Introduction to Cloud Computing | Neil Anderson | Skillshare

Playback Speed


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

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

16 Lessons (2h 30m)
    • 1. 01-01 Traditional IT Deployment Models - On Prem and Colo

      9:01
    • 2. 01-02 Server Virtualization Simplified

      19:54
    • 3. 02-01 Defining Cloud Computing

      3:20
    • 4. 02-02 Cloud On Demand Self Service

      10:02
    • 5. 02-03 Cloud Rapid Elasticity

      5:47
    • 6. 02-04 Cloud Broad Network Access

      3:28
    • 7. 02-05 Cloud Resource Pooling

      11:10
    • 8. 02-06 Cloud Measured Service

      2:28
    • 9. 03-01 Cloud Service Models

      8:45
    • 10. 03-02 Cloud IaaS Infrastructure as a Service

      21:11
    • 11. 03-03 Cloud IaaS Infrastructure as a Service Design Example

      15:53
    • 12. 03-04 Cloud PaaS Platform as a Service

      5:12
    • 13. 03-05 Cloud SaaS Software as a Service

      4:48
    • 14. 03-06 Cloud XaaS Anything as a Service

      7:31
    • 15. 04-01 Cloud Deployment Models - Public, Private and Hybrid Cloud

      9:56
    • 16. 05-01 Advantages of Cloud Computing and Calculating TCO

      11:17
  • --
  • 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.

217

Students

--

Projects

About This Class

The 2nd top hit for 'Cloud Computing' on Google says 'Cloud computing is a general term for the delivery of hosted services over the internet.' Unfortunately, that definition is completely wrong.

This course is designed to clear up the many misunderstandings about Cloud Computing and to give you a crystal clear and easy to understand explanation of exactly what it is, how it works, the different options available, the advantages provided and how much it's going to cost.

And I'm not just going to give you the theory, I'm going to show you practical examples which you can follow along with yourself so you'll understand exactly how Cloud Computing works in the real world.

I start off with a review in Module 1 of the traditional deployment models of On Premises and Colo so you can understand how Cloud changes things, then explain how server virtualisation revolutionised data center design and paved the way for Cloud.

Module 2 defines Cloud Computing by listing the Essential Characteristics of Cloud services.

You'll learn about the Cloud Service Models of IaaS, PaaS, SaaS, and XaaS in Module 3. I start with an overview of the different models and then dive deep into the details and options of each. I also give lab demos of how to deploy each type.

The Cloud Deployment Models of Public Cloud, Private Cloud, Hybrid Cloud and Community Cloud are covered in Module 4.

Finally in Module 5 I list out the advantages of Cloud Computing and then break out the spreadsheet to show you how to do a TCO cost comparison when choosing between an On Premise or Cloud deployment for a real world workload.

Meet Your Teacher

Teacher Profile Image

Neil Anderson

Building your Cloud and Data Center career

Teacher

Hi, I'm Neil Anderson, and my mission is to build the best Cloud and Data Center systems engineers possible.

For the last 9 years my main focus has been delivery of technical training and development of course content for large enterprise and service provider customers such as NetApp, Cisco, Verizon and IBM. I train the engineers who design and implement the largest Cloud and Data Center deployments throughout the Asia Pacific region and the rest of the world.

I also provide consultancy support and mentoring for the services division systems engineers in the company I work for.

Prior to focusing on training I worked in the field for over 10 years in a variety of systems and network engineering roles.

