments. If you are new to IT or have been in Technologies for sometime, this course will give you the foudational skills needed to manage your own Servers.

This class covers -

  • What is a Server
  • Looking at a Rack Server

  • Virtual and Cloud hosted Servers

  • Different Server Types & Functions
  • What is DNS

  • and much more

Do I need to be an expert?

No, this course is designed for people who are excited about tech and may be either working in tech or looking at getting into the IT industry. It’ll provide a foundation for anyone aspiring to work and learn more about Server technologies.

Meet Your Teacher

Teacher Profile Image

Emilio Aguero

Technology Professional | Educator

Teacher

Class Ratings

Expectations Met?
  • Exceeded!
    0%
  • Yes
    0%
  • Somewhat
    0%
  • Not really
    0%
Reviews Archive

In October 2018, we updated our review system to improve the way we collect feedback. Below are the reviews written before that update.

Why Join Skillshare?

Take award-winning Skillshare Original Classes

Each class has short lessons, hands-on projects

Your membership supports Skillshare teachers

Learn From Anywhere

Take classes on the go with the Skillshare app. Stream or download to watch on the plane, the subway, or wherever you learn best.

Transcripts

1. Introduction to this class: Hi and welcome to this class. My name is Emilia. I'm a technology professional and trainer and had been working in technology for over 20 years. Really appreciate that you can spend the time and watch this class and the next few lessons where we talk about servers and virtualization. This course is going to cover a lot of material around what a server is, the differences between a server and a virtual server. And also showing you some of the concepts around servers. We're also going to be defining different types of service to give you some ideas about what service you could be building yourself and what servers are common across the IT industry. Of course, servers or something that is used really across most organizations. Thousands of companies around the world are using servers in some form or fashion. So we're going to be giving you the skills and the knowledge on water server is and how to actually make and use service your self. Now, I myself am going to be demoing some of this stuff on a Windows Server environment. So I've got myself a home lab environment. If you have a home lab environment, and that is great as well. What I have is I've got something that's running VMware and I've got some virtual environments set up with multiple service built. You could be doing that yourself. But even if you're in a workplace that has some servers will still give you the skills and the knowledge to then be able to go and try this yourself. It's great that you're watching this and you're going to get the knowledge that you need, but then you've got to put it into practice. So for example, one of the videos we're gonna be talking about the top servers. What are the top service at a commonly used across industries? One thing that I'd recommend for you is then going and building your own servers to better understand what those services are and how they function. We are going to be talking about something like a domain controller or a DNS server. So then knowing what a domain controller and a DNS server, building them yourself and trial he knows, will also help you out. So what will be giving you a good foundation and knowledge around service? Re-watch these videos because this is stuff that I wish I had of had yeas and easy ago. And I know that you will definitely find this helpful. Do also let me know throughout each of these lessons, when you complete a lesson, what are you let me know how you went. Let me know whether you understood that. Let me know if you have some further questions. As I said, this is a very foundational course, but then the journey commences. From there, you're always going to be learning I myself, even having worked in technology for so long, having worked with service for so long, I'm still learning. There's always new things to learn. There's always new service to build. So this just formed as a foundation while you continue on your learning journey. 2. What is a Server : Essentially a service, it a voice that is a lot more powerful and ascended and use a desktop or laptop computer. Generally, the comparators inside the server will be very similar to what you find is our standard desktop or laptop, but a lot more higher-end. You can also add a lot more resources to a server. You can also have hour of additional source of cards or peripherals that I stated desktop or laptop would not have. As surveys generally going to be used in a corporate environment in a business. You might have uses to do some service testing at home. If you want to be doing some, you know, you're in lab urine learning as well. But essentially a Rack Server is going to be sunlight is going to be connected into something like what's behind me. This is IRAC. I cabinet were essentially it's a place that stores servers, switches, firewalls, storage unit first sort of things. And the fact that it's called Rack Server means it's going to be racked somewhere. It's not going to be like a tower device, like a tower desktop or a server that sits on the flow of, sits on top of something. This is actually going to have some rails that connects to the side of the server. And then it's actually going to be riled and rack into something like this behind me as server can be configured in a number of different ways. What I'm going to go into detail of the different types of service that we have. Different videos that talk about the different types of faults service such as file server, web server, DHCP server, DNS server domain controllers, database service. There's a whole range of different service, but really the end goal is to service something, to service the end users, to serve as an application, to serve as a website service, whatever it may be, essentially is going to be set up to have a particular server function to be able to provide a service to some end, end device or end user. So this is a rack serve up. Essentially its main purpose is to go into a server rack into a data center. So I'm going to just sit like a normal Tower server may say it's a full length server which will require rails on either end on the sides, on the left, on the right, and then wrapped into a server cabin into a rack cabinets. This is a one IU server, essentially taken up one rack space. And it's a quite sleek looking device, looks quite nice. Obviously the fronts of this is just a cover to cover that's covering the internal components or the front components of the actual unit. Easily lockable. So there's a little area here, we can just lock it so that you don't econ actually take this off. But you can easily just remove the front of it by just clicking this little knob, little latch, and it just comes off straight like that. We've got the unit, this particular unit doesn't have any hard drives inside of it because we are connecting this to a ES6 I farm, to a VMware ESX I farm. And it's connecting to a storage device called a sandwich has all of my hard drive. So the back of this unit is quite simple, being a one RU unit. We've actually got limited peripherals that we can put into this unit. We'll see that at the top, there's a couple of slots available, so I can actually add additional cards if I need to quite easily by opening it up for moving these slots. And then I can put it in the same way as a normal pace. See, we'll just go through some of these poets. So here I've got an, I drag polar. This is essentially a management port where I can run into an Ethernet switch. And the next is this whole server remotely on the network. Got a COM port, I've got a VGA port, I've got a couple of USB here, I've got my Ethernet ports. This is going to be my one gigabit Ethernet and then a couple of slots before 10 gigabits for NSF pace. And then a couple of powers here as well. We've got two of them for redundancy and high availability, because obviously we don't want power to go out from your server itself. So we always sit them up in pairs. These would be running into two different power sources. So one into one power rail or until you different UPS than the other one. So that if one power rail, one PDU when UPS goes down, I've got access to the other one and gives me power. These are easily removable by pulling a little LacI and the whole unit just comes out like so. I will buy a brand new one if a particular unit guys, and then just slot it in as normal clicks right into space. Essentially these rails are what's going to be connected into your server cabinet and also into the side of the server like this. So you'll see that the rail itself moves left and right. So you're going to connect one of these into your server cabinet. And then these little notches on the rails match up to the notches on the side of the server. So here we've got the inside of the unit, so I've just taken that cover off by unlocking it and just removing it. And you said that it is quite a detailed piece of equipment. So the fans will keep all of this unit stuff he called and the whole server