Hello, everybody. And welcome to this lecture and the structure we're gonna be covering the introduction to eight of us computes more specifically E C. Two. For the certified cloud practitioner, there will be some questions related to easy to, uh, easy to basically is a cloud server.
It's a virtual server, ran an AWS cloud,
and it was designed to reduce the time required to build a new server Instance on Prem from weeks into minutes, and it literally is that quick and this course we're actually going to spend a penny see to you get to walk through the process and you'll get to see just how fast and efficient
it is to get exactly what she needs. So you can get your project out the door now. You can quickly scale your resource is capacity up or down
based on the requirements of your computing needs, and I'll get to show you that's too later on throughout this course. So let's jump into the sea to pricing model the freshly model, uh, down for a couple different options you have on demand reserve spot and dedicated. Essentially, this is all that
on demands you're paying a fixed hourly rates. This is the most common one. Um, or you can pay by seconds. And there's no commitments if you just need something up and going real quickly. This is the one that will be selected by default. It's on demands now. If you wanted to reserve capacity,
you're gonna get a pretty steep discounts when you do that.
Basically, these are for those projects that you know you're gonna be using this server all the time. You already made the decision to go with eight of US E c two and you're gonna be locked in for a 1 to 3 year contract. Now, this shouldn't be any surprise if you've been working on it for any period of time.
You probably know that when you get a server you're renting, it's and it's ah, multiyear contract as well.
It costs a lot of money. Ah, and you're stuck with the hardware which may or may not do what you need. So with eight of us, at least you can reserve that capacity
and you're still saving a lot of money opposed to getting a server on premise.
The third pricing model would be spot. This is going to be when you're actually bidding for the press that you want on the servers here that you are looking for the capacity that you're looking for. This is really good for those people who,
you know, they have a product that they need to compute. Maybe they're doing like, some kind of data science or something like that.
And they need some processing power and, you know, maybe they want it at nights. Maybe they want it, Uh, for a very short period of time, like, let's say, Saturday from 4 a.m. To 7 a.m. That kind of thing
spot allows you to do that. And because you're bidding, you can normally get an even better price than you would if you were to reserve
the thesis. To instance, capacity and then the most secure one would be the dedicated. Now,
once again, if you've been working in I t. For any period of time, you've probably heard of dedicated servers. You can rent them online. They usually costs a boatload of money, but you are being given a dedicated hardware that's you know that your information it can be
stored on and it's compliant, it's safe. It's not being shared with other people, and you're not being split within virtual environments.
Ah, that's something that eight of us does offer. It is quite expensive. I do encourage you to take a look at it. But as far as the scope of thes certified cloud practitioner exam, this is just about all that you'll need to know. Let's talk a little bit more about the tears individually, E. C. To on demand is your standard D. C. To pressing option.
It doesn't offer any upfront costs or long term commitments.
It's good for APS with short term, spiky or unpredictable workloads that shouldn't be interrupted. And it's also good for applications that are being built on Amazon for the first time. Maybe you don't know what the use is gonna be for the application, and you just need to get this thing out the door.
This is a good choice to get started with. Then, of course, you can always switch your application over to a different here later on. Easy to reserved as I mentioned before, is foes applications that you already know you're gonna be running for a long period of time it gives you that predictability. Ah, applications needing
some reserves capacity because, you know, you're gonna have to use this for a long time. Like, let's say, an exchange server for your organization. You're trying to move this whole thing's to the cloud and off the
ah on promise environment. Obviously, you know that you're gonna be using your server, you need to emails running. So building that reserve capacity,
uh, using an easy to reserved instances, a great way to do that.
The reserve pricing types kind of go like this. You can have a standard reserve pricing gives you up to 75% off. They do request that you pay more upfronts, but then you get a better discount later on.
Then you have convertible reserves, which is up to 54% off.