I've always embraced change and I've worked in contract rol... See full profile

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. 01-01 Traditional IT Deployment Models - On Prem and Colo: high on Welcome to the Introduction to Cloud Computing Court. Invest First module I'm going Teoh, talk about what we hade before cloud because for me to be able to explain how cloud is different, you need to know what we had before. So we're going Teoh, discuss traditional on premise solutions Ock soo Cool location solutions and then in the next lacks in. I'll talk about server virtualization server virtualization. Give us a lot more flexibility in how we can deploy. I were daycare centers and it's actually one of the main enablers that mean cloud possible . Your first lesson. I'll talk about one playing on cool. So on Pam on premises first with on premises Solutions This is what we've had real eats and I t first begun with on premises all equipment is located in your building with you being the enterprise on all equipment is owned by you. There's clear lines of demarcation. Unless you've only got one office, you're very small company. You're gonna have multiple officers and you're going to want tohave corning to the aid between most offices. All of the equipment is in the offices is owned by you and it's your responsibility, but the network connections between your offices. That's going to be the responsibility of your network service provider To maintain those and make sure that they're up and you'll have an agreement with your network service provided, or about about all of the equipment in your officers has to be appeared for upfront. You've got about one time up front fee on it's a capital expenditure cost. New equipment will typically take at least a week to deploy because it needs to be approved by management. For the purchase, you need to order it. It needs to get delivered to your location. You then need Teoh on pocket on, install it in Iraq and you then need to configure it as well. So really one week is are pretty optimistic. Timeframe. Also, new equipment is going to need regular technology. Fresh is because equipment gets out of the If you look at the GP use that we have now, there are a lot more powerful than what we had five or 10 years ago. So you're not gonna went toe have the equipment you're going to need to replace it fairly regularly. Okay, so you're gonna want to consider redundancy is. Well, if you're running mission critical applications on your equipment, things that our view important for your business to function, you're going to want to make sure that it's always available. So you're gonna want to eliminate any single points of failure. So if one piece of equipment feels there's another one beers a back up on you waiting ahead of time. So what we're talking about mission critical applications. If you do have these, you're gonna want to make sure you go high availability. You're gonna want to make sure that the facility that your equipment is well kiddin is hardened and was actually standards for high pardoned have available. I were DEA Sanders are on the hire of a number of the tear. The more highly available media center will be starting done it the ball and layer. We've got a Tier one, which is expected to have a 99.671% availability, which equates to an annual dying time off 28.8 hours with Tier one. There's new requirements for a redundancy there. With a tier two D of Santer, it needs to be more available to an annual dying time of only 22 hours. It only requires toe have one path of power and cooling, so we don't need tohave physically separate paths. But we will expect toe have somebody done, etc. For some of the power components moving up a tear. 30 Dia Center annual dang time of only 1.6 hours on it is required to have multiple power on cooling paths, and it needs to support and plus one fault tolerance and plus one means. For example, if you had 4 January or if you require for January's to supply the acquire power, then you would put in five. You would put in a spear on with the power and needs to be able to sustain a 72 hour power outage from the grid and still be able to provide backup power from generators. And then we have the highest level. A tier four day of ST are has an annual dining time of just zero point for hours. Obviously, the higher tier, the more expensive it's going to be to build that facility on. If you did want tohave highly available facility and you're building yourself, that's obviously going to be a huge upfront capital expenditure costs. So a way that you can make this more affordable is using equal location facility, a cool location sent or cool. Oh is a data center location where the owner of that facility rents space to external customers. The facility owner provides the power. The cooling on physical security on the customer will put their own servers storage networking equipment into the facility. Independent cool providers such as quinyx offer customers multiple network connectivity options through a choice off network service providers that got canon connections coming into the facility and was network service providers will also typically Pier with each other in booze. Cool facilities. So of your big network service providers. Weight revising E, T and T. They'll have connections coming into the cool moves, and the oaks will appear in there as well. In fact, that's how network, according to the works across the Internet, looking at the characteristics of cool new solutions, your server infrastructure. From the point of view, the customer is, well, kid in an external cool new building. But it's just your servers that you're gonna happen there. Obviously, your users aren't going to be beast in the cool a building as well. You're still gonna have your all face where you're gonna have your own staff with for when dash talks. And you're going to require connectivity A from your office to the cool little facility so that your users get access to their servers, The cool provider ones the day ascent facility and is responsible for providing highly available power cooling on physical secured A according to the terms off this service level agreement that you have with him, you on your own server storage networking equipment within the cool move facility and the connections between your offices and the cool new are going to be your network service providers. Responsibility. You'll also have an agreement with them as well. More characteristics. Your equipment within the cool of facility is a cop ex cost. You still had to buy it as an upfront costs about capital expenditure, But the monthly cool hosting fees are now an operational expenditure difference between carpets and or picks. Carpets is a one off up front, cost on or pecs. Cost is an ongoing monthly feet. This makes having a highly available D ascent are more affordable because rubber and having a P for about huge big upfront Capex cost. You can pay it eyes, a monthly, all pecs cost. New equipment is still typically gonna take over a week to deploy because you still own and provide the equipment in the daycare center. So you're still gonna have to go through the same process again off ordering it, having it delivered, cabling it on, configuring it as well on again. You still own the equipment you're still going to have to up the regularly to keep up to D A swell. You need to consider redundancy for the hardware that you are. For example, if you've got mission critical servers and there were maybe gonna wicket clustering boys server so that was no a single point of failure. But the redundancy provide power and cooling is handled by the facility 2. 01-02 Server Virtualization Simplified: in this lesson, I'm going to discuss server virtualization, which is one of the mean enablers off Cloud computing, its allies for resource pooling, where multiple customers share the seam. Underlying server hardware on resource pooling is one of the essential characteristics off cloud. But server virtualization has been around for a lot longer than Cloud Computing house. So let's take a look at what we had before. Virtual ization. And this example. It's actually from one of the first companies I worked for my view first job in I t. We back in the nineties. So I was working for a fairly small company and we were looking after our own servers in a server room, so we had Iraq to put the equipment in on. We had a few servers in there. We had an email server which was running maker soft exchange on. We had a database server that was running maker Soft Sequel on we also ha Dato on Web server in there as well. Obviously we're gonna need power for both servers. So we had a couple off under interruptible power supplies. These provide power. Andi, Also, if the great powered was down, they've got bodies in there, so they will continue to provide power for a pdf tank. We're also gonna need switches in there as well for according to the E, between my servers and also for connectivity to the PC's in our offices. Rail on. We had some other officers, so we need connectivity to get there. We also needed connectivity to the Internet. So we had a writer and a fireable for not that I wouldn't a fireable they could be part of the scene piece of hardware or they could be sent. But it also you'd maybe when I have two of them in there for redundancy is Well, so that's what we had before virtualization existed. Hey, if we're going to get schematic, view this to the blue rectangle here about represents a single physical server. Done it the bottom level. We've got the hardware in the server. So we've got this CPU, the processor, the Iran mammy and the Nick the networking Orpheus car for our network corner to the directly on top. Off the hardware, we install the operative interest in one there and then after the opening system is being installed, we'll install our application on there. The first instance was our email server. So we're going to do that for our meal server, and we're gonna have the same thing for the DB server and Old Sue for a Web server as well . So it's see servers running on 30 physically separate pieces of hardware. When we do it like this, the server utilization leaves a lot to be desired. When I see a server utilization, I mean, how busy the CPU beram with Nick etcetera is the hardware that makes up the server. Typically, when you're running each single application or not server like this, it's going to be running it maybe around 15% utilization, which obviously is no very cost efficient, particularly as we had two p for each individual server under all using power, taking up rocks, peace and requiring cooling. So what I would like to do to get better utilization Is this so here again, I've got one server. I've got the hardware down at the bottom level, CPU there on and Nick. I've got my operating system installed directly on the hardware and then on not one server . I'm going Teoh, install my milk server application on my database server application onto my Web server application as well. So if I did this, obviously this is going to give me much bear utilization because Robert and having the 50 servers running on three separate pieces of hardware, I'm not running them all on the same piece of hard work's about one piece of hardware is doing more work. It's getting bear utilization, but putting multiple applications. One machine server is very bad practice because if I have a problem with any one of those applications, then it's liable to take all three days. I don't want to do that. So there is a way that we can actually get the higher utilization running all the servers on the scene piece of hardware, but without running them all on the CME. Instance that the same operating system and we can do that through server virtualization. So here the big blue square that is our physical server again again, donut. The ball in level. I've got my hardware resources, may CPU my Ron and my neck, and then what I do. On top of their night, Robert been installing a normal opening existent like windows. I installed my hyper visor on there. My high provides er is going to come from one of the server virtualization companies like VM Ware. So this example, Let's see him running the M. Where s X I as my hyper visor, the hyper visor ox Osby opening. Existen or not, Physical server on it, folks. It is another job, which is It allocates access to the underlying hardware to my virtual machines, which is coming next. What's ability it up? Teoh. I've got my first virtual machine, which gets installed in my hyper visor on In there. It's God's operating system. What say that virtual machine One is running on windows and then on top of windows. I install my email server on there. Then I install a second virtual machine onto that seem physical server on in my second virtual machine. It's also running windows, as it's all pretty insistent, but this is a separate instance. Off the operating system. It's a virtual machine one and virtual machine, too. The act like they're two completely separate servers, actually virtual machine one in virtual machine to they don't know that their virtual machines, they think they're just normal servers and then on top of Windows and Virtual Machine to I once stole my database server application in there and then virtual machine 30. What? See that this is actually running Lennix now. So even on the scene piece of underlying hardware seem underlying server. I can have virtual machines running when they're running different opening systems that work just fine. And then on top off the Olympics operating system, I installed my Web server application that so this is similar to what I wanted to achieve before where I'm running all fevers, different applications on the one box. So I get much better utilization. But it's still fulfills my security and my performance and my availability issues because voice for the virtual machines, the hot like they're completely separate servers. Okay, so that's the media that when she hey vis actually works with a wide damage. I've connected into my VM Ware lab here on day. I'm on the workstation which is connected into the V Sphere server, which is the monitoring application for monitoring all of my s ex I hosts, which are the physical servers on the virtual machines that are running on new servers. So if you have a look over in the left. You can see that I've got to. He s excited. Horse cheer. I've got tandoor $2.1.11 on I've got tanned up to $1. 12. So, actually, two different physical servers. And then if I expand these out, you can see that I've got four virtual machines that are running on 10 dot to dot wonder 11 . I've got a couple of XP machines on here. Nostalgia and open fail er on on my other physical host, which is 10 dot to 10.0.1 dot 12. I can expand it out and see, I've got a couple of virtual machines running on There is, well, it to show you how these are different instances running on regime underlying server. Achieve a click on XP one. I can slink the consul in there, and I will show me the screen of this actual virtual machine. So I'll click in here on dykan. See the log in screen for my first instance of Windows XP latched Sand control. Delete, Teoh. Allow me Teoh Morgan here on the island are my passion birds. And then click on OK on that walks me and you can see on the desktop of XP one. I've got a tax fail here, which is named one. If I know, go on Teoh XP to on Click in the Control. You can see that it is each separate console. It's a different dash stop. It's a completely different operating system, through XP one and XP to their completely different virtual machines that are running on the seam underlying server hardware. Going back to example, we had earlier this could be my exchange server on my DB server. While I'm in here, I also want to show you that killer feature off server virtualization, which is weaken my great. I were virtual machines from one underlying server to be different underlying server on Ba fly without requiring to shut them down and start them up again. Let's have a look how you do that. So I'm going to right click on XP two and then choose the option to my great on this will open up the Wizard. I'm going to change the underlying host. I click on next. I'm gonna move it from $10 to 100.1 dot 11 to 10 dot to 10.0.1 dot 12 so show act that and click on next and then click next again aren't finish. And if you look down in the bottom here, you can see the job window. You can see that it's doing the make great of a virtual machine. What it's doing now is to be able to do this. The virtual machine it needs to have its virtual hard desk on external storage. The hard disk couldn't have bean using the hard drive in tandoor $2.1.11. So the virtual machine it's got a virtual hard disk, which is where the operating system and all the applications are installed. That's on an external knives or sign started system. Then when I do the migration, what it does is it copies the contents of mammary from Dog 11 over to Doc 12. The actual virtual hard drive steers where there is on the external sign. It moves to the contents of mammary over to $10.2.1 dot 12 and then it cuts it over. And when it does this, it does it on before I It's transparent to anybody that's connected into that machine. You'll maybe miss one ping battle through the implications of this service is great if you're the servant, administer because if you wanted to take tandoor two don't wonder. 