itself called. These are the RAM slots, okay, so the ram is going to be inserted right under H0 and then your two big heat sinks for your CPU. So this is actually got two separate processes or what they call sockets. A socket where your CPU is going to sit on top and then your big heat sinks on top of those CPUs. Inserting RAM is quite simple. You literally just click on the little tabs here, on the side. Plastic little slot comes out. This is really just to protect the actual groove of where your RAM sits. And then your RAM would just slot into place to insert some more ran the same way that you do it on a normal PC. You apply your RAM and use flawed it in. Where you're going to need to do is you're gonna need to pull these levers up left and right. I should remove this cover to be able to access the RAM slot on this side here. So here's a couple of power units and from the top got a couple of other peripherals where I can connect additional functionality into. Here's a, here's another riser, essentially allow me to add other cards. Can remove this by lifting this tab up, the whole unit just comes out. I can add additional cards into here. So as you can see, it's laid out very, very nicely, is actually quite nice and quite packed full of features. So we are here now inside of our columns room because there comes cabinets. We are going to be racking the unit. And you'll see that I've just got a number of slots here. Association, I'm going to attach the rails on the left and on the right. And then once that he's in on the front and on the back of the unit. I can then slide my device in as normal. And then I'm good to go OK. And then it connected, I can connect it into my particular, to my particular switches, into my particular powers on my left and then right. And then get that on the network and go ahead and configure it and then use it. So here is an example of a device that is rats. So you've got the cabinet right here. You've got the rial that's hooked up into the cabinet. And then you've got the actual system itself, which is connected to the rail. We saw the top, this is the actual unit itself. And then the rail runs all along the side here and it's actually mounted into the actual side of the system. Most units will either be to IU or one IU or more depending on how big it is. This particular unit that the example that we're looking at here, this is a two IU units to saying enough to rock units faces. I wanna, you would essentially be half of that space. And you see that right here. It's actually going to be clicked into one spot. But he's capable of taking up to the two ru. So obviously whenever you're racking one of the systems, you want to make sure that you've got enough capacity, enough storage space to be able to accommodate other the one Ru with a tool or largest space. 3. Physical vs Virtual Servers: So we're really talking about service and server infrastructure that is on-premise well, that he's heard bothered basis for the IT managed provider of a company. We are, you know what, I'm going to talk about things like Cloud because you can have a service sitting on the Cloud or managed by other parties. For example, we're really talking about service that our in-house, they could be in your physical building, that could be in a data center or in a co-location site, still owned by the business. So we're really focusing on service, physical and virtual that are within your business. So this is an example of what a physical server infrastructure would look like. Physical servers had been around for a very long time. In the olden days, there was no such thing as a virtual server that we're all physical. Every single service that you would have would be a physical server. So physical seven, he's literally like a computer that has a lot more smarts and a lot more brainpower behind it, and can perform a lot better than your average desktop can, is physical in the sense that I can physically see it and touch it and actually go in and open it up and do all those sorts of things with that physical server. In this case, I've got an example here. A full physical servers. One is hosting a domain controller in DNS. The other one is a mail server, perhaps something like exchange occur a fossa which all of my files sitting on. And I've got a database server which could be used for multiple databases. Now, this is a standard operating structure for physical servers. You've got a physical server built that has functionality within it. So each physical server would have its own function, its own purpose. In some cases, a physical server could have multiple functions. So in some businesses, for example, one physical server could be the domain controller, DNS, the mail server, and the file server, for example. In most places you would, you generally want to split up those services into multiple service. Otherwise, you're going to be having performance issues and you have to upgrade after that. Essentially have a very, very powerful so to be able to cope with the operations. And the other scenario is if you lose that 17 I contains everything, you're in big trouble. So it's always good to look at load balancing your services across multiple service. All of the service would have some sort of storage attached to it. The physical servers themselves could have their own storage built-in. So they said a ride worth of hard drives built-in, and all of the data is sitting on that sign with the mouse over the file server, the database server that could have their own built-in storage. In the case of most, most larger businesses, you're going to have a dedicated storage device attached to it. So this would be a saint or run as a storage area network or a network attached storage. And that is essentially connected into all of my service. So the service might not actually contain most of the data. The data is actually sitting on some sort of a dedicated storage solution. Now, this seems like a very good model, but you're going to consider this is a high cost model in the sense that I've gotta have physical server for each individual application or each individual service about wanting to provide to the business. If I lose one server, I lose all of those functions on the observer, right? If I lose the mail server, I've lost my own, that could be detrimental to the business. The business may not be able to operate unless it has active email systems. Of course, the advantage of having physical service is that they physically, that they are dedicated and I can go in and upgrade them and they are dedicated for one individual service. But I do file in the era of redundancy, there is actually no redundancy built-in out of the box for physical service. As well as that thinking about costs when we're talking about things like electricity and physical Rackspace as well. The service have to physically sit somewhere. They have to be powered, that have to be cooled by a conditioning units. So if you consider all of those options, the cost of physical service is generally going to be higher than if we say look at the alternative, which would be the virtual server infrastructure. If a server does fail, for example, and I've lost my DC, my domain control MID and S. Of course I'm going to have services interrupted. Then I have to organize a replacement server, rebuild it. And if it's running something like the Windows Server, I have to reconfigure Windows Server, reconfigure my DC and my DNS. I have to restore from backups, but I'm essentially rebuilding a so after source that survive to reconfigure it, Racket, cable it up, reconfigure all of the applications and the software before it's often an operational scalability on a physical server is always going to be a bit of a challenge if I need to upgrade the resources, for example, uh, my mail server that is running 30 to eat your brand and I need to upgrade it. I think you have to power down the server source. The RAM from rubbery may be I have to white that amount of time for that rant coming off to go down to the local IT store and purchase some additional RAM, open it up, put it in, and they guard right, and then I can bring everything back up. Upgraded from 32, 64 gig, but that amount of time, the amount of effort to do that is quite challenging. If I have to upgrade the CPU, I'm maybe stuck. If, if I've got 16 cause on two CPUs and then I need more resources. I need to upgrade or double it to 32. You're going to be in trouble as well. So the scalability of a physical server is a bit of a challenge. Generally, when you buy a physical server, you want to purchase it already with enough resources for intended 12 and 24 months, even five years worth of growth, it is a challenge to look at operating those services later on. So if we compare these two, a virtual server infrastructure, I'm like the physical where I've got multiple servers hosting different operations, different services. I've got in this case of virtual server, I thought one physical server hosting multiple virtual service to the physical server, in this case, could be what's called a hypervisor. There are the big vendors which are your, Citrix and VMWare. Microsoft have their own virtualization technology. Essentially it's a physical server still