Understand the Point of the Cloud

0:00[MUSIC PLAYING]

0:06So, you're sitting around the dinner table at Thanksgiving

0:09with your entire family, and they know

0:12that you are the tech person.

0:13You're the person who's interested in tech,

0:15and your uncle looks over you and asks you,

0:19what is the cloud?

0:20What do you answer to them?

0:22What do you say at that point?

0:24This video seeks to answer that question.

0:26What is the point of the cloud?

0:28And how does the cloud do things differently

0:31than we've done in the past?

0:32This sets up even bigger discussions

0:35about things like data centers.

0:36So let's start to lay that foundation.

0:39Let's plant that seed right now and talk

0:41about what is the cloud.

0:42Let's go.

0:43This meme to me never gets old.

0:46When you think about the cloud, one

0:48of the best answers that there are is there is no cloud.

0:51It's just someone else's computer,

0:53and the fact of the matter is that's mostly true.

0:57That's mostly accurate.

0:59Where did the cloud come from?

1:01And what is it really all about?

1:03There's an OK chance that when we think about the cloud

1:06your exposure to the cloud is through cloud storage apps.

1:11You've probably heard of something

1:13like Dropbox, or OneDrive, or Google Drive, or Apple's

1:22iCloud, or they have it called iCloud drive,

1:25I think is what it's called.

1:26All of these things that you have maybe been exposed

1:30to these are storage, right?

1:34The idea here is that I can take my files

1:37like a text file, or an Excel file, or any type of file.

1:43And I could just drop it on OneDrive,

1:46or I could drop it on Dropbox.

1:48And the cool thing about it is I know

1:50that file throughout the rest of history

1:54is probably going to be safe.

1:56Nothing is going to happen to that file.

1:58If I delete my files, or my images,

2:02or whatever off of my hard drive on my computer,

2:05I know that I can go to OneDrive, or Dropbox, or Google

2:09Drive, or whatever, and download them again.

2:12Or even cooler, I could upload something to OneDrive,

2:15drive across town to my dad's house

2:18and download it on his computer over there.

2:21It becomes a way of almost sharing the files

2:24that we want to collaborate on.

2:26And this is storage.

2:28This is one type of cloud resource out of 100

2:32that can be deployed.

2:37And there's a decent chance that you pay something

2:39like a small monthly fee to get an allocation of storage.

2:44I think for a long time, iCloud had something

2:47like for $0.99 a month you could up your storage of cloud

2:52storage like 1 terabyte, and that

2:55was used to back up our iPhones, and get all of our images,

2:58and videos, and pictures of our family,

3:01or vacations, all of that was automatically synchronized

3:05to the iCloud storage that way I could

3:08access those files, or those images,

3:10or those videos any time I want.

3:12And with all of this extra storage,

3:14I felt very comfortable knowing that I could take basically

3:18an unlimited amount of pictures or videos of my vacations,

3:21or my kids, or whatever, and it would get automatically stored

3:25in the cloud.

3:25And if I got a new phone, it would just

3:27download from the cloud.

3:29And that was a really cool thing.

3:30But what's really going on there?

3:33What is this storage when it uploads all of these images?

3:38Here is my phone right here.

3:41I'll put the little circle here, because there

3:43used to be a button on iPhones right there in iPads.

3:45So when I take a picture snap, there's my picture like that.

3:49And it's got, of my face on a vacation

3:52or something at a beach.

3:55What happens is whenever my phone typically

3:58gets connected to Wi-Fi again, it immediately

4:01transmits that image to a secure facility where they just have

4:08gigantic pools of hard drives.

4:11Just an absolutely insane amount of storage

4:15that they have running in Apple's own hardware.

4:18Apple has their own facilities with just extraordinary amounts

4:23of storage and hard drives upwards of the eta bytes.

4:28That's right, they're even bigger

4:30than petabytes or terabytes.

4:32There are eta bytes.

4:32It's just massive amounts of storage.

4:35And all around the world when people

4:37are snapping these images, these images

4:40are just getting stored securely on hard drives.

4:45And of course, they're associated with your account,

4:47and you've paid out of all of these massive blocks

4:50of storage.

4:51You've paid to carve out 1 terabyte a month

4:55worth of storage for your images to land in.

4:58And whenever you need to recall one of those images,

5:02because maybe you deleted it off your phone

5:05or your iPhone is such that it automatically

5:07pushes old images to the cloud, it just

5:10automatically seeks out that exact spot in that image

5:13on those hard drives and just redownload it back

5:15to your computer.