11 Done for Harbor me an inch. Then you can just move over virtual machines on the fly to a different Horst Oxo for redundancy. If we had, say, a power age on 10 dot to 10.0.1 dollar 11 we can automatically move all the virtual machines over to Doc 12. So this is a super useful feature. Okay, But everything I wanted to show you their lack school back to this whole age again through the paper visors. Assured you there. VM ware. Yes, X I that's actually e type one. Hyper visors to tapes was type one on tight to explain what the difference is coming up in a second. Our most popular tight one hyper visors are vm ware s X I. That's what I showed you the demonstration on. We've also got maker soft Taper V Red Hot Key VM Article V M server on Citrix Zan server actually include environments. Citric zand server is very popular because it is the cheapest option to use for the cloud service providers Give what a tight one hyper visor means is the high provides or runs directly on Miss System hardware, meaning the hyper visor is installed directly on the hardware and axes the operating system . And like I said earlier, it's over. Job is to give access to the underlying hardware to the different virtual machines. We also have tape to hyper visors. A tape to hyper visor runs on top of the host opening system. You'll see what that means on the next flight. I were popular type to hyper visors. We also have some of those from the M wear with VMware Workstation VM Ware player on VM Ware Fusion, which is for the mock workstation and player, or for a Windows PC. We also have virtual box Can you on part of Wales. So let's take a look at how the type to hyper visor works. I've got a schematic for you here again, and the big blue box is an individual host. This is probably going to be your laptop now. Take one. Hyper visors are used in the Deer center on tight to hyper visors are used on an individual pc. Again, I'll explain why that is coming up in a second, Let's will get how it works first. So this big blue box Ventura laptop on on your laptop you've got your underlying hardware like your GBU your mom on your nick again. And then on top of their we don't install the hyper visor when it's a tight to hyper visor directly on top of the hardware. This is your laptop, and you've got your normal dash top or pretty interesting on your laptops. That's probably gonna be running windows or Lennix our Mac, then on top of your open. Interesting. Because this is your normal laptop, you'll have your normal applications. Why make your soft office adobe actor by except running on there? Then we install the tape to high provides our eyes. Another application on top of our normal desktop opening system and then on top of their in the hyper visor will have are different virtual machines. Here we've got virtual machine one, which is our email server virtual machine, too, which is I were DFB server. I went to have a look at how tight to hyper visors work with another lab, Daniel. So I'm on my normal laptop here. It's the same one that I'm recording this video on Andi. If I open up Windows Explorer, if I can find it, here is I've actually got the tape to high provides are already installed on here. I'm using the M where workstation and I've built some virtual machines on here for a different course that I wrote. So what's going on? My netapp blob folder on? Do you see? I've got a whole bunch of different virtual machines in here. In which school? I'd open up clinics, E. So I'll fire up this virtual machine on this will open up in VM where Workstation on Alpi. Word on the virtual machine on you actually actually starts firing up from the bios level. You'll see linen actually booting up here. So there's Monex booting up. I'll go back to windows. Explode are on. Go up a level. You see, I've got ah, Winnie. Virtual machine here is overseas. Are Windows server on? I'll open this on powered on my windows server on. You'll see that this will start being up windows. So there is window is being up. I can see it in the control of my second virtual machine. I can flip back to the council of my first virtual machine, and you can see there is Lennox. So I've got my normal Windows operating system running on my laptop and on talker there, I've got a winex virtual machine running on. I've also got a windows Virtual machine. Running is well, show which nagel back to the slides and see where we would use a tight one and where we would use a tight to hyper visor. So, looking at the differences between them, I've got my type one hyper visor on the laughed. I've got my tight to hyper visor on the right, you see with a tight one hyper visor. The hyper visor is installed directly on top off the hardware with a tight to hyper visor. We've got a normal desktop or pretty insistent installed on the hardware, and then we have the hyper visor installed on top of our normal. Who asked a tight one hyper visor. This is used in E. D. Ascent are this is a purpose built bulks, which is dedicated for running dia center servers. We use a tight one. Height provides er because we have the least amount of overhead. We only have the hyper visor between the virtual machines on the underlying hardware. But this would not be very suitable to run on your own laptop, because you don't. We have virtual machines running on there. You can't do it. Run! This is ah, laptop operating system. Then with our tight too high, provide services what we are gonna have on our laptop. So I've got my normal operating system running on here. I'm running windows. I've got my normal applications running on top of that, which is what's allowing media record this video and also a what's allowing the issue of its Power point presentation. I've also got the hyper visor installed as an application, and I can have my different virtual machines running on top of their. So where this is really useful is if you're an I t engineer on, you want to do some testing on Lennix or Windows servers, for example. That I can through that as a virtual machine onto my laptop allows me to do view we convenient testing, and I don't have to go and find another hardware box to install Lennix or windows onto Okay , so maybe you're in technical seals on your selling a product that you'd like to give a demonstration off to your customers on it runs on Lenox again. You don't want tohave to carry that around to your customers. Offices with you is a set, but Harbor box. You can run it as a virtual machine on your normal laptop, So in your normal laptop, you can give them a power point demonstration. You can also give them a demonstration off the different operating system as well. 3. 02-01 Defining Cloud Computing: we covered our traditional I T deployment models in the last module. In this more job book, start taking a look at cloud and we're gonna define war. Cloud computing is now if you ask, the average man on the street to define card computing will probably tell you that it means where I t services are located in the cloud meaning there somewhere not on premises, but are cool. Facilities are off premises and they're not clothed on private cloud. Deployments that will talk about layer are often on premises. So we can't describe cloud computing by seeing that it's were the i t services are. Well, kid in the cloud or off premises. That's no correct definition, Teoh. Okay, boys, we correct definition then? Well, thankfully ever is actually a kneel down de facto or definition off what clothed computing is on. It comes from the n i S t. So what's have a look at that? Pdf now. So this is it. It's the n i s t definition off cloud computing on a nasty starts for the National Institute Off Standards on Technology, which is based in the U. S. It's only seven pages long, actually, the 1st 5 pages is basically a preamble. To get to the definition, I can go straight to Ph six on Dhere is the n I s t definition off client computing. So quite computing is a model for enabling ubiquitous, convenient on demand network access to a shared pool off configurable computing resources, for example, networks, servers, storage applications and services that can be rapidly provisioned and released with minimal management effort or service provider interaction. This cloud model is composed of five essential characteristics three service mortals on four deployment models. Okay, so I said that there was a de facto a definition. I didn't see it, but it was a very short and concise definition, so that's quite a mouthful. But actually, when we take this document is a hormone scroll down and you'll receive it vers essential characteristics on service, more goals on deployment models because the problem is that cloud computing, it can't really be defined in just one sentence. But when you take all of these things is ah whole. It actually does a view a good job of defining at this document, and it makes it very clear what cloud computing it to bear with me. I'll go through each of those three sections essential characteristics, service mortals on deployment models. I'll do that in the elections coming up, and I promise with Italian that we're done with that you're gonna have a full understanding of cloud. You're gonna understand exactly what it means. 4. 02-02 Cloud On Demand Self Service: the first of the essential characteristics that I want to cover is on the mind of self service. The definition from the n I s t is a consumer can unilaterally provisioning computing capabilities such as server time, A network storage as needed, automatically without requiring human interaction with each service provider. The easiest week to destroy versus to show you how it actually works. So let's have a look at Amazon Web services. Andi. I will automatically provisions e virtual machine running in the AWS cloud. So I'm here on the AWS control page on I'm going Teoh sign in with my account and then I'm going to click on the services button open The talked laughed on Choose the E. C to service that Stein's for elastic Cloud Computing on This is arms and service when I can prevision virtual machines. Sure, I'm going. Teoh, Fire up a new virtual machine. Incidents are quick on the instances. Top over on the left on. Then click on the launch instance Bond. This is gonna take me through a wizard where I can configure all the same things for my virtual machine. It with AWS, you cannot just sign up for a year's worth off free service. It only allows you to provision the war and servers, but it's still a really good way Teoh. Find out by cloud computing by practicing it yourself. So that's what I'm being You're not saying dot Put a year's worth of feet service on the first page for my virtual machine. Asked me. Well, opening system I want to install on there. I'm going. Teoh, choose Windows Server 2016 base, which I can She is included in the feet here. I'm not gonna have to pee for this virtual machine, So I select that. And then on the next page, it's gonna ask me how powerful I want this virtual machine to be in terms of how many virtual sea use it's gonna have on how much memory it's gonna have a swell again. I'm gonna choose with free option here, which gives me one virtual sheep. You on one gig off, Mammy. When I click on next to configure the instance, details in here asked me how many instances that I want to configure it. This is very useful if I want it to spin up five new servers at the same time with her seems settings. I could put five in here, and it would let me do it in one go. I only want one of them. I also specify high. I want the networking to be configured as well. I could set up different network settings ahead of time where I can have different sub nets . And then when I create my virtual machines, different virtual machines in different sub net, and that's going to control the connectivity between them again. I'm happy with just using the default submit in here. Another option. Dignity bought. Mr Tenancy. I can choose either shared or dedicated. Shared means that this virtual machine is going to be running on underlying physical server hardware that can be shared with other customers. So I'm customer a customer. Be could also have virtual machines. Running on may seem physical server underneath that if I chose dedicated, I'm gon anteed that it's only my virtual machines that are on this physical server. I would maybe do that if I had to four compliance divisions, but that's obviously going to be a more expansive auction, so I'll just choose shared in here and click on next for odd storage in here. It's going to default to giving me a 30 gig disk for my boot disk. I'm happy with that. I've also got options if I choose other types of storage of very the kind of performance I'm going to get for Pistorius to get, I'm just going to take the default here. I click on next and it's gonna take me to the tax page. This is for if you've got on wall of virtual machines, you can put different tags on your different virtual machines toe, organize them and you concerts for boys Layer. I've just got one. Virtual machines or tags aren't doing relevant for me. Click on next for the security group on. This is where I can configure the fire bottles. This is a Windows server. I'm going to need to be able to manage it to obey default. Remote desktop is allowed to come in. If I was going to run a Web server on here, for example, I would also add another rule in here to allow incoming Web connections on TCP Port. Ate it. I don't need to do that. For example, Shah again, I'm just going to accept the defaults here. I then click on review on launch. It's going to give me out so many page that shows me all of the different sayings. And then I can click on launch on Given next page. It's going. You prompt me to create a keep here if I don't have one own 80 I've got one already. But what's making you one? So I'll create and you keep here on. I'll give it a name. Let's just call this damn you on then Damewood to keep here. What vics is for is for when I initially connect to it the first time for management purposes, I'm going to need to use this security key. I'm sure you have this works and only or lesson when we actually connect into this virtual machine. So exited the fail. It's going to see if it into my Dame Woods folder that was a W. Ashes review important. Do not lose right. Fail because you can't get it back again. Then I click on watch Instances on it's going to create the virtual machine for me. So what it's doing in the background is it's configuring a virtual machine with all of those settings to on the server side, the compute side. It's creating a virtual machine with the amount of V C P use on the amount off memory that I selected it. It's folks are going to install the windows all pretty interesting in there for me. It's going to install that on E 30 gig boot drive, because that's why I selected for the storage, and it's going to configure my firewall rules as well. If you think about how this would work in a traditional deployment, you would have Teoh physically get the server racked up. And then this regarded different I t teams to configure all of always different. Xiang's actually rack up the server and stole We all pitying system. Do the patching on there and install any applications that would be done by the server team , Then to configure the networking, saying so the sub net that this virtual machine is going to be in on also be firewall rules still over incoming rdp connection that would be configured by the networking team on then for the storage for this virtual machine, that ther e gig boot disk that would have to be provisioned by the server team and attached to this particular server so that we'd all take a lot of time on it would all be done as individual manual tasks by the different teams. You see when I do it in a cloud environment, I just use a really convenient Web front end, and everything is done for me automatically in the background. This doesn't raise a ticket and then send it to somebody to do this manually. Behind this Web based front end all automation software at the bike and which is talking to my server systems to my networking systems on my storage systems on is configuring everything or Mike Lee for me. So the benefits I get from this is vist Server is gonna be up and running in west than 50 minute time on because it's old Donald automatically. It's much easier for my I t stuff. I didn't need Teoh have members of my I T department running around doing all of these time consuming jobs. So that's how we spend up a year virtual machine or on AWS wets go back to this lady aged to wrap up this last. Okay, so you so that we can have a virtual server or a much any other software service up and running in the cloud very quickly, usually in less than 50 minutes. Wentz compare that with our traditional on premises on Kool Moe deployment models. So in most traditional deployment models, it's usually going to take at least a week to get a server up and running. The purchase needs to be approved on the server ordered that then needs to be delivered. It needs to be physically rocked out and cabled the server team need to install and configure the opening existen patches Any standard software on the applications. For about server, the networking team need to configure switches, routers and firewalls. The storage team need to configure Historic Tristen on San switches Job by poltics. A long time, often several weeks that the lead time to get up and running will be quicker if your provisioning a virtual machine Robert on E traditional beer mail server. But it still takes time for the different teams to manually complete their tasks. So undermined self service is a big advantage of cloud computing it Let's just get our servers on. Our service is deployed much more quickly than we could with the traditional models 5. 