running a hypervisor operating system. If we're talking about VMware, for example, the m ways hypervisor is called ES6. You install ES6 island to this, and then you would go and configure multiple VMs virtual machines within that one server. So now rather than having multiple hard multiple pages of highway multiple service, yeah, I've got the one server hosting multiple VMs. Now the great thing about this, if I do a comparison, let's add what the domain controller in the physical environment, heme. And it has ran and it has one CPU. If I need to upgrade that, I would have to power those services down, open the box out because that additional hardware and install an additional hybrid if I can even upgraded in the first place. In this scenario, this hardware, the server, has got the races already built in. So you would go and purchase this hypervisor, this physical server, with enough resources to be able to cater to all of this and then some more intended growth, right? So if my DC now requires more resources, I can literally upgrade the race horses on he using and physical resources that are sitting on my hypervisor up. So I can easily boost up the RAM, the CPU of any of these directly using the resources from my harder for my hypervisor. So essentially a chunk of RAM, let's say this particular server has 64 gig of rent physical 64 gig of RAM. I can allocate components of that 8, 16, 24, 32 gig of RAM, for example, across these vigils. And it's going to use 32 gig of RAM physical on my physical server. Besides the CPU, the CPU can be shared. So I can use the CPU that is unhappy and shape components of it across my virtual service. Now the thing that's very important is when you are speaking up or scoping out your virtual server infrastructure is you got to think about capacity. You're going to either think of it, hi, going to provision those the end that he can under and over provision VMs. You don't want to essentially create too many VMs in here that they start to run sluggish. Because what can happen is there are, there are operations in place yeah, for sharing CPU resources in usage. Where if this bar here it is, which virtual server hogging up or taking up most of the resources and the other service could potentially perform poorly as a result of this server using up most of the resources. So you always have to speak up and allocate the resources accordingly on your physical server across your virtual service. So straight away you can see that this is a lower upfront cost model. Rather than me having to purchase full service. I've just purchased one and I'll be sold at multiple virtual service on top of it. The maintenance of the soma is significantly easier as well because I've got less of them. If I have to go and maintain all of the surveys, routinely going in and checking and making sure that The renewing the warranties up reading the pods, updating firmware, all those sort of things. I wouldn't do it across all of my physical service if I've got hundreds of them, that could take a long time compared to a hypervisor or a physical server running virtualization technology. I do it essentially on the less infrastructure. The only consideration that you need to make is when you're choosing one or the other as well, is the compatibility of the applications, all the programs that are running on that physical server. And there are still some services that can not run virtualized. And that is changing, is changing very, very quickly. But just keep that in mind and look, I'm going to say that majority of your core and services are going to be able to be run virtually. So you don't have too much of a consideration then, but do double-check if you are running third party applications, potty applications that may be tight and just for your business or is not very well known in the industry, can see that whether they can actually run on virtual service as well. Ideally, when you're sitting up things like high availability and redundancy, you don't want to have just one physical server here. I'm running virtualization technology because in the case of this, if in the physical world, I lost one sever, I've lost 17. Services could still potentially still operate. In this scenario. If I lose this physical server, I lose every one of these virtual service. So what you do is set up multiple, at least two physical servers, set them up in redundancy so that I have high availability. So services can flip between the two. So I can have a second hypervisor configured between them. They connected, they're what's called highly available, high availability. So there's the technology and configures and the two service know about each other. And I can have the service split across the two. So I can have, for example, I can turn the day, say often him, uh, be running it over. He. And I can actually split the services, all the VMs between the two service. In the event of a disaster and a physical server die, the psi dot server died, right? All of these VMs could be spun up over here. So in this scenario, I lose a setup, I'm in big trouble. I can't really get anything unless I restore it. I've to build any set of potentially restore the software, restore the backups. In this scenario, I lose a server. I can have everything is spinning up and running on the secondary server right here. If we're talking about much bigger picture, if you lose, if you lost both of them, if they both like head and saying the same data center in the same columns room, you can set up multiple servers across other locations and then have them replicated from one side to the other. So that in the event of a disaster and your operations do file, you can easily spin them up in another location. While in the olden days, in this physical server infrastructure, unless you've got it configured properly. If you lost your on-premise service, if you lost your data center, you'll comes regular server and you can lose everything. And then to be able to restore everything could be a huge challenge and could have a significant financial costs of the business. So as you can see it really in the scheme of things, virtual service, definitely a better option for, but for most businesses, you're thinking about cost-effectiveness. Performance can be improved because I can easily increase the performance of these VMs quite easily. And disaster recovery is much easier as well because I can have services such as high availability configuring services file and one that can fail over to the other. So the things you gotta consider when you are looking at physical or virtual and comparisons between the two is really you starting off with the budget, right? You're starting off with how much is it going to cost me? How much physical hardware do I need? How much is that going to cost me in terms of procuring the hard way, setting it up so and maintenance, somebody physically going out and maintaining it and building the server, right in a scenario of let's say, of an IT professional and have to go and source IN particular piece of hardware. And you either they could do it in-house or organize a third party. But you have somebody needs to physically set up a server in a virtual environment, spinning up a seventies, considerably easy. I could just literally I virtual server that's already sitting on some form of physical server infrastructure. Consider your disaster recovery and business continuity. In any business, you want to make sure that services are operational as often as possible and not in the event of a disaster, in the event of something going wrong, you can restore operations as quickly as possible. Restoring operations in this model is much easier and much more timely. Like you can actually restore things in a much quicker fashion. This model is a bit more challenging. It can be done effectively if you are willing to put an invest the money and you have full replication set up across multiple physical service. Virtualized disaster recovery and business continuity is a lot more consistent. So remember that if you lost a server in this environment, That's seventies to be rebuilt in this environment, you lose it. If you've got high availability configured, it just spins up and runs on E and you potentially have no to little as each little could be only a few minutes while the server is booting up again. If you lost everything, you're in big trouble. If you lost everything he it could be haven't replicated over to alternate sites. You lose this. You can spin up everything over he quite easily. So most surface will have some form of backup in place. He's actively reviewing BI service and backing them up on a regular basis, whether it be daily, weekly, or monthly, always have to then go off to ice, some sort of a backup media and retained for a certain amount of time, et cetera. But in the event of a server to server dying, I've to restore everything from scratch. You can take a long time. It will take a long time if I have to procure the new hardware and ever and everything, if I lost a mail server, like, let's say this is like high, but I've actually lost a virtual server. The restoration of that video seventies a lot quicker. Also, in a virtual environment, you can do Full snapshots of the VMs