This is going to be, um, you know, you're you're giving a little bit upfront, and then you have the ability to change your reserved instance attributes. Maybe you're amounts of processing power. Things like that and those granted changes will result in greater or equal value,
but it is modifiable
based on your needs, and then you can also schedule the reserved instances and, you know, have them launch within a reserve time window. R n a predictable, re occurring fashion. Easy to spot instances are basically the TC two instances that,
um, applications have a flexible start an end date. You know that you're not depending on us solely all the time.
And you only need it for, you know, a couple hours, maybe a couple of days. Whatever. You have something that's coming up, maybe a big, uh, campaign, and you need to have more server capacity to handle the workload for a short period of time. The TC two spots
option is a really good option that you can add to maybe your pre existing easy to
fleet. So that's something to consider applications that are only feasible at a very low compute process. Easy to dedicated hosts are ideal for your more regulated applications things like hippo or stuff like that. They're great for licensing, which does not support multi Tennessee or cloud deployments.
They can be purchased on the man which is hourly, or they can be purchased at a reserve three, which is
which you can get up to 70% off the on demand pricing Here are the different easy to instant types. You have your field programmable. You have high speed storage. You have graphic intensive. You have high disk through puts
lowest cost general purpose. And you have been storage memory optimized
general purpose computer optimized graphics and general purpose GPU and memory optimized. Now, this is a very popular charts that you'll find on Google.
And if you look here, the instance types, initials actually spell out a pneumonic, and this is, ah, fights Dr MC picks. So when you're taking the exam of you may come across a question or two where they ask you, please identify A.
Please identify one of the instant types based on
the pneumonic and your please identify one of the instant types based on the initials, and you're going to be given a couple different answers, and you'll have to identify which one of them is correct. And they will throw in maybe ah z or another letter that isn't in this list.
You'll know that those are not really easy to instant types.
And ah, you will probably want to make sure you memorize this demonic and maybe even draw it out. Dropped for and maybe even draw it out and the beginning of the exam.
That way you can reference this charts and be able to identify the correct instant types and get a couple extra points that way. Well, so what is an easy to conversation without talking about TV S E B s are the actual storage volumes that you attach to the DC to instance. So just like when you're building a computer at home,
you need to have a hard drive you can get on actual hard drive for a solid state drive
E. B s is kind of the same idea. It's something that's Amazon allows you to make some decisions on in order to optimize the performance of your server. Ah, after being attached, you can actually create file systems, run databases, operating systems, whatever and use them similarly to the
hard drive system of the storage system that you have within your computer at home.
EBS volumes are strategically placed and specific availability zones to replicate and protect your data from failure.
So here are the different options that you get for E. T s. You have SST or you can go with magnetic for SST You can go with the general purpose s to see which is GP to this is gonna be balanced on price of performance. Then you also have a more intense more
Ah, more powerful SSC, which is gonna be provisions I offs.
This is high performing SST used for a mission critical low latency and high throughput workloads. Under a magnetic, you're going to have through optimized hard drive, which is an S t one.
This is going to offer a low cost hard drive volume designed for frequent access through put intense workloads and then you'll have cold HDD, which is an S C one. This is the lowest cost hard drive designed for less frequently access workloads, things like file servers and whatnot.
And for those legacy type systems, you can also
choose magnetic, which is the previous generation. Uh, though probably there's not a lot of use cases for that. So
the exam tips for this lecture would be a W S E C. To stands for elastic Compute cloud.
Easy to allows you to spin up. Server's almost instantly, allowing you to save time and money. Make sure you memorize 24 82 Pricing models and also make sure you memorized the easy to instance new Monix. That way you can identify which ones are true instances in which ones are gonna be false.
And then also, you want to make sure that you memorized the different CBS volumes as well.
And the next lecture will be covering how to set up an easy to instance, and you'll get a chance to see the options on how you can add EBS volumes to your A C two instance. But for this lecture, that about summarizes it up, I will see you and the next one.