02-03 Cloud Rapid Elasticity: the next essential characteristic to cover is rapid. You last to see the definition of this from the n I s t is capabilities convey Be elastic Lee provisioned on released in some cases automatically to skill rapidly outward and inward . Command Cherie with demand to the consumer, the capability is available for a provisioning often appear to be unlimited and can be appropriated in any quantity at any time. So let's have a look and see what this actually means you saw in the last lands in when we covered on demined self service that servers can be very quickly provisioned on up and running in around 15 minutes. About was for our virtual machines. If we're provisioning any other kind off quite software, VAT can almost always be provisioned very quickly as well. On Just a zit could be provisioned very quickly. It can be D commissions taken down Justus quickly as well, so we can very quickly spin up or decommission our services as and when they're required. This is what's referred to is elasticity. So if you think of that wake out rubber band when you need to scale out, you can do that easily and when you need to contract backing, you can do that easily on demand as well. On you only pay for services oz and when you're actually using them. So elasticity allows the cloud providers customers to achieve cost savings on this is often a core one of the main reasons for adoption off cloud services. Let's have a look at some examples of where we would use this first example. See that tan servers are required for e theme on projects with cloud services of the company can provisioned them within minutes p A small monthly OPEC's fee to run them robber than a large upfront Capex cost on decommission them at the end of the three months. If you compare this to what we would have to do before Cloud Computing was available, see that we're a company on a customer. Come to us with an opportunity, but we're gonna have to scale out. To be able to meet that opportunity. We're gonna have to buy 10 more servers. Well, we would have Teoh by those servers is a large up front capital cost on. It ended a few months when the project was done. We're going to be left with voiced servers again, we don't need to use them anymore, so it's really not cost efficient. And this may mean that were no evil a tick. Our opportunity with Cloud services because it's much more cost efficient were much more likely to be able to take that opportunity. And this concave us a competitive advantage over companies that were competing with the next example. Let's see that we're an e commerce store. We're probably going to get higher seasonal demand Christmas time. What we can do if we're using cloud computing is we can automatically spin up new servers as our demand is going up as well. So what we can do is we can monitor the load on the servers CPU, the man, many of the networking mood, and when it reaches a certain threshold, we can automatically odd new servers to the pool to help meet that demand. And then when the demand goes by doing again again, we can have another threshold. And when it goes below about financial, then we can start automatically Decommissioning servers. So this is great. It doesn't require any manual intervention at all. We can do this all automatically so those last two examples were for when we're using virtual servers and declared, Let's have a look at If we're using Ekl ABC Software service on I'll used still time d dot com for this example. What we can do here is we can use software for a short term and just pay for about short term Robert and buying a one off permanently rations. Let's have a look at showtime d dot com So I'm here on their Web page on what seals hunt. The is mainly is an email service, so I see that it's me that wants to use us. I've just completed recording. This car's on. I'd like Teoh all of my contacts, no so that they can go and download it. Well, I don't want a email, everybody, manually, individually, about with ages. So I'm going to use to kill time. Do you do it? But I don't want to use tools handy permanently because I'm not normally sending out a lot of emails to everybody. This is really just a one off thing when I've completed the cars. What I can do is I can use this second option here of irregular Robert than going for the long term, $7 a month e. I can pay a one off $9 fee, and I can just use it once. That's way cheaper. Then. If I wanted Teoh actually buy this offer permanently, that would be more expensive. No, it's that if I do take it one hour long term basis, I get a discount there at $7 Robert than Name doors, Joe aching. Get good cost savings on my software licensing with quite services. I could just pee for the services odds and when I need them. 6. 02-04 Cloud Broad Network Access: the next essential characteristic off cloud computing to cover is broad network access. As usual, will have a look at the n I S t definition first. So they see capabilities are available over the network on access through standard mechanisms that promote use by Hector Genius. Thin or thick client platforms, for example. Mobile phones, tablets, laptops on workstations. So let's have a look at what this means is well, first off, let's have a look our traditional on premise solutions without using cloud. I'm using the same example that I used before here, where we've got small company, we've got a rank out their headquarters, which is gold, are milks over our baby server on our web server in there. Nay, the also have other officers as well and also telly workers. So we're gonna need to have network connectivity there. Through I were company data center is connected out through e network service provider on we've got another branch office. The connection between the branch office on the mean say is maybe a virtual private network over the Internet. Or maybe it's a direct we explain between the two officers in the branch office. We've got our users working in there. Maybe some of them it on Windows PC's. Some are on Lennix on. Some are on mark doesn't do Emaar whichever kind of desperate they're using, they're all able to access the server's in the company Daycare center in our means site. We also have voice tally workers. I mentioned her maybe working from a hotel or working from whom? Maybe they wanna p She may be there on the tablet. Maybe they're on mobile. Doesn't really matter. As long as the services were running in our mean site, support was different types of clients. Ondas. Long as we've got network connectivity between everywhere been, everything's gonna work just fine. Okay, so that's how it works With a traditional on payment solution on the next light will have a look at how it works with a cloud based solution. So that's what looks like on premise. You ready for cloud? Here we go. There is It's exactly the seem Exactly. We see him network topology. The only difference is that our servers over here in our d ascent are Birnerova not clothed . The ascent are pervaded by a closed service provider robber than in our own Dia Center. So this fire is the network Corning to the Eagles. It's exactly the scene. So if you're a network engineer, who's going to be doing the designing as you transition from an on premise solution to a cloud based solution? This is great news because you don't need to learn anything. You you just do the network design exactly the same as before. It's just like before. The only difference is that the servers are in E cloud providers facility Robert and in your facility. 7. 02-05 Cloud Resource Pooling: the next essential characteristic off cloud computing to cover is resource pooling, but she had the n i S t defined this. The providers computing resources are pooled to serve multiple consumers using a multi tenant model with different physical on virtual resources dynamically assigned on reassigned According to consumer demand, there is a sensitive location independence and that the customer generally has no control or knowledge over the exact location. Off provided resources but may be able to specify location. Ah, higher level off abstraction, for example, the country state or D ascent or level examples of resources include storage, processing, ma'am Adi on Network Bind with So what's have a look at this in some more detail? Teoh. The first thing Teoh talk about that we can pool our resources for is the process er on the man money off the underlying servers that were running virtual machines on two went school back to I were hyper visor lobbed a mole for fish. So I'm back here in my VM Ware lob on in my management station, and this is similar to the kind of software that cloaked invaders would be using to mine age their horse on virtual machines. Maybe they'd be using the m wear. Or maybe they're using song upper vendors. High providers are like maybe Citrix Jane Server. So I'm using VM were for the example Here on you can see in my lab I've got to Horst 10 dot to dot wonder 11 on I've got another Horst, our daughter 12 and I'm clicked on 10 dot to 10.0.1 11 right now on on the summit E page And you can she in this physical server, but I'm gonna be running virtual machines on. It's good to process or sockets. Meaning it's got to physical sheep he used on two cores per CPU. Andi, I've got two gig Avraam in this server was real world in aerial Cloud Service writers Dia Center They're going to be using much more powerful hosting this subject school a little powered host for my wild demonstration here. So those cheap you on memory resources that can be divided up amongst the virtual machines that I've got running on Miss host. So I can see I've got three virtual machines on here. Open fire where one nostalgia to on XP one again. These are just a little powered virtual machines for my live demonstration that cook on open fail er I can. She is running with four virtual sheep he used on a little over 300 meg off Mammon E. Its function using storage is, well, it's got 100 gig of provisions storage that's going to be provisioned on My Dear Store, which is living on my external signs storage, if it look it in histology to this is just a really small virtual machine that can be used to run or dosh game civilly. It's for Daniel purposes on its Got One virtual CPU on its Got just 32 meg of memory. She can. She don't the underlying Horst server physical level. I've got the physical resources there and in my virtual machines were running on that horse . They can get access to the underlying physical resources. It's the job off the hyper visor to meet, sure, but the virtual machines get their fear share In our client environment. Very often we're gonna have different virtual machines from different customers that are going to be running on may seem physical server, and that's no problem. The cloud provider will make sure that we doing too many virtual machines on any single server, so they can all get good levels off Performance Oaks. Execute A isn't a concern either. The virtual machines air can't completely set, but on secure from each other. Okay, let's go back to the slayings again on the next resources. We're going to look at what we can. Pool is the storage. So in my example here, I've got a storage system which is represented by the big blue box There on. It's got lots of hard drives inside, botched or IT system. The hard drives are represented by each of these smaller white squares. With my centralized storage, I can slang shop my storage. However, I want Teoh on Give the virtual machines of our own small part about storage, for however much storage made require. You see the example here on my first disc, I have taken a slice of up and I'm going to give that amount of storage to the boot disk on 10 and one Server one with cloud computing. With the concept of tenants, E. Tennant is a different customer to customer, a would be one tenant customer be would be a different tenant because we can have multiple customers using the seam underlying infrastructure. It's e multi tenant system back to restore its just them again. So I had my boot disk for Attendant one server one almost gonna take a slice of my storage on provision. That is the boot disk for tenant to server one. So you can see, but having this cheered centralized storage, it makes it view efficient. Rather than having give whole discs two different servers, I can just give them exactly how much storage they require. Fervor. Savings could be made for you. Storage efficiency techniques. Such eyes thin provisioning where I can actually make it to the servers. Like the half goal more storage than I've got underlying physical disks underneath. I can also do D duplication where if I've got duplicate data on the storage system, I can just keep one copy of it, and I can remove a duplicate copies that gives me more started space on that can also do complexion where I can get rid off repeating strings or wait species and my dear, if you want to find out more about centralized storage sign and knives evokes ago another course on here Introduction, Teoh sign and nice. And that will teach you all of the basics that you need to know about storage. OK, moving on the next resource that can be pooled our our network infrastructure on our services as well. If you look on the slate here, you see, we've got a fireable up at the top about represents e physical firewall. Well, all of our different tenants are going to have firewall rules controlling what traffic is allowed to come in. You remember the example before we were all I wing rvp for our management trophic? If it was a Web server, we would allow Web traffic on port et a swell. Well, we don't need to give every single customer for one physical fireball we can share the seeing physical firewall between different customers oaks you If the require e would violence or see if they've got a server pool of your front end on incoming client connections Concave. Any one of those servers were going toward violence was incoming connections across the different servers using over with violence, sir, Again Via would violence, sir, can also be virtualized and shared between multiple customers. Oak shoe in the main section you don't know after she We've got multiple switches there on . We've got voters as well. Those switches on rotors are shared with traffic for different customers going through the same physical switches on writers Over on the right hand side, the clay provider here is also providing video services to the customers. Such a swindler's up the servers from the patching we can also patch Red Heart. From there is whale DNs. Except so Robert been having separate DNS servers for the different customers. We can provide the unanswered as a centralized service. So with all of these things, we have got a few underlying physical pieces of equipment and we're running multiple customers through village cheered pieces of equipment. Because we've got shared equipment robber and having a date, a k a separate one. To each customer, it means that we need to put last equipment in there so we get economies of scale there. We get bad efficiency. We get cost savings on from the service providers. Point of view we can pass was cost savings on to the customer, which makes it more viable solution. From the financial point of view, the last thing I want to mention in this clash in If I go back to the first slight and the n i s t definition you exceeding See that the customer generally has no knowledge or control over the exact location. Off pervaded resources, but maybe evil. A specified location. Ah, higher level of abstraction, for example, countries 30 or the center. So what we're talking about here, which used aws for the example again when I spawn out virtual machine I did it in the Singapore Dia Center because I'm based in the Southeast Asia region Knots closest to me. If I was based in the US, I would have chosen the newest location in the U. S. To do it to be doing that, having it close to me, I'm going to get normalised network late and see I don't get the best performance. So with AWS, you know the DS center that is in. But if we look at the last trade again, you see that within the DS center, I don't know which actual physical server my virtual machine is on. It could be running anywhere in not particular Dia center. It could be using any of the individual storage systems the AWS ago in there, it could be using any of the individual fire. Most the specifics of that really do admire to the customer there No important. As long as I know what Dia center it's in. The actual underlying server is irrelevant, really. 8. 02-06 Cloud Measured Service: the last of the essential characteristics off clothes as defined by the N I s T is supports e measured service as the STC close systems automatically control on optimized resource use by leveraging a meeting capability at some level of abstraction. Approved it to the tape or service, for example, storage processing bandwidth on active user accounts. Resource usage can be monitored, controlled on reported pervading transparency for booth, the PRA Vader on consumer off the utilized service. So with cloud services, typically they're going to be multi tenant. You're not going to just have one enterprise on there. There's going to be multiple enterprises, so the quote provider needs to be able to measure how much off the service each customer is using on build them accordingly. Billing will typically be monthly. On is an ongoing operational expenditure cost from the customers Viewpoint. The choice off piers you go or fixed monthly plans is often available. If you go for a fixed monthly plan, you'll usually get a discount in there compared to what you would pay for the more convenient P as you go. And this is going to be just a really short lasting because exactly What is billed depends on what service model is being used wherever its infrastructure as a service, platforms of service or software as a service. And I haven't covered what the different service models are yet. So in the upcoming lashings, it's actually the next more Joel, where I'm gonna cover the service models for each one of the different types of I E. S. P s on SCS. I'll give you a bit more details in boys individual actions about how the building were work for the actual type off service. So she in the next morning. 