themselves. Essentially, the VMs are just a big file containing a lot of mini files that resides somewhere. So restoring I virtual server is much quicker and more effective than a physical server. Performance is the other consideration. Purchasing some hardware for your physical server, then purchasing a hypervisor and adequately specking out your virtual service and you can have adequate performance right now down the track as your server acquires better performance, we need to improve the performance. It's more challenging to do that physical side than it is on Virtual WAN virtual you can just literally spin up that VM and just give it more CPU, more ran literally on the flight. In some scenarios, you may not even have to power down that VM to do that. Sorry, it is much more adequate and much easier and so much simpler in a virtual infrastructure as well. And then let's think down the track. Let's say you need to migrate services. You're changing offices, you are acquiring a new business or you've been acquired by business and you need to migrate services, or you're running an old operating system and you need to migrate services from one to a new operating system, for example, whatever that scenario might be, migration of virtual service ease, easier than migration of a physical server. In a scenario of hea, funny to move service from one to another. I literally can just move them on the fly from these hosts to this horse. I can procure a new piece of server infrastructure, physical server, and fragile and sit up here. And if I need to move VMs from here to here, is essentially once that's all configured, I just right-click on it and I can move it on the fly from one to the other. I can't really do that on the physical infrastructure. If you're changing offices, a virtual infrastructure, I literally just pick up that hypervisor, one physical server or the multiple contained by multiple VMs. And I just really like hate them. In the physical. There's obviously a lot more, so it's a lot more effort to move things between one side to the other. So that is a overview of physical and virtual. So really in summary, physical, you're going to have to have physical pieces of hardware, physical servers running multiple different types of applications. While in a virtual infrastructure, you can literally commission one or more physical servers running hypervisor technology, and that had multiple virtual servers running within it. 4. Servers and the Cloud: Hi, welcome to this next video. Let's talk a little bit about service in the Cloud, right? So we've got a good understanding now that you've got service that can sort of sit in various places, you've got physical service where you go down to your local computer store or you distribute out whoever it is that gives you physical servers. These can be ranked based servers, these can be serviced that blade service, that could be tau, a service, that could even be a standard desktop or laptop PC that is acting as a Service. You're installing server software onto them. And then of course, you've got the cloud. You've got cloud services that are sitting in the Cloud somewhere. Now, the big ones, the big players when you're talking about Cloud Service or Cloud infrastructure specifically are of course, AWS and Microsoft. Google also has their own stuff, but I primarily stick with Microsoft and AWS, which is Amazon's AWS or Amazon Web Service. And Microsoft Paint, Microsoft Azure as yup, however you depend however you want to say it. So these are if we're talking about service specifically, there are servers that sit in their infrastructure, right? So you could have wherever you're working, if you're looking at getting a job, you're working in a job right now. You could have servers that are both sitting on-premise, which you call on-prem or on cloud, on the cloud or cloud based or cloud hosted. And they're going to be sitting on generally one of these three, AWS, azure or the Cloud, google Cloud, but there are other services out there that are also cloud base. Now of course, the servers that are sitting on the cloud generally going to be virtual service, right? So we talked about physical servers and you talked, we talked about installing some virtual operating systems. So you've got VMware and you've also got Microsoft Hyper-V. There is a sort of software technologies that let you install virtual machines on top of physical infrastructure, service in the Cloud. It's really the same. There are virtual VM's, virtual machines sitting in, let's say AWS, you could access to what's called an EC2 instance, which is an AWS technology. And that is sitting within this VM and ACE2, a VM itself. And that in turn is sitting on some physical infrastructure. And this is the great thing, is most people don't think about it this way. That when you're thinking about AWS and you think that mocks and all these big guys who have cloud-based computing, there's actually some physical server, storage and networking. Someone out there. It's all configured. It's all configured for you. And then I just essentially lease or loin a bit of that physical server space for you to use and install your virtual machine within it. Now a lot of companies have got birth. They'll have on-premise and they'll also have Cloud. This is what's called a hybrid model, where I've worked across many, many different companies, many, many years. I've always seen hybrid models. It's very, very rare that you find a company that is 100% on the Cloud. They do exist, but it's not that common. Most companies will have a hybrid model where they'll have both on-premise or physical servers with perhaps virtual machines installed on them, and then other virtual machines that are now cloud-based. Now of course, because it is Cloud-based, there are pros and cons, of course, right? If you've got a VM sitting in your company where you work, or perhaps where you want to work. And that server goes down responsible, you lose that. You need to get it back up and running. If it's sitting in the Cloud, they're responsible for it. I mean, you sort of are in a way because you you manage it and you paying for it, etc. But they're also responsible because it's their infrastructure. And what you'll find is when you have your service on-premise, you need to make sure that they're always up, that they're secure, and that they've got backup power UPS is always sort of stuff all configured so that if the sighted side there is a power outage, you don't lose them. Well, if they're in the Cloud, then Amazon, Microsoft, Google, they're responsible in a way for keeping your server up and running. So that's why you pay them. Because you're insured more by uptime. They have like a very, very good percentage of time where it's very, very rare that a server will go down on the Cloud. It does happen. They're not a 100 percent foolproof. But there are so many levels of what's called redundancy that Goswami levels of high availability where if one thing fails, this thing picks up the slack and it doesn't file actually because it's just sort of this course, then this one will still operate the server. And then if that one goes is still something else that can still make sure that the server doesn't go down. So the cloud is brilliant because of that, because you are paying for a service that is more reliable. And, but of course, the thing with the Cloud as well is that you need to now learn all these new technologies. So if you're familiar at all with on-premise technologies, you're familiar with different sorts of servers and the different technologies around the service. The eye pays, the networking, the DNS, dhcp, all of that sort of stuff. Well, on the Cloud, they have their own technologies. There are names that are in everything that is sitting in the Cloud that you now need to learn and understand. It's something completely new. And if you look at AWS, if you look in at a zoo or even if you look in at Google causes, there are certifications that you can go and get to become experts in those particular fields because that is a whole new world. So just because you are an expert with servers that are on-premise, doesn't mean you're going to learn and be an expert straight away with the Cloud-based technologies, you will have to learn some of that. That is a bit of an overview. You can go into a lot more detail around the cloud. The cloud is awesome. More and more companies are going towards the Cloud. What we're going to be talking about in this series is the foundational stuff, right? We're going to be talking about the foundation stuff in an on-premise, on-site environment. But you can use all these practices, everything that you're going to learn in this course can also be used in the Cloud. So nothing here is going to stop me from understanding at least the basics to then give you a better foundation when you do want to implement more Cloud technology. So it's not one or the other. This is the foundation stuff that you need to know that can then be used in a cloud when you are configuring all of your Cloud technologies. 