9. 03-01 Cloud Service Models: in this module. I'm going to cover the Cloud Service models. The N I s t defined three service models of higher cloud services Congee offered. We've got I A s, which is infrastructure of the service P A s, which is platforms of service on Ash E s, which is software. As a service. I'm gonna have individual rations in this module where we're going to cover what each of the three different service models are in detail. Nate, large cloud service Raiders are not going. It offered just one of these. Typically, they're gonna offer multiple models or service. Like have a look at AWS. Teoh much Do that. Now you have got to open in my browser Already. Click on services up here in the talk laughed. You can see the AWS have got heat off available services here. The have got all the different types of models. If we would make her soft, it's the same thing there. They've got maker soft zur for i e S p a s services on the got making soft office 36 5 which is a software as a service. So you might have smaller quote providers, mate Specialize in just one, but the big players will offer multiple different services. Okay, let's get back. Teoh the strange, the three models and define where the customer on provider it is off. Responsibility are on it wont Lambiel. The customer gains access to the service on the three models Build on top of one another. So we've got I e s bought M and N p. A s built on top of up. And then Ash A s would be the next level up from there as well through to explain. It's when you start off by looking at the DS and our stack at the bottom of the stock. We've got the facility of actual physical location on the power on the cooling, The next level up. We've got the network infrastructure. Sure, I were. Rorer's IR switches are fireballs, Exanta. The next level up is I were storage system. A baby. We were talking about physical things at these levels right now. Then we have got compute, which is I were servers then the hyper visor on top of there. So getting into the software level now, then on top of the hyper visor, because the opening system then we've got our applications on. Finally, I were data. So when we're looking at the different models were available for cloud. It's easiest if we look a day or sent our stock to explain this. So what I'm gonna do in the following Swain's is we'll talk about what all the different deployment models are. Starting off with it traditional mortals of on premise on cool Move and then moving into the cloud models on bond. We'll look at the data center stack on in their Wuxi. What is the customer idea of responsibility? War. They manage on what the provider managers so starting off with on premise, first on payments. Obviously, the customer manages everything. There is no cloud provider when we're using on premise. Moving on to e. Kool Moe facility and then this is a traditional model. Was does not count as cloud with cool. The facility is going to be pervaded by that provider on Villach Sumit. Sure that network connections are available, there is well the customer will negotiate e network connection from one of the network providers and then all of the other levels in the stock up there from storage. Compute hyper visor or pretty interesting applications, and their equipment is going to be owned by unmanaged by the customer. The next model to look Guys are first Cloud model, which is infrastructure as e service. You could see this differs from cooler with cool. No, we've got the providers facility, but the hardwood equipment in the facility is i words from the customer point of view. This is where it is different when we look at declared models with cloud. It's not just the facility which is owned and run by the provider, but also the hardware equipment is owned and maintained by the clothes provider as well. So the facility the network hardware started hardware of the computer hardware on the high provides or software is all a wound on maintained by the provider. The operating system also will be installed by the provider Onda. We will get access from the operating system level, so whoever patches your operating system, it depends on the particular provider. Maybe they do the patching for you, or maybe you have to do the patching. Everything above the opening system is going to be the customer responsibility, so the customer is going to be installing very one applications on managing their own theater. And if we look at where the customer actually gets access with I E. S, they will get access at the operating system labels. If it's a window server, the customer will be able to remote desktop into the the actual desktop, the or pretty insistent on manage everything up from there in the following lashings. I'm going to actually sure you this because it might still be a bit confusing right now. I don't want this lesson to be a huge Lassen. So I'm not going to give you a damaged creation of all of the different models in this one lashing. It would make it too long. I'm gonna break out into different lessons. I don't want if I es. Seems a little bit confusing now. I'll be covering that in the next election. I don't give you a demonstration of it. I don't. I'll make it really simple to understand. Okay. The next model we haven't This is the one that actually can be quite confusing. Is P A s platform Oz a service with platform as a service. The level goes up a bit in the ds enter Stack, so the provider is going to manage from the facility up to the operating system. The customer looks after the applications India, but really, this doesn't tell the story. I need to move on and slight because with platforms of surveys, what this is used for is for developing applications. So the customer will come in to a custom environment, which is on top of the opening systems. There's not like with aii Ash, where they can get access to the desktop level. They're gonna come into e purpose built environment, which is designed for building applications on again mentioned a bit confusing night in a couple of lessons. Time will have the PS slash, and I'll show you exactly what it means There. The last model is software Eyes E service on with software as a service provider is going to manage everything from the facility up to the D L level. What software as a service is cloud based software. For example, make your soft office 56 5 or salesforce that came to think so the customer is going to come in at the application level on they're going to use the application. Also, they can have it when data in there that everything is going to be managed by the provider . Okay, so that was just a quick introduction to the different models. In the next lessons that we've got coming up, I'll explain these in our lot more detail. I don't also show you how they work. 10. 03-02 Cloud IaaS Infrastructure as a Service: This is the first of our lashings on the cloud Service models on Dhere. We're going to cover I E s infrastructure eyes e service, Mrs How the n i s t define it. So the capability pervaded to the consumer is to provisioned processing storage networks and other fundamental computing resources where the consumer is able to deploy and run opportunity software, which can include opening existence on applications. The consumer does not manage or control the underlying cloud infrastructure, but has control over operating systems, started on deployed applications and possibly eliminated control of select networking components. For example, host firewalls. So with I E. S. This is the service model that gives the customer the most control. They get access down at the opening system level, but let's have a look it that way. Looking at the DS and or stocks, this is a remainder from the last lashing with I e s. The PRA Vader will manage from the facility up to the hyper visor and improve a durable. Also, install the opening system and me patches as well. The customer gets access from the opening system label so they can customize the operating system as they want the also installed the applications we went on there on will be working after their own data Joe, which have a work on she had about works in AWS. So I've longed into the damaging Web services console and clicked on the instances top and you conceive the virtual machine that I created in the earlier lash in. I'm going to click on the connect button on I'll get a pop up. First thing I need to do is to find out what the administrator password is, because this is the first time I'm connecting Teoh this virtual machine. So I'll click on the get password button, and then I need to browse to keep here. The idea mooted earlier. So that was in my dang words folder was that damn adopt pan file? I'll double click on that, Andi. Now I can click on the decrypt password button on. It shows me the password that was created for the administrator account through ongoing Teoh. Copy that into my clipboard on. Next thing I'm going to do is dang, would Buddha more desktop fail, which is going to make it easy for me to connect with Rdp or click on all came to Damewood . Got into my downwards folder can close Vish. And then I will go to my dang Woods folder There is the Rdp fail. I'll double click on that on Click on, Connect on And I'm gonna piece thin the password that I copied. Andi, Quick on. Okay, on Yes, to the warning message. And then this should log me into the desktop of the virtual machine that I created on there . I can see my virtual machine is ready. I'm on Modesto now, so I'm in windows on. What I would do now is I would install whichever applications I wanted to use this virtual machine for. Another thing that I would do at this point is also change the administrator password. There's a bit that misconception where people think that if you go on infrastructure as a service that the clothes for Vader all tool have access to your machine. So it's insecure that that's not the case. All its best practice. That first thing you do is change the administrator password, and then it's only you that has got access to the desktop off your virtual machines. The provider does not have any access at all. So it is a secure solution. So you can see with infrastructure as a service, the provider is pervading the underlying infrastructure on the installed. The opening system for me. I get in the operating system level at the desktop, and I can do anything I want with the virtual machine from there. Okay, it went school back to this range again on we'll have a look at the most. Well, no one i e s providers. You can see we've got Amazon web services up here in the top, right? This is the Gartner I. E s magic quadrant that we're looking at now. Gartner wrote a search company on the research. You are the biggest players in cloud services amongst a whole heap of other things. AWS are by far the biggest cloud provider there actually currently bigger than all of their competitors combined. However, make their soft as you are. Are gaining market share Been able to do that? Because it can be very cost effective option for their customers because pretty much all companies are maker soft customers in some shape or form. Probably using window is eyes. They're destined opening system so because of that already, make yourself customer. The king get cheaper options for using maker soft workload as well. But it'll be s are still by far the biggest player with I e s. Right now you have a week in the bottom section is Well, you see Google also pervade cloud services. Other well knew in providers are VM ware have got the ruined surveys. Also IBM software on the rocks, please. Okay, There's different flavors available for I es quite Providers will often offer 30 of these. The first is virtual machines on shared physical servers. We also have virtual machines on dedicated physical servers On results will dedicated bare metal physical servers. If you're an I. E s customer, you don't have to choose one of the three you can mix and match between the three of them. What's have a look at what boys? Meaning a bit more detail. So the 1st 1 was virtual machines on shared physical servers. Here, different customers can have their virtual machines. One Besim shared underlying physical servers to customer A could have a virtual machine on physical server one on customer be could also have a virtual machine on that seeing shared underlying physical server. This is the waste expensive option because you're using shared resources to its most cost. Effective for the provider they can possible is cost savings on to you as well as the customer. Typically, it's gonna have the least amount of options in terms of how many V. C. P. Use rahm on storage sayings unavailable for the virtual machine. I have the three possible flavors. The virtual machines can usually be provisioned more quickly, then the dedicated options. These can usually be provisioned very quickly, typically in Western 15 minutes on, because it's the least expensive option. This is also the most commonly deployed auction of the 30 as well. Next option we have is virtual machines on dedicated physical servers. Here, the customer is God. Indeed, the underlying physical server is dedicated to them. So if customers have got a virtual machine on physical host, one new other customers are gonna have any virtual machines on that physical host. One, the physical host is dedicated to them. This is a substantially more expansive option. Been virtual machines on shared physical servers because the provider has dedicated physical hardware to that customer it's going to be a more expensive option. There's typically MAWR options here in terms of high. Many V C P. Use rahm on storage options out available for about virtual machine because the customer has got dedicated hardware for them. The may be required to saying a minute and went contract for this, but not necessarily. It depends on the particular climate provider. The last of the three options is dedicated bare metal servers. With these, a customer is given access to their own physical server. Dominate the lower server Landel. I'll explain what that means. A bit more in the next late hyper visor is no installed and managed by the cloud provider. The customer can either install an operating system directly on the server or vacant, install and monitor When hyper visor. This is the most expensive option. Office 30 and it typically has been most options in terms off Virtual CP use Rahm and storage options that are available again. The customer may be required to sign a minimum length contract on AWS for the biggest Chloe I. E. S provider. Do no offer this option with either, Will you ask, You only get the 1st 2 options. Currently, you can't get a dedicated bear mail server with them. So looking at the data sent or stock where the customer gets not is diffident if they're using dedicated bare metal servers, was quickly have a look back at how it works with virtual machines. So with virtual machines, the provider is going to work after everything up to the operating system level. When the customer gets in on the open interest in, ah, level at the desktop and they can do whatever they want up from there. That is how it works with the 1st 2 I. E. S believers, the virtual machines on shared physical servers or virtual machines on dedicated physical servers with dedicated bare metal servers. Fool, it's different. The provider is going to manage the facility up to the compute level. The customers get in at the compute level, too Big A and B move the level of the operating system. The operating system is actually no even installed yet, so the customer will get access using some kind of management application. 8 p.m. I. Or late down to the physical server. I'm they're able to install the opening system from there so they can install any open interest in the one that could install Windows on their it could install when extradite. We want the hardware. Or we could install that on hyper visor like the M Wear or Citrix Zane server. The choice of the opening insistent is up to them, so I provides. That is optional. Maybe the institute have a hyper visor. Or maybe the OS isn't stole directly on the hardware. If the customer wanted to run just one workload or not particular physical server like let's say they're going to run an Article D A base on their they want tohave. A high performance server on the article database is going to be the only thing running on the server. In that case, they would install the operating system directly on the hardware. They wouldn't put a hyper visor in there because it's another layer. It's another thing that can go wrong. And it we'd odd overhead as well. We're gonna want the best performance for that particular work. Would it would install it, evidently on the hardware. If, however, they want it sit on multiple workloads or not physical server, then they would install hyper visor on it on, they would run their workloads as separate virtual machines moving on. The most common reason to choose virtual machines on dedicated physical servers is for compliance. The customer me have son kind all regularly, a requirement that means that they can't have virtual machines on shared physical servers. Dedicated bare metal servers will also fulfill Besim compliance requirements. Both of these options require dedicated physical servers for the customer, so the cost is typically similar, with bare metal servers maybe being a little more expansive, a reason E customer may prefer virtual machines on dedicated physical servers. Out of these two options is, if we do know, have expertise in house to manage the hyper visor. Looking back again, you see with dedicated bare metal servers if they're going to be running a hyper visor on their than they need to install it and manage it themselves. So we're gonna need veteran staff who have got expertise in that area. If, however, the guru with virtual machines on dedicated servers, then the provider is going to install and manage the height provides air for them, they just choose the operating system. The one when we spend up the virtual machine, and they get in at that level. So that's what their option if they don't have I t staff who have got expertise and server virtualization moving on again, just like the customer may be offered options for shared or dedicated servers. They can also be offered auctions for shared or dedicated network infrastructure appliances like the firewalls and the load balancers. Again, it depends on the particular cloud provider. If we're going to offer his options are no. Customers can typically connect into the cloud provide Rose D ascent or over the Internet on door via a direct network connection with the storage options, customers will typically have the option of local hard drives in the server or external sign or not started because there are actual will often have the option of managing their own storage operating system on a virtual machine or be our mail server with I E s. The customer gets into the operating system level so they could install some storage management software in the opening system and look after their own storage. The most common reason for doing this would be if they want to look after their own encryption. The customer me also to be able to install for a one whole physical started Justin in declared providers DIA Center. So maybe they've got started system from a company like now. Or am she? Maybe they can actually install that in the D ascent are on. Connect their servers into that again. Wake everything here. It depends on the individual cloud provider. The customer can manage for servers to install applications and patches accepted a through standard removed management methods, such as remote desktop for Windows servers on secure shell for Lennox you saw earlier in this last year. When I give you the Daniel and A P, I is also typically available to a loafer automation of common tasks such as provisioning, a new virtual machine. The customer may also have the auction off applications such as Microsoft Sequel or Auntie Vega's. They can either install the application and, after delay syncing themselves, which would be a capital expenditure, or they can have the cloud provider do it for them, which would be an operational expenditure. So what I'm talking about here is you know that the customer get in at the opening system level so I see that they want to run sequel server under virtual machine. They could install sequel server themselves, in which case they would have to prove a delay since or if the provider pervades this auction, the provider can install sequel for them on then the customer just peas for delay since is a monthly fee to the service provider. Robert been pervading the license themselves the cloud provider me Oaks who offered to manage the application as well they may have in our example sequel db Saad Mons on the provider staff who can look after the database for the customer. Okay, let's have a look at Harry Billing works with I E. S for virtual machines on shared physical servers. The cheap you and Rahm will typically only be build when the virtual machine is powered on the physical cheap. You and Rahm and the underlying server hardware will be available for use by other customers when the virtual machine is powered off. So the provider isn't going to charge you as the customer receipt. You and Rahm usage when you're not using it because you've pared to serve it off so you can get some cost savings there. Network Bonwit will be billed as it's used. Some use IDs will typically be bundled in. If you've got a monthly plan, data storage will typically be billed. Weber of a virtual machine as powered on our off as Vidia is always going to be there and taking up physical started space. Optional software extras, such as a Windows operating system or sequel server, will be billed as a flat monthly fee. If you've got Lennox in your virtual machine, Will Minutes is a fi? Ah, pretty interesting. Silver's new additional charge for it up that if you want to be running windows in your virtual machine, then there is a fee for about opening interest in so the service provider will include it in the charge. Okay, let's have a look at some examples of billing. Will have a look at AWS first, so I'm here on the AWS simple monthly calculator, which is a tool that you can use to estimate what your monthly charge is going to be every month. First thing we do is up. We talked. We choose your region because was slightly different charges for different regions. I will choose to sing a port in here and then under there you odd your virtual machines that are on shared servers. So I click under possible and and in select the tape, so went See that we have got a virtual machine which has got four V CPU cores on 16 gig of memory that should take t two extra large in Amazon. Select that and clues and shave, and you can see that the monthly fee for a vist server if I had powered on ah, 100% of Italian would be a little over $175. You can also select your other options in here as well. For example, let's see that for about server, I also required 500 gig abates off SST storage. I can improve that on you. See up top here estimate of the monthly bill, $175. When I enter my additional storage, that cost will be added as well. That was one of the another $50 per month further down you can on all of the other options as well. That's how you can figure out your bill on AWS. Let's look at another example. Have a look at Telstra's placing structure. So I've opened up their place engaged, which comes as a PdF Telstra are the mean telco in Australia and they also offer I e s services. Tell Strub offer bulls virtual servers on dedicated servers. Israel have a look at how the billing works for virtual servers on shared underlying infrastructure. You can see that Telstra use a monthly plan structure. So if you spend $200 a month with them, you get the extra small plan where you get to Vichy pews and four Gig of Rahm. If you go up to $4000 per month, you get 64 CP use on 256 and given no these VC views and Ron that can be divided up amongst multiple different virtual machines. Shia Queued, for example, have eight virtual machines with it cores each or you could have 60 in virtual machines with four cores each or however you want to mix and match. They also have the piers you go applying as well, but you get a bit of a discount if you go into one of the monthly plans 11. 03-03 Cloud IaaS Infrastructure as a Service Design Example: in the last election. We covered the details off infrastructure as a service in this lash in, I want Teoh show you the basics of how to do an I e. A s design. But he's not going to do this is because if you are moving from purely on premises to cloud solution on you being tasked with doing the design, Miss consumed a daunting at first. But it's actually really simple. Andi, I'm going to explain why, as we go through this lesson, it's because designing and I es solution is just like designing and on premises solution, which is accessed from office that uses exactly the seam data center design prints. Suppose it's just that the data center hardware is in the Cloud Providers facility instead of in yours. The harbor components that you're gonna use ever seem the way its own network together is we seem that way. It's access to the scene on the way it secured is also the same if you remember back from the broad network access slash in, we covered how the network looks like in a traditional on payment solution when we're accessing our servers and the company data center over on the left year from branch offices over on the right or from telly workers working from a hotel where? From a home, for example. So that's how it looks like that. So we do the network design for the traditional on premise solution on how the network looks like for a cloud. I s solution is this It looks exactly the same. The only difference is that the servers are now in the cloud providers. Theus sent our robber than in our data center. So for doing the design, really, we do the design just the same way as we've always done it. Traditionally, wets have a walk through of doing this on in the example, I'm going to use a pretty standard three tier e commerce application. So we've got the web servers front end that customers are on the Internet, are gonna connect into, and then trophic goes through to our middleware applications server, and then we've got a d A B server at the back end. So first thing it considers what are we going to do for compute on storage on? We need to figure out why. Divorce free flavors off i E. S What are we gonna use for the different types of servers are front end Web servers are going to be picky. Lewin servers. I don't need a lot of power there. And also, I don't have any kind of compliance requirements. So I'm going to go for virtual machines on shared physical servers because Ratzmann was cost effective option for the middleware applications server. Let's just say for this example that we've got a compliance requirement where we can't run both servers on shared underlying physical compute infrastructure. So we're going to need to put the always on dedicated servers. So for about accuse virtual machines on dedicated physical servers and then the last one is the server at the back and on. For this example, we need a powerful server for the database server. We're gonna need that high amount of Vichy pews on what of Rahman there. Wilson might have mawr stringent storage requirements as well that aren't available on virtual machines from this particular cloud service provider. So not case. We're gonna put our backend DB servers on dedicated bare metal servers get so that's the compute taking kids off made boys decisions. The next thing to consider is What are we going to do for the storage for the front end Web server on for the middleware applications servers. We're gonna have multiple levels. Servers, but their organ It have exactly the seemed content on there. So it seem content. We're going to put them into a server farm for bull tapes of the two different servers. Easiest option we're gonna have for storage there is to use sign storage for them for the backend db server. Lancisi of it for this example. We have got high performance requirements for the storage as well. We did a certain amount of I Alz there. So in that case, we're going to use local discs in not dedicated bear mail server to get the highest possible stories. Performance gets the back tire. Compute on stories, decisions taken, kid off. The next thing burger. No, Wolcott is the networking. There's quite a lot to consider on this slide here. So with this three tier e commerce application, traffic is going to come in from external customers over the Internet. It's then going to hit. I were front and Web servers. Where the customers will be able to drive is our catalogue and be able to pay Putin things into their shopping cart. From there, the traffic then hit the application server middleware on. From there, it goes to our database servers at the back end to watch the traffic flu. I'm gonna have a firewall in front off my web servers to make sure that traffic can we come in his web traffic on poor 80. I'm also gonna have a Lord violence or here as well, Because I don't just have one web server. I'm gonna have more connections coming in than one server can handle. Also, I don't want to have a single point of failure, so I'm gonna have multiple Web servers, but all identical copies of each other. But good to see and content on. I'm gonna put them in our server pool on. I'm gonna have a Lord violence or in front of them. It's going to bow into the incoming connections to the different servers that are in my server farm. I've also got a global would violence from the outside here as well. I'll talk about what that is there for when we talk about disaster to cover the layer. Okay, so I've got my firewall on my wood by Winsor in front of my front end Web servers. Then my application servers. I'm gonna put those into a different sub network because traffic should never hit the application servers directly from the Internet. I'm gonna have a fireable in front of them, and they're going to be in a different sub net. And traffic is only going to be allowed to get to the application server if it's come from the Web servers and it's coming through on the cadet port number. So I'm doing that to secure them again. I don't just have a single application server. I'm going toe have multiple servers there to handle the volume of traffic on also because I don't want a single point of failure again. I'm gonna have a mood violence, sir, in front of my application servers to ward balance foods, incoming connections to them, then at the back end from a database server again traffics, you know, hit the database servers directly from the Internet or from the Web servers of gonna have a fireball in front of than I put him in a different sub net on my fireable rules. alive traffic from the application servers on the correct courts. I don't have a load violence or in front of my database servers because for this example application, but it's handled within the application itself. I'm gonna have at least two database servers because I don't want to have a single point of failure. Other things to talk about here. The server farms can be automatically scaled with rooibos Web servers on the application servers. But identical, they've got exactly the same content on them, so I can build an image of those ahead of time. Then I can configure a fresh world where I see that if the wood on my existing servers goes above a certain level, I'm going to automatically spin up an additional server, an odd to the server pool again. The Lord violence terrible. Add it to the servers that it's going to be sending the incoming connections to. So this is great. I can automatically scale up and scale down, and if servers I have in line with the current demined with the traffic will we discussed there but was for traffic coming from external customers to do their shopping. We also need to consider management traffic as well, because no one I t engineers are sometimes going to need to get one of those servers to do me in. It's so for incoming management connections. I were engineers can either use a begin a virtual private network over the Internet, or we could set up a direct connection from our office into the Cloud Providers facility. Next thing to talk about is back up, so we need to consider these again the same way as we would with an on premise solution. Super importantly, the cloud per Vader will know automatically back up your data. This is another bit of a misconception. Are misunderstanding? Some people have a bit cloud to think if they have their servers deployed is a cloud solution. It's in a hard and DEA sant nervous. No single points of failure. Backups will be automatically taken as well. That is no. The kiss, the service provider is not going. Teoh back up your day are by default. You need to provision that the data center is, ah, hard and facility with no single points of failure. If you've designed your solution like that, but that doesn't protect your against regional disasters of the entire day ascent or going down or the corruption If we look back at the PV slide, you see, with my database servers here, I've put two of them in there for redundancy. But if my DEA gets corrupted, it's going to get replicated between boom. It's going to be corrupted on both servers, so having to servers isn't gonna help me. I need to take backups in case I need to do a restore from a previous version. You have network connectivity to the clothed facility. It So one of the ways you could configure your backup is you could back up back to your on premises office and use your existing backup solution so you could back up to tape in your office, for example, if you want to, you can also back up to the clothes elevator storage. If you are going to back up your DEA to the clay providers started, make sure you're backing up to a different descent are been where your servers are like it again. We might have that regional disaster. If we lose the entire data center, it's not gonna help us. Much of our backups are also witness India Center. There should always be brought up to an off site location. Next thing to talk about is disaster recovery. If the DEA center isn't lost, you'll be able to recover to a different location two different Dia center from those backups as long as they were stored off site. In that kiss, your moves all new data since the last backup was taken. So we're talking about AARP. You'll hear the recovery point objective. What AARP you'll means is in the worst case scenario, how much data could you lose if you have to restore to a different location? So, for example, if you're taking by cops every