5. Top Server Functions: So let's now go through this top 20. We'll go through them really quickly. Give you a bit of an overview around each of these and what they are. Number one is a file server, a very, very common server. A lot of companies will have these. And it's essentially I set up that is for storing files and folders. So you're going to be citing business data, documents, images, whatever the actual company does is going to be stored on a file server or a fleet of file servers. And then these file server is going to be accessible on the network for people to connect to it. So commonly, if you are working in a company and if you're running on a Windows computer, you open up your mic computer, you'll fall Explorer up, and you've got other drive you have, of course, your C drive, which is your primary Windows installation. But then you might have a G Drive and H drive and F drive as then drive all these other drives. Most of the time they're going to be pointing perhaps to a file server of some sort where that particular, you know, if you're in marketing, if you're in sales, if you're in finance, that could be data that is relevant to that department and it is stored on a file server. Commonly, a file server would use protocols such as SMB, NFS, as well as common. It could be on a Windows-based file server environments, could be on a Linux Space File Server environments. You could even have a nares, which is a network attached storage, which is acting as a file server as well. Second one is what's called a domain controller. This is almost like the elementary server to ensure that your network and all of your servers and all you need to buy your devices, laptops, desktops, Mac, PC, Linux are all communicating within this environment called Active Directory. And Active Directory is stored on a domain controller. So you'd build a Windows Server. You then go and install a domain controller role. You then go and have all of your Active Directory tools where you manage all your users, your permissions, what people can and can't access that computers that are bound to this thing called a domain is all managed within a domain controller. Within the domain controller, you also have things such as group policies that are essentially allow you to push out policies to a fleet of computers on your network, not just desktops and laptops, but also serve as making sure that the passwords, for example, are complex. Perhaps you want every computer on the network to have the same wallpaper, the same screensaver. Perhaps you want to disable right-click, you want to disable access to control panel. All of these is all controlled by a group policies on a domain controller, you think got a DNS server. Now commonly a DNS server will also sit on a domain controller as well. If you are in a larger business, you may have multiple DNS servers. And of course, DNS is going to convert your host name. You're fully qualified domain name, which is a FQDN, to a IP address. So the whole DNS server controls when you ping a particular server, when you ping a website, when you ping whatever it is within your organization via a normal host name or FQDN. It then we'll resolve to an IP address. You can also do reverse lookups on an IP address to go and resolve against the host name. But this is almost like the address book for all of your IPs and for all of your systems on your network to sort of resolve between each other. And before is your DHCP server now computer on the network, how does it get an IP address? Well, commonly it's going to be plugged into a network, is going to be connected to a Wi-Fi device, is gonna go and scan your network and ask for an IP address from something called a DHCP server. The DHCP server manages all of your IPs for all of your devices, and then pushes out an irrelevant IP address to a computer, to a server, to a Internet of Things and IoT device, whatever it may be. And it gives it perhaps not just an IPA, but also the route up the gateway address, the DNS addresses and other information as well. A web server, of course, you could run a website some place that is stored on a web server. You could have something like Microsoft's IIS, which is common with the Windows server fleets. Alternatively, you can use something like Apache. You can use PHP, and you can store that on there. You can even run things such as WordPress or Drupal or other sorts of web applications on a web server. Web servers and other servers will communicate with something like a database server. Perhaps a database server could be running something like Microsoft's SQL. It could be running Oracle, for example. It could be running MySQL as well, which is a free version. And it's essentially just a database, essentially thinking about as a big repository of information and a website and application. Something on your network goes and queries your database to retrieve information. The database is almost elementary for a lot of applications and websites nowadays, especially when you're looking at hundreds and thousands and thousands, perhaps millions of rows and cells that need to be contained in a centralized location such as a database, an email server, most commonly being something like an Exchange Server. Nowadays, these may be a little bit less common because a lot of companies nowadays are going on the Cloud and they using the Microsoft 365 services essentially exchange on the Cloud. But a lot of places still do have some form of a Internal e-mail server, whether that be exchange, whether that be some sort of an SMTP service or a relay or something like that. But something that is hosting e-mail services or at least managing some sort of e-mail mail services within a company. Application servers are a very broad term, but an application server can store one or multiple different types of applications. These could be apps that are specific to your organization, that they may have been developed in house. They use just by a particular company and they're not used anywhere else. And that's application is stored on a server and that server becomes an application server, there are other apps that you may have, other third party applications that are installed on an app. And that could be an application server as well. There could be an application server storing multiple apps, and it's just a common name application server. And that could be housing multiple apps, providing multiple different types of services out to the network and out to your users and computers and servers perhaps as well. And that's very important to make sure that all of your data is kept safe, that all of your data is backed up. It's elementary and almost every single organization needs some sort of a backup mechanism, backup process in place to make sure that all of your data is backed up. Now whether that be backing up your end user computers, your desktops and laptops. At least you need to make sure that your servers are being backed up. That data is being backed up to type to a hard drive to the Cloud, sent off site to another replication server in another location, whatever that is, it needs to be managed by some sort of a backup server. Now, the most common sorts of applications or backup processes or apps that you could use would be the big ones being Vin, you've got CommVault, you've got Net Backup, you've got IBM's TSM as well, and there are some others out there as well. So backup server, using these apps, configuring your backups. So if you do lose data, you have that confidence that your backups are there and that you can restore the data. Every computer on your network, every server on your network have updates that need to be installed from time to time. Of course, if you are familiar with IT a little bit and you've worked in IT and accompany, you will know that Microsoft release patches every second Tuesday of the month they call it Patch Tuesday. Other companies, Apple, Linux as well, different versions there and even applications are releasing updates from time to time. Apps need to be updated, the operating system needs to be updated. These vendors don't just release patches for the sake of releasing patches, often they have identified some sort of a security vulnerability. There is something, there's a bug, there is something that needs to be fixed. There are perhaps updates, feature updates, and it'll be deployed. And all of these can be managed in a patching serve up a server that is dedicated for patching. Common ones would be something like W SUS, which is used for patching out windows, computers and servers, SCCM. You've also got other ones such as jam for the Mac. And you've even got SolarWinds Patch Manager and others. Yeah, there was a third party patch management solutions for patching your fleet of computers and service management of phones. It'll be done somehow. So these are desk phones, perhaps even mobile phones, but we'll focus on desk phones, specifically VOIP, voice over IP. These phones need to be managed centrally somewhere. They're going to be connecting or at least communicating with sort of like some sort of a VOIP system of PBX, something out to the outside world you're gonna be doing SIP trunks are going to be connecting to perhaps phone switching devices that you've got in your CMS cabinet, in your server cabinet as well. But it's all needs to be centrally managed within some system that can manage extensions, manage groups, all done through a phone server. If you do a lot of printing, even if you do a little bit of printing, you can have a print server also sit up. Commonly if you're in a smaller organization or even in a medium organization, the print server services could actually be run on serve up. Some, some places actually have it running on the domain controller or their file server that just installed the printers onto there. But if you've got a lot of printers, perhaps you have multiple floors, you've got multiple buildings, multiple sites. It's sometimes a good idea to actually have a dedicated print server that is managing all of your printers, all of your printed queues, and all the relevant drivers for all of your printers, for both Mac and PC are the things that print servers are very common for as well, is now something like follow me printing or printing where you send a print. Again, C gets added to acute. Within there. There's some specific print server software that is now tied to ID cards, to fobs, to sort of some sort of swiping device with a user now goes to a printer, swipes, they pass onto the printer and then the print will come out. It's more like a secure printing and all of that is managed through print service, FTP, and SMTP servers. If you have an external company who you need to send files to and from, you could house this within one of these FTP or SFTP servers. Of course, the differences between FTP SFTP is SFTP is more secure and is encrypted, encrypted and things like that. So generally I would recommend SFTP, but the very common to use still nowadays. You need to send a file to an external person. You store it into this SFTP. Perhaps the server is external facing, it has proper security. It's in what's called a DMZ or DMZ zone. And then you give credentials, set up an account for an external company that can then login and they have access to that file that you've just shared with them. And that's really the whole purpose of that server. So to ensure the health of all of your servers, you could set up some sort of a monitoring server. And this is a server where you actually install some software. You then scan your network, you add devices, you add servers into these monitoring software. And it essentially monitors the health of your equipment. The CPU, The ran, the hard-drive space. How is your sweet how your switches performing, how's your firewall performing intrusion detection sort of stuff. And there's a lot of apps that can do this really, really well. You've got things such as Ryan, you've got PRT, gee, you've got Nike OS is all of these other apps out there as well that can essentially control and monitor your service, for example. And then when things are triggered, when things are going wrong, when the CPU is too high, it sends alerts to the right people so that you know, when things go wrong and you can prevent things from going worse because you are alerted to something that may have happened on your network accordingly and then you can respond to it. If you're in a building that has some sort of building security, you have your doors that need passes. Perhaps you have cctv cameras, you have other sorts of security in your organization. I serve out or fleet of servers needs to manage this. Essentially you're building security. Again, passes cctv and other things like that about the security of your network. Now you can have firewalls that are software-based firewalls. You could have proxies that a software-based proxies. You can also have hardware firewalls where actually you go and buy yourself an actual hardware-based firewall, hardware based proxies. Or you can have software versions of those which are running on a server. So an example of this would be something like PF sense, which is a software-based firewall. You can also use it as a proxy or configure it with some sort of proxy services. And that is, that is sitting on a server. It's a Linux-based serve up, but it is acting as a firewall. And then you point your devices to go and filter the traffic in and out through your network, through a software-based firewall which is sitting on a server similar to that, you've also got load balancing service. Again, load balances can also be in a hardware form when you purchase load balancing service, or you can go software based load balancing service. Very good example of this is a website that get, that gets a lot of traffic. The last thing that you want to happen is you get tens of thousands, millions of hits on a website. And then that one website crashes because it cannot handle the traffic or you need traffic to be filtered between different locations. So what it does instead is it hits some sort of a load balancing server that then can spread the load across multiple locations based on a number of different requirements or prerequisites that you can figure out in that load balancing server. The next one is a terminal server or a remote desktop server. This is essentially like a jump box. So this is a place where people login to the server to then access something, access perhaps another server. So for example, you may not want your staff to have direct access to a particular server. You might not want your IT staff don't have access to a particular server, but they can access services on another server through almost like a jump box or entry gate server, called a terminal server. And that could be for a specific purpose. So for example, you may know what your staff to access your domain controller directly. You know, you don't want Active Directory tools available to your IT staff by them logging into a domain controller. So you could set up a terminal server with the Active Directory tools installed onto there. And then of course that in turn talks to your domain controller. So your stuff access the terminal server to access the services from another server or some other application that could even be configured within a terminal server. Then we're talking about virtualization. Now we didn't talk about this too much. But of course you can have physical servers, you can have virtual service is equal servers in the olden days, you would actually go and install physical servers, lots of them, and install one particular service on a physical server, you would install a file server on one physical server, mail server on another physical server, a DNS server on another physical server. Nowadays, a lot of this is all virtualized, so you get a physical server, you install some software onto that, you convert it into what's called a hypervisor. And then you can build multiple virtual machines within that hypervisor up. I would recommend going and checking out some of my videos when it badly. So if you do want to learn more, do check out that video right there. Well, we are talking about physical versus virtual service, so you can get a good understanding around that. But essentially, some physical servers you will keep could be converted into hypervisors. The most common ones being VMware as ES6, you've got Citrix, Xen server, you even got Microsoft Hyper-V. And essentially this is a host snail, a virtual host or physical host acting as a virtual hypervisor to allow you to install Virtual Machines so that you can then go and build your network and then you go to serve it to essentially manage your fleet of servers. An example of this is VMware vCenter. This is a piece of application software that you install and set up on a server. And then you can go and manage your fleet of ESX. I harvest your VMware ESX host, so your, your fleet of hypervisors and managed by one central server, in this case called a vCenter Server. And the other vendors such as Microsoft and Citrix have similar things. Because the last thing that you wanna do is be logging into each individual hypervisor. We talked about that one before, one by one. So you have one central or a pool of central vCenter servers or virtualization server host servers to go and manage all the fleet in one centralized location. So that was the 20 we did a very, very brief snapshot on each of these, but hopefully you learned something. 