night, your AARP you'll would be 24 hours because worst case scenario would be you have a disaster just before you take the next back up. So all within you DEA that was written today, since the last backup is going to be lost. Best case scenario would be that the disaster occurs just after we taken the backup. But when we talk about our Pulitzer worst case, an idea we talk about. So if you're if you're recovering for backups and you take a back up every day you're our people would be 24 hours. It could take a significant amount of time to deploy the infrastructure in the new location on restore body as well. So just like we've got the AARP you'll but a company point objective. We need to consider that a company time objective as well. Using are seeing example again what she we're going to just restore from backup. So our AARP you'll is 24 hours. But when we do feel over to the new location, it's not like buying weaken. Just click our fingers and everything is going to be back up and running. We're gonna have to do the restore, which is gonna take time. We're also gonna have to deploy our new servers, but also going to need to configure I were fireable rules configure. I were would balancing, except that's all going to take time. So the art you'll is going to be how long it takes to get back up and running again. It's no Aguila. Calculate this like it is with our peel. The rt overview we to calculate that you need to do a task recovery should you have taxed feel over in your sight and see how long it takes you to get back up and running again. Okay, So if we are just restoring from backup so you can see there that AARP you'll and art you is going to be quite long and not might not be acceptable. You may want to provision a disaster recovery solution to reduce the AARP. You on our to you. So that's what we're doing here. You can see over on the left. This is the same quite solution that we deployed on eighties. This is with me in sight here. Customers are going to be coming in over the Internet on we're going to be heading. I were free tier application in the main Claudia sent are on the left, but we want to have a fast disaster recovery solution available as well. So what we're going to do for dies in a different Dia center are we are going to prevision Web server application server on database server on configure our Lord violence er on fire ball rules as well. So we're gonna have infrastructure oddity shelf ahead of time. So if we do have to feel over, this is going to give us a fast arc. You'll because we're gonna ready to feel over when we need to. We're also going to need the data to be available and not disaster, the company say as well. So we're going to need to replicate the from the database servers on the left, in the main site to the database server in the D R site. I don't need to replicate my Web servers and applications servers in this example because they're just using started content so I can deploy these from images. The last thing to mention here is my global load balancers. They're there to direct incoming connections to the correct DIA Center. Normal operations. Incoming connections will get directed to the main site. If the mean site goes dying, I will feel over to the D R site on my global would violence. Terrible direct new incoming connections there. You only need the global would balancer if you've got a disaster recovery solution. If we only hide IRA service running in one site, we wouldn't need dot component. They Obviously, if you're going for this kind of disaster recovery solution robber, then just back up, it's going to be more expensive because you do need to deploy additional infrastructure in the disaster recovery site. But this is going to give you reduced AARP You'll on our to you. We're typically not going to deploy exactly the Siem infrastructure in the disaster recovery, say is in the mean site because this is just a backup. Say, well, just put minimal infrastructure in there to give us the most cost effective way of doing this. 12. 03-04 Cloud PaaS Platform as a Service: in this lesson. I'm going Teoh cover the next off are clothed models, which is P A s platform Eyes e service. And as usual, books start with the N I s t definition P A s is very capability pervaded to the consumer is to deploy onto the cloud infrastructure consumer created or acquired applications created using programming languages, liberties services and to supported by the provider. The consumer does not manage or control the underlying cloud infrastructure, including network servers, operating systems or storage, but has control over the deployed applications on possibly configuration sayings for the application hosting environment. So you're looking at high bish mops to our d ascent or stock again with PES, the provider is going to provide all of the underlying hardware infrastructure on the underlying operating system and then the provider will provide a custom environment which is used for building applications. And that is very Webo, that the customer gets in act. So the customer is going to be managing the applications that they build and any DEA on top of up the point off p e. A ash is to make it quicker for customers to be able to get their applications into deployment because we don't have to worry about building out the underlying infrastructure on supporting that, they can just get straight into this custom environment where they can quickly build our applications. Also, with this custom environment, it will have videos, plug ins there that make it easier to build the applications so it foot example. See, the customer is building an e commerce application. There will be modules in the PES environment for things like adding a shopping cart or adding life chat, that kind of things to the developers. Evil would just prove at ST in without having to do all recording themselves from scratch. So it makes it very quick, very convenient, on very cost effective. The popular PS providers we've got AWS have got elastic beanstalk Make us off. There's your pervades. Bull's eye es on P A s. Google Lapses. Another example. Silks forest force dot com on IBM Blue Mix with songs. Forest Just first have traditionally Beanie software as a service provider that devote soon moved into the PES species Well, nay, with for Stockholm for billing for platforms of service, it will typically be billed based on mammary usage Let's have a look at the IBM blue mix pricing structure for an example. So I'm here on the blue mix placing page for school. Dana Little. You see that there is a free trail available where you get up to two gigs worth of run time and container Manatee, or you can go with the P as you go plan. You only pay for what you use were thought. You get half a gig of freed on time and container mammy for unsold Sewell BDs subscription plans as well. So actually, the placing works. Let's have e quick. Look, how it actually works is well for using this. To be honest with you, I am I dear, sent our engineer by trade. So I know compute. I know server virtualization, networking and storage. I can talk about what subjects all day, but I'm really know an application developer. So I'm just going to give you a quick overview of this. I've already signed up for bid believe Mix trail, so I'll click on the log in button here on because I'd already entered my credentials. It's taking me straight to the creates peace pop up. I've already added my organization is flat Box is telling me I don't have a space in the U . S region just after me to enter a name here. I'll call it Flak Box damn or on Click on Create. Then I get the option to create an apple click on that, I'll choose a cloud floundering out, and you see, I get choice of the boards of different programming languages so I can use Java nor GS. Payton, etcetera. They went to the containers pH underneath. There you'll see that Docker is also supported. If I go to watch and you'll see some of the services that I can pull in, for example, I could pool in text to speech or vice versa, speech to text that can also do personal a insights there. I'll get back with the There was a quick overview off PS See you in the next lengths in for software as a service 13. 03-05 Cloud SaaS Software as a Service: in this western. We're going to cover the last of the N I S T defined cloud models, which is S E s software eyes E service. This is going to be a a quick Klassen, because S A s is simple Onda. Also, even if you don't know already, you're almost certainly a software as a service customer yourself already. You already know how it works. As usual will start with the n i s t definition. The capability pervaded to the consumer is to use the providers applications running on E cloud infrastructure, the applications accessible from videos. Client of A sues through either Ethan Klein interface such as a Web browser, for example, Web based email or a program interface. The consumer does not manage or control the underlying clothes infrastructure, including network servers, all paying systems storage or even individual application keep abilities with the possible exception of limited user specific application configuration settings. So I'm looking at our data center stack. This is the opposite of the on premises solution with software as a service, it's the provider that monitors everything the entire stock all the way from the physical facility, up to the data you as the customer will get in the application labels, you'll be able work with the application. You'll be able to create the Azrael, but it's the provider that manages everything. Some examples of software as a service are making soft office for the 65 salesforce dot com into it to make financial applications. Adobe Creative Cloud on Gmail Yeah, our normal cloud based email services like cult meal on Gmail. They were actually available before the term cloud was even coined. So we doing normally think of than eyes closed based software as a service. But that's actually exactly what they are. The provider is providing an application to you, the customer on their pervading and managing the entire D ascent or stock. So basically any kind off cloud software that you use is software as a service for the billing with ash, a X, it will typically have a monthly fee per user on vour. Maybe multiple pacing. Tears offered based on the usage is well, let's have a look at an example of some SCS billing structure will have a look at Microsoft Office 56 5 so I'm going to scroll down a little and you can see I'm here on replacing page for office 56 5 For business, the places are per customer per month or per user per month on verse. 30 different tiers year. Different tiers you get different applications available was part of each year starting off well, but on the right office. FFI 65 Business San chose that includes email. It includes Make us off one drive storage escape for business on Make it Soft Teens on the left office. 365 Business. That package includes all of the normal office applications, like Make Us off Word. Excel Access, etcetera Office 36 5 Business and Business of San Chose Candid the opposite of each other. So obviously five business includes all the normal APs, but it doesn't include email. Skype for business or teens, which is on business essentials. Both of them do have the one drive storage Will Really combining the two is what you get with office 365 Business premium that includes all of the different packages, so business essentials is $7 per user per month. Businesses for in dollars, 20 on office 365 Business premium is $17.50 per month. This is pretty kind of standard, praising that you would see with software as a service. Often the provider will give you different tiers onto the higher prices. You're going to be able to do more with my application. Quite often, the least expensive one will actually be a free package as well. On it's usually based on per user on, it's going to be billed per month. 14. 03-06 Cloud XaaS Anything as a Service: in this lesson, I'm going to cover XPS. Anything is a service which is no included in the n i s T s standard definition off cloud computing. Many collective aiders actual offer other eyes of services apart from i e s p e s on S E s . These are sometimes described as X e x for anything as a service. Examples of these include D A s, which is desktop is of surveys D r A s, which is disaster to cover. He is a service B A s backup as a service on storage ISA service. And there's many more as well different clothes Per Vader's will pervade different services . First example that we mentioned there was B s backup eyes e service. This is suitable for small companies because it allows them to back up to an off site location without having a go through the harsh ALOF buying tip hardware. And also transporting the tapes offsite every day with backup is a service. The company has got their own on planet situation, so they've got servers and when all face and boost servers and when office are backed up to storage was a which is at the clothes for Vader Teoh, the customer. The often will not have any actual servers at the Copa Vader. We just make use off the storage here. So what we're doing on an example here, you can see that customer A are backing up their servers to the cloud provider storage. We've also got Company B, and they're doing the same thing. So from the clay providers point of view, there can be multiple customers backing up to the seam started system. So this gives us I were resorts fueling, which makes things more cost efficient. And the provider can pass on voice cost efficiencies to the customer. So this is often attractive solution for small companies. Another example is D. R. E s disaster recovery eyes a service in the example here we've got an active Stein by model to the company. Have got their own on premises solution. They've got their own servers there on the auction would appoint minimal infrastructure The clothes elevator, so similar about previous examples back up is a service the customer is running for servers in their office on they're just using the cloud provider as a disaster. The company solution the doing half servers permanently being used over at peak load. Vader is just for their disaster recovery. Dia is rectal kid, from the company office to the cloud provider. Storage on. If the company has a disaster at their company office, this means that they can quickly feel over to the cloud provider. So this GAN. For small or medium sized companies, this can be a view attractive solution for them because they would love to have a disaster recovery solutions. If they do have a problem, they can get back up and running quickly. But if they were going to do this to the traditional way of building their own disaster the company DIA Center, and putting the infrastructure in there that would obviously be hugely expensive bay using , ah clo provider to give them not capability, they can get it for a much more feasible cost. Next example is De Es Desktop. As a service here, the customer has got thin clients in their location. What I think line is is it's basically are viewing wool powered terminal. It doesn't have enough CPU Ron, except resources to run a standard destined opening system like windows on there. Then the customer. The are using infrastructure at the cloud provider with hyper visors There on the virtual machines are running the normal user dash stop. So this is different than what we would normally do with cloud services with normal quite services. Typically, we're going to be using the cloud environment to host our servers with D. A s were using the cloud environment to host our normal user Dash dolls to the desktop opening system is running as a virtual machine. The cloud provider on we have got a thin client in our office is used to connect to the quote provider on Run the desktop virtual machine From there, the reason that a customer would do this is it's going to save the money on technology. Fresh is because with a normal office seem, we've got 3000 users in there on. We've got all with the PC's, what most PC's get out of debt. So every several years we're gonna have to replace all with voice species, and this can be expensive. If we use a virtual desktop infrastructure like we're doing here, we don't have to the place it was 1000 PC's and our office we just have to build new virtual machines at the club. Per Vader, on it's up to the cloud provider to have the underlying hardware that will support his dad starts Joe again. We are moving the cost from a capital expenditure costs that we have to pee over to a monthly operational cost which could make things more cost efficient. The last example I want to cover is started as a service. Ah, well known example of this is Amazon s free with Amazon s three. You can use Web a storage over r a w asked. You can use it to store things like fails images, etcetera on again. This congee cost effective for you because Amazon have got pretty much unlimited amounts of storage. It seems you having a pervade the storage yourself of examples of storages of service which are not normally cold storage is a service, but they are our dropbox. Make us off one drive on Google drive and you're probably using some of the always already yourself. The reason that these examples are no included in the n i STS standard definition is that actually these could old fall under the other models that were covered in not definition like storages of service. Dropbox makes off one dive Google drive. This could be classed eyes software as a service on the earlier examples I covered, like desktop is of service on disaster. To cover these or service our view, we types off infrastructure as a service. So it's not that there's anything missing in the n i S. T s definition. It does give e complete definition of all the cloud services is just the providers wait to also use X e s because that can help with their marketing. 15. 04-01 Cloud Deployment Models - Public, Private and Hybrid Cloud: in this module I'm going to cover the clothed deployment modeled on die Can fit it all into one rash in the n I s t defined foreclosed deployment models Rights Public cloud Private Cloud Community Cloud on hybrid cloud I will cover each of George different deployment models in Miss slash In starting off with public cloud the way the n i s t defined that the cloud infrastructure is provisioned for open use by the general public it may be owned managed on operated by a business, academic or government organisation or some combination off them. It exists on the premises off declared provider examples of public cloud Our view we all the well known cloud providers that you know about why Amazon Web services maker soft zur a couple of examples there off i e. S on P A s IBM blue mixes a pes example and seals forces an S E s example. So all of those are quite providers which sell their services to the general population on public cloud is by far the most common deployment model. Next up we have private cloud the way that the n i S t defined not the clothes Infrastructure is provisioned for exclusive use by a single organization comprising multiple consumers, for example, business units. It may be owned, managed on operated by the organization, 1/3 party A or some combination of them on it. Me exist on or off premises. Private Cloud works the seam we as public cloud. But services are no pervaded to external public enterprises like they are with public cloud there provided to the organizations or internal business units. So there could be a bit of confusion about how is play that closed different than on premises van. If it's just operated for a single company. Well, the difference is that private, clothed fulfills of the cloud essential characteristics that we spoke about earlier, so those characteristics are undermined. Self surveys rapid us this in broad network access resource pooling on measured service. Private client will fulfill all with those characteristics, whereas an on premises solution, those no most obvious example without is with the traditional on premises model. A business unit orders a new server by raising a ticket with the IT department. The server is van provisioned on, configured by the server network. On started teams are separate manual cool sensors with private Plato. Ever a business unit, orders a new server, typically through a Web portal. The server is van automatically provisioned without requiring any manual intervention. So if you remember, the Western recovered earlier on the mind self service and she would you know about how weak, provisioned a virtual server and twx we went to the Web. Poor old. I configure it all the settings that I wanted for my virtual machine and then in the background automation software deployed everything or unlikely on virtual machine was up and running in 15 minutes. That's different than would happen with the traditional on premises model, where everything would be done manually, and typically it would take a week or more to get the server up and running. So we have Private cloud is different than on premise and that we've got all of those characteristics. So, just like with public Cloud, when the business unit provision a virtual machine, it's all going to be done in the back end automatically. That's going to be using automation software like BMC See A Technologies or Cisco UCS director. It was a lot of other ordination software available and also for companies big enough, they may even end up developing it themselves. Private Cloud is most suitable for large companies, where the long term Otto I and efficiency gains that will get from the solution can outweigh the initial effort and cost to set up the infrastructure on alternated work flows It. Private cloud is an expensive solution because the dance center is dedicated just for that . One customer on all of the infrastructure in there as well, again is just for the one customer, so it's going to be very expensive to get the Deer center set up. I'm sure for everything to be all amid this all needs to be set up ahead of time. Israel's about automation software will need to be deployed, and all overworked floors will need to be written and all of the integration between the front end on the back end components such as storage and networking and the server. We'll all need to be developed and tested a swell so it's expensive and time consuming for private clouds to get up and running. But if the company is big enough, they can make long term cost savings from doing this. There aren't many well known examples off Private Cloud because companies of private client don't usually advertise the fact because it's, well, private. However, ah, well known example is U. S Department of Defense on Private Clothes, which is pervaded by AWS. That's an example of private cloud owned managed aren't operated by 1/3 party rather than the company who's using it themselves. I also want to give you some information on something that's sometimes called Private Cloud but isn't really here as well. Public cloud I es Per Vader's work sometimes market dedicated servers as private cloud because the underlying servers are dedicated for a particular customer. But it's not too private cloud because it's only the servers that are dedicated for the particular customer. The supporting network infrastructure, like the switches, the rowers with fireballs, etcetera is shared. So this is not true. Private cloud for two private cloud. The entire solution is dedicated for the particular customer, not just the servers. The next deployment model is Community clay AIDS and I asked. He defined this. The cloud infrastructure is provisioned for exclusive use by a specific community of consumers from organizations that have shared concerns. For example, mission security requirements. Policy on compliance considerations. It may be owned, managed on operated by one or more of the organizations in the Cuban A. 1/3 party or some combination of them on it. Me exist on our off premises. Community Cloud is similar to a traditional extra net giving example about I worked for an oil and gas company before on we hade private network corn into the with other oil and gas companies such odds, BP and Shell so that we could share information with each other. But community quite is a little bit different than not. It's food share data center services instead of just network connectivity. A Between on premise offices Community Choir is Ville Waste Coleman Deployment model. The final model we have is hybrid quote. This is defined by NASD. Cloud Infrastructure is a composition of two or more distinct cloud infrastructures, private community or public that remain unique entities but are bunk together by standardized or propriety, technology that enables data on application portability. For example, cloud bursting for load balancing between clouds Have a look at this in a little bit more detail. So you saw the term, their cloud bursting what This is about companies with limited private cloud infrastructure . Me cloud burst into public cloud for additional capacity when required. So maybe I've got my own private cloud infrastructure at my company and my dear center, but I've got limited capacity there. I've only got through many servers. They whether I could scale, is by growing my own private cloud. But I would have to pee for the hardware for that, which is to be expensive, a way that I could make it. More cost effective is if I'm running out of capacity and my private cloud, then I can burst. I can expand into e Public Cloud. A company could also have private Clodagh remain, say, and use public cloud for their disaster recovery solution. That's quite common Israel, because building the private cloud infrastructure, which we talked about already is expensive. If we wanted to double that for the disaster recovery, saying is, well, that's going to take the cost even higher. We could bring the cost down by having the main Dia center as private cloud and then use public cloud for our disaster to company site 16. 05-01 Advantages of Cloud Computing and Calculating TCO: this module is going to have a single action, and here I'm going to cover the advantages off cloud computing. Now all the advantages of youi being covered as we've being going through the different lashings. But I want it to summarize them all for you here in one lash in. Also, at the end of this election, I'll give you a quick and basic example of how to do a teach your calculation, for we can see if you're going to move to a cloud solution or not. Okay, so we'll start with the clothes advantages. I'm gonna go through them quite quickly because, like I said before we do, we covered most of these old 80. Starting off with scalability Clothed computing pervades businesses with the ability to regulate the service. In accordance with their current requirements, you can easily skill capacity up and diving eyes needed on you get the appearance of infinite computing capacity on demand. AWS, for example, are not going to run out server space any time soon. On. You also get flexibility through clothed bursting where you can have your own on premises solution or a private clothes solution on you can expand out by bursting into the cloud. We also have business agility actually ability to handle expected or unexpected changes in mood. We can very quickly deploy a new service in the cloud whenever the need arises, so we get reduced trying to deploy an application into production. Next, we have cost efficiency the customer peas just for what they need, resulting in directly proportional costs. The customer avoids provisioning and peeing for the peak as a permanent fixture. For example. Actually, we've got a customer on, on average, be require of Tang servers. But it peaked. Eins Viet acquire 20 servers. Well, if they were deploying and on premises solution, they would have to deploy 20 servers to cope with the peak on those 20 servers would be there all the time. With a cloud situation. We can use automatic scaling so we can have 10 servers normally, but as and when needed weaken skill that up automatically operative it 20. And when we're not redid anymore, we can steal what, Back down to the 10. So we end up on Lee peeing exactly what we need as and when we need it. We move from a large upfront Capex cost to a competitively small monthly or packs cost if we were going to go with a traditional on payment solution. We have to peer for over equipment as an upfront cost. But if we're using cloud services, we're not buying the equipment. Our sales were using the service providers equipment, and we pay for about an ongoing monthly or pecs cost on. This also makes the I C T costume or transponder to the business as well. Continuing with cost efficiency, the customer does not have depreciated ble hardware assets on technology. Re fresh is responsibility off the cloud provider. The provider passes hardware. Meeting its costs on the customer is part of the predictable monthly fee, so we don't have any unexpected costs. We can also gain a competitive advantage from using cloud services as well. Organizations can respond quickly to evolving market trans and focus on growing their core business, reducing capital, expand on infrastructure, releases funds we can invest in innovation or other priority. It is for product to the and this one's a little bit controversial. I T staff can focus more on strategic decisions on developing on improving core applications, Robert and maintaining or troubleshooting in house icy tea. Okay, so something that I sometimes hear is that in the future there is not going to be any I t engineers anymore because everything's going to be going to the cloud. But really, that's not kiss. Remember when we had to look? The I. E s design less in the way that cloud services work is viewing the Seimas. It works from an on premised solution as well. We're still going to need engineers to do the design on. We're still going to need engine years to manage everything up from the operating system level Israel. So we're not going to be replaced Oaks you on leave the cloud providers are also going to need engineers as well. So this is not putting jobs at risk. It's actually making jobs bear, because if I'm working for a company, I don't need to four kids on Monday. In tasks like hardware maintenance, I can focus on mawr productive core tasks that are going to enable new business on a competitive advantage for my company. Next up, we have availability on July ability. All major quote to Vegas facilities are located and hard India centers with redundant power . No single points of failure on on site security. The service will be certified to be relevant industry standards such a Zeisel with 9000 and one, which covers quality A and to seven or one, which covers security. The data center is built by facilities, server networking and started qualified specialists according to best practice. So if you were looking at doing it on payments situation, you're going to maybe be a bit concerned. If you're management, are we actually falling? Best practice? Well, if you're using equal situation the cloud providers of hired specialists in each of the different idiots who are very expedience in building out best in class daycare centers. So you already know that everything is falling. The best practice on you're getting the best possible service there. Check the service available agreement to see what is guaranteed. I'm the compensation if they actually only for the power and the availability off the facility is not man. Next thing costs, and this is where we're going to talk about getting into doing our tissue or calculation as well. So those advantages are all great toe have. But a decision to deploy cloud computing or no usually is gonna come down to the overall long term cost. The total cost of ownership off maintaining and on premises solution should be compared to the tissue of maintaining a cloud equivalent on the advantages and disadvantages of each factored in. When making the final decision. It's knowing either our decision. The majority of companies who used quote services will have a mix of on premise on cloud solutions as well. So the different cost that we need to take into account. If we were going to maintain our own on premise data center, we need to know these so we can compare these costs to the course of a cloud solution. So for an on premise data center solution, we need to factor in the Capex cost of the hardware procurement, the actual equipment we're gonna have in there. We also need to add in the OPEC's cost for taxpayers Paradyne cooling and ongoing management. So this is a picture of exponentially that I made up with its sample Tszyu calculation. Just I really basic calculation to give you the idea of how to do this. Now, don't be using this for actual viewer WorldCom partisans because this is just an example. Spend shit. I made up using made up figures. So looking at the example here, we're looking at the cost of if we were going to use an on premise solution. So the different components that make up the cost of each server for not example it's $65 on. We're going to refresh replace the servers every five years. The cost of running the servers, power cooling, rocks, peace and maintenance will see its $50,000 on we've got 12 servers. The cost of I t support per year for hardware and backups is $50,000. So this is not the actual silent of its being paid to the staff. We figured a off all the money that is going and stuff how much of our is actually used for hardware on Baku's next up. The tape Liberty and Backup software, which is going to be a one off costs at the start of the five years $20,000. So if we ideas together the Capex cost, which is the number of servers times because preserver, pill acidity, Lavery and backup software comes to $92,000 the OPEC's cost, which is the number of servers, tames the cost of running reserve. Er's times vidi fish cycle plus I t support is $430,000. So if we add those both together the tool Compaq's plus OPEC's cost over five years, if we were going to use on on payments tuition would be $522,000. Next, we're going to calculate the cost. If we were going to run, this seems service in E cloud solution. So the way we calculate that is we look at the monthly cost for running it in the cloud. So that is the cost off the servers. Plus, if we've got any optional extras in their wake, if we have to pee any fees for additional software, additional storage back anything, we add that all in. So we've seen that I our monthly fee is $6000. Times are 12. It will give us the yearly fee of $72,000 on. Typically, it'll be new installation fee for going with the cloud solution, So $72,000 per year times that by five gives us $360,000 so the on payment solution was gonna cost $522,000. The cloud solution is $360,000. So we're going to get a cost saving off $162,000 over five years. If we go with the clam situation again, please don't be calling this. This is just an example. Assure you of how you would do the calculation. You need to do this with actual real world figures for your particular scenario. Check the cost of the on premise solution compared with the cloud solution. Factored in boy's advantages that we get with clothes as well, and that's gonna he'll help you to make your decision.