7. Naming your Servers: We are going to be talking about server naming convention. How should you be structuring the name of a server in a business? Wouldn't it be wonderful if just from the server name, you could find out a whole bunch of information about that server. You may be able to know exactly where the center is located, what sort of server it is, and what sequence of numbers is just from the server name. So you don't have to necessarily go in, look up an asset register to find out another server because it's already built into the name of the server itself. Now before we even get started with the server name, you want to have make sure that it is a total. It's generally good practice to be a total length of 16 characters. Don't make the server name any longer than 16 characters. As a good practice, the server name at the very start should always contain the business name or at least an abbreviation, but ideally an abbreviation of the business name. And you may be saying, well look, I've only got one business and I've only got business. Abc fruits. I'm never going to have that name again, right? What if in the future that business gets acquired? What if in the future you've grown and you've acquired a new business, what if you don't have the business name in there and then you've got another business deal doing business with the same sort of name. Server, a one and that's it. And I've got another server, I wonder, do you think I know what am I supposed to do? Well, what if you have the name ABC server or a one and they had the FF server or one banded together. You now know that ABC was from business ABC and DEF was from business df, have the business name in there. So the business end could be a combination of two characters or three characters depending on how you want to abbreviate it. But again, keeping in mind that if you want to keep this 16 character limit, just form your structure around that. Perhaps in groups of three, halves, in groups of two, I preference generally groups of three. So if you've got a business called ABC fruits, I'll use the name ABC as the very start three characters of my server name. The next part of my server name is the location or the site. If you have multiple sites, perhaps in different suburbs, in different regions, in different areas, in different states and different cities in different countries, whatever it may be, incorporate that bit into the name. So let's say for example, I am ion in Australia, so I am in Melbourne. You've got Sydney, you've got brings them. So my abbreviation would be mgL for Melvin, SID for Sydney or be our AI for Brisbane. So I would have a name of a, B, C, and E L. So a straight away know from the name, it's Company ABC in Melbourne in MEL. So that's really the first bit. That is the information about the company, right? So if the company and the location of the particular office or whether particular server is residing. If a server resides in the Cloud, you may want to have it as AWS or I, as I said, IZ you for a zoo or you may want to be able to change that, but generally try to have it as a location of the actual server, company name and location, but then move into the distinctives of the server itself. What sort of server is it? Is it a production server? Is that a development server? Is that a false server? There's more than one file said, What do I do? The next three characters really is the environment of the server. What sort of TR, It's part of now generally, in most IT spaces, there is what's called a 40-odd model. So a server would generally fall into one of these four tiers, would fall under development stage at dP, dV for development could fall under a test stage. Txt protests before and the staging STG all fall under production PRD. So depending on where that server lives, you'll put it into a particular TEA, into a particular phase. So again, if it's a staging server, I'm going to say ABC, MEL, STG to straight away. I know in server that it's from company ABC. It's based in Melbourne and it's a staging server. All right. It's not a production server which you could potentially power the server down. I could change the cell without too much complications. If I see PRD in the name, I know straight away that it is a production server. And I need to be more cautious of what I'm going to be doing with that server. We then move into the function of the server. So we now know that it's in a particular tier, in a particular phase production, staging Dev or Test. Now, what does the server and do is an application server? Is that a domain controller? Is that a web server? Is that a database server, whatever it may be, put that into the name. So if it's a database server, you may wanna do dB. If it's a domain controller, dc for domain controller or AD for Active Directory. If it's a file server, you may want to be calling it FL or fs file server or FL file in or something like that. If it's a What's another example? An application server. So some applications, it could be a dumpy. You still have characters available and you'll be more specific. You can actually maybe give the name of the application, APP or F if it's application of office or ABP, EX application exchange, the name couldn't, the type would be male, ME NIL. So the function could be very varied depending on what the function of the server is. You don't have to be that specific on this specifically saying that it's just DB. You may want to be able to give it a bit more information. Again, keeping under 16 character limit, it could be calling it Db, Db EX because it's the exchange database server, something around those lines, but making sure and function of the server is listed in there. And then finally, the number or the sequence. So in many organizations you will have more than one of the same type of server. If you have a file server and you're calling it fs, or if you have three or four of these FS service, it may be a bit complicated. Just have fs and then what are you gonna hold? An excellent, so good practice to have fs 01 and the next all be fs 2, 0, 3, 0, 4, and onwards, the same deal, db 0, 1, 0, 2. So good practice. Always have a number at the end of your server name. Do not keep that blank. Have db, O1, O2, O3, O4 as your sequence. So there you have it. That is my general recommendations around a server naming convention. To summarize, we're really talking about firstly, the company name, the customer name, whatever it may be, which would be two or three characters, followed by the location or the site where it's actually physically located in a state, in a city, in a location and an office, whatever it may be. You then may want to spit into their potentially in the inner spaces. Or you could have the location if you have a very large office or if you have multilevels having, let's say if it's MEL, one made, maybe, maybe familiar one office on level 1 or MEL n Maybe it's the Melbourne office in the north side or in the south side or on the west side. You can actually abbreviate that a little bit more if you need to. That is followed by the environment. So whether if it's in production, in staging, development, or in the test phase followed by the function with its database server, a domain controller, file server, and application server. And then the sequence number of O1, O2 onwards. 8. Understanding DNS: All right, so what is DNS? Dns stands for a domain name system and its integral for any form of computer systems, specifically for talking about computers and servers, to be able to translate a domain name into an IP address and vice versa. This is almost like central to the internet. All right, so when you go to an Internet website and you're typing in www.google.com and you type in into behind the actual domain name that you're saying that there's actually an IP address which points to the same location. The same way that it works out on the internet, works within your environment. So you set up a DNS server, or at least on a domain controller, you'd have a DNS service installed. And this now manages all of these translations between a domain name which is more like a human-readable. It's something very easy for people to remember people to understand. And that translates that into an IP address, which generally people don't remember, right? Some people will remember lots of IP addresses. I'm sure as you work more and more in IT, you start to remember and learn IP addresses. But for the most part, just knowing the domain name is the best thing. And then you let the DNS server doing the translating for you, translates it all from a domain name from a DNS to a IP address. When you open up a web browser, when you open up a command prompt, when you try to log into a server or even do some file-sharing, you want to access another server where there's some file stored. All of that will check DNS. It'll check some sort of a DNS register on your network, generally, as I said, hosted on some sort of a domain controller. It will then go, okay, my server name, ABC server 0 one actually points to this IP address and the guys goes and connects to that IP address. You can also go and type in the IP address if you remember that. But how much easier is it? Just remember the names of service and names of devices, the names of websites. Let DNS do its thing, let it go and convert it. So within DNS, you have a big catalog, essentially like a big phonebook of all of these records. But that is a brief overview around what DNS is. And it's something that is integral for really an IT person. If you want essentially become an IT professional or you want to improve on your skills in IT. Learning about DNS. Hello, students. 9. Servers vs Supercomputer vs Mainframe: What are the differences between a server, a mainframe, and a supercomputer? Let's firstly define what i server is. A server in reality is any sort of computer, but I can provide a service to a suite of other systems or other computers servicing a particular product or a particular software or application. They can host applications on a physical infrastructure, physical piece of equipment, or within a virtual machine environment. So just really think about a server as a standard out home desktop computer, but a lot more powerful. They have a lot more grant, a lot more CPU, a lot more RAM, a lot more graphics processing power to be able to run Up software and provide better services out to a network. You can actually install server software onto most computers, your home PC or your laptop. And it actually acts like a serve up, generally serve as run Windows Server Edition. So the current version being Windows Server 2019. You can also run Linux servers, different flavors or kernel versions of Linux. Or you can also run these in a virtual environment running something like virtualization technology hypervisors such as VMware, microsoft Hyper-V, or Citrix Xen server. So what about the size? What is a service size? What does it look like? If you're thinking of server hardware, specifically, servicing come in different sizes, different capacities for different form factors. You can get service in a standard tower setups, which is sort of like a desktop PC tau up, but a little bit bigger. And of course they packed full of a lot more power, a lot more resources inside of them serve as also come as Rec Service, which is, which is one of the more common ones that you'll see serve as they're essentially wreck base units, very, very large, long units in one IU to IU foreign aid, IU size units. You've got rials attached to them and then they attract and riled and screwed in a server cabinets. The other 4-factor is a blade server essentially can be different sizes. You would have combinations of these smaller blade service within Chelsea's, which then make up multiple pools of service to be able to give you certain services for your environment and for your network. So when you go to any sort of data center, a CMS room, a server, him, this is what makes up the majority of that space. You've got servers wrecked across multiple server cabinets, rack cabinets, along with other things such as network equipment and storage in server rooms. What about the cost of the civil? Well, a server really varying costs depending on how you customize it. Generally, a server can be customized the same way that you're a standard desktop or laptop can be with the amount of CPUs and RAM and processing power that you want to actually including the server. But commonly you'd find that service with starting the low thousands, working their way up to the high thousands, perhaps even up to 102030, $40 thousand, depending on how many CPUs it's got, how many caused the CPU has got, and how much RAM has been installed on that particular server. The next one is a mainframe. Mainframes generally, they've been around for a long time, so they're generally going to be a bit older. Not always the case, but mainframes had been around really since the 1950s. They have built, custom-built to serve a particular purpose. They're very different to modern day servers as what we know them now. They handle generally large workloads, large processing power requirements. These aren't commonly used today as much as they used to be, but they are still in use quite regularly in say, the banking and finance places, especially if you need some sort of dedicated device that requires a lot of processing power and a lot of data management. What about the size of a mainframe? Well, mainframes, they're large. They're not going to be the same size of the server. They generally going to be larger. Think about a full server cabinet. So a standard cabinet where you have multiple servers, one of those could be a standard size for a mainframe around the size of a refrigerator. They can be smaller. They can be larger. The processes inside and not your standard sort of processes that you may find in a server fleet of products. Some of these processes can be custom built and they're more of a niche market around the mainframe space, generally containing more than one CPU, sometimes up to a 100 CPUs within a single mainframe. Mainframes will not commonly run something like Windows Server, which is your standard operating system that you'd find in the service space. They generally will run some sort of a standardized operating system, perhaps something that has been built by the vendor. But you can also have custom built versions or kernels of Linux, as well as z OS, the applications or the services and the software that run on mainframes generally cannot run on standard, conventional service. They've been custom built for the operating system and for the hardware that is being hosted by that mainframe. As I mentioned previously, a lot of these mainframes had been around for a very long time. They are generally been phased out. A lot of organizations who do have them. I'm moving away from them, opting for servers that have a lot more grants power with versions of Linux that can provide the services that they, that they need, but they still needed in certain industries. Now the cost of mainframes is a lot more than a standard server. A server being in the thousands, a mainframe, and getting to the tens of thousands into the hundreds of thousands, really depending on who the distributor, who the manufacturer is, and how custom-built that mainframe has been for specific purpose in an organization. What about supercomputers? The term sounds amazing. What are these magical devices? Well, as the name suggests, these are super computers known as standard base computer, a standard based server or mainframe supercomputers really generally are in a league of their Arne. And similarly to mainframe supercomputers are built, custom built for specific purposes. They can be used for things such as research or for scientific purposes. Something that requires a lot of number crunching, a lot of smarts behind it, a lot of processing power to be able to crunch numbers and gather as much information as quickly as possible. You generally will not find supercomputers in your everyday business or organization. So places that can include these could be things like defense agencies, other examples such as university specific university's supercomputer laboratories that are specifically built for the purposes of research. A very common use of a supercomputer is for space research, going out and discovering what space looks like. Things such as AI, artificial intelligence, a supercomputers, internal pieces are insane. These can have thousands and thousands of CPUs inside of them. And that can even title up to millions of calls depending on exactly what the purpose of that stupid computer is. So this is just in a complete League of its iron and would blow away any form of standard computer at harm in a workplace, even a server or a mainframe. Supercomputers also generally run some form of Linux kernel, sometimes custom built for that particular supercomputer. So what are these things look like? What are the size of these things? Think about you're walking into a server room, into a data center. You've got rows and rows of what look like. They serve a cabinets. They cabinets housing multiple servers or multiple mainframes. Well, no, these are potentially one single supercomputer that looks like a pool of computers. What about the cost of these things? Now, similarly to serve as and mainframes, async need to be custom build. These things can take years and years to build a lot of manpower behind them to be able to build it, to design into project managing. And so these things can go easily into the hundreds of thousands and into the high millions, hundreds of millions of dollars for various types and designed supercomputers at this current stage, the world's fastest, most powerful supercomputer in existence. Simply computed that he's base adding Japan. It's called the fugato supercomputer. This space is insane. It is so big, it is sorry, powerful. It costs over 1 billion US dollars to set up and get all up and running. It is only recently taken the title of the most powerful supercomputer, surpassing the number two, which was the summit compete based out in the US, like many other supercomputers, this amazing piece of machinery also runs a kernel, custom-built version of Linux. 10. Your Turn - Summary: So we've covered a lot of material around servers. You now understand what a server is, how it works on the Cloud versus on-premise. We've talked about different server types also to serve as it can be building how service should be named. And we've also looked at things such as DNS and these other server technologies that are very foundational and elementary for anybody who is in the IT industry. But now it's your turn. It's now your turn to go and actually build some service. Go and put some of these foundational content into practice. Whether that be in your own home lab environment, which is what I've got, or whether that be in a business environment. You've now got the tools, the knowledge, to be able to better understand and build your own and manage and administer your own servers. But that's it. Thank you so much for letting me spend the time showing you and talking through this stuff with you. Please do let me know how you went. Let me know in the comments throughout the entire class how you've been going. If you have any questions, do reach out as well. But if you did find this helpful, please do share it with other people and also check out some of my other classes where we talk about a lot of other things around all things tech. Thanks again. We'll talk to you next time.