5:16Put this another way, the cloud, in this case,

5:19the iCloud storage is just somebody else's hard drives.

5:24In that, that's really the idea that

5:27comes to mind when we start to think about the cloud.

5:31It really is just somebody else's computer.

5:34I want you to think again for a moment, the cloud it really

5:38found its feet with Amazon.

5:41And you think about amazon.com and how big amazon.com is now.

5:46At least in the United States, but worldwide,

5:48amazon.com now has its presence.

5:51Wherever you are, you can get very quickly

5:54redirected to an amazon.com website

5:57where you can shop and buy things

6:00and with prime delivery have it dropped off

6:02in front of your house within two days in most

6:05locations, one day in some major Metropolitan locations.

6:09Think about what it took to get amazon.com spread out all

6:13around the world, right?

6:15All around the world.

6:16Wherever you go, amazon.com has web front

6:19ends that you can quickly access,

6:21because you just connect to their website.

6:24This took a tremendous amount of computers and servers spread

6:30out all around the world.

6:32So, that way, whenever you want to connect to their website

6:35and buy things, these servers are serving the content to you.

6:41They're serving the ability to shop and buy things and charge

6:46your credit card and send of to logistics

6:48to have the product shipped to you.

6:50To make all this happen, Amazon spun up

6:53really humongous facilities all around the world

6:57called data centers.

6:58We're going to talk about data centers

7:00more in an upcoming video.

7:02And what they found out pretty quickly

7:04is they invested so much in this infrastructure

7:09right here that they actually had

7:11excess servers sitting around.

7:14They had excess storage, excess compute, excess network,

7:19and excess RAM.

7:20But they had spent a humongous amount of money

7:23on building all this stuff out.

7:25So how can they get some of this money back?

7:29Effectively, they rent it.

7:32That's what the cloud is.

7:34Now, if you are a business, and you want your web front

7:40end or your applications to run with the same availability,

7:45with the same reliability of amazon.com you can simply

7:51rent these servers from amazon.com

7:55and run your infrastructure, run your applications

7:59on the exact same hardware that Amazon is running Amazon on.

8:05That's the idea with the cloud.

8:07When we say it's just somebody else's computer,

8:09if you're going to, AWS, which is Amazon Web Services.

8:13It's just Amazon's computers.

8:16But we're here to talk about Microsoft again, right?

8:19Take the exact same idea, except in yellow

8:24where I wrote Amazon things, we're going to do blue,

8:27and it's Microsoft things.

8:28We've got Microsoft websites, we've got Bing.

8:31We've got all of this infrastructure that

8:33makes Microsoft stuff work, and guess what?

8:36They did the exact same thing, spinning up servers

8:39all around the world, all around the world.

8:41It's just servers, servers, servers, servers, and storage,

8:44and hard drives, and massive networks,

8:46that do things like 300 terabits a second of data transfers.

8:52That's literally what's happened here

8:54is they have spun up these gigantic facilities again,

8:58called data centers.

8:59And they had so much success compute and storage,

9:02they started renting it out.

9:05That is Azure.

9:07Azure is where you can go to say,

9:09hey, I want to borrow some of your servers,

9:12and some of your storage, and some of your networking

9:15to run my application, and my website in.

9:19And this is what businesses do.

9:21Now, of course, there's a personal side of things.

9:24You experience this with something

9:26like OneDrive, where they're actually leasing

9:29some storage out to you.

9:31But when it comes to Azure, Azure

9:33is really for businesses to deploy business scale

9:38applications.

9:39To deploy their websites.

9:40And to deploy all of the pieces that make tracking and shipping

9:45work.

9:47If somebody places an order on our website

9:50that we were running in Azure, a big piece of that

9:53is that order needs to make its way to a warehouse

9:57to box up a package and ship it to via UPS. So

10:01that's a whole different component of the application

10:04that has to also run in the cloud.

10:07And the end result is when you use something

10:10like Azure or any other public cloud,

10:13the idea is that you are getting this reliability.

10:17And the cool thing about renting it

10:19is, instead of ping this humongous fee

10:22to deploy all of these servers, and the electricity,

10:26and networking, instead we just get paid an hourly rate.

10:32The ultimate result of this is Azure

10:35and us are now the most profitable portions

10:40of those businesses.

10:42Amazon, Microsoft, Google, their public cloud operations

10:46are now where they generate their revenue.

10:49It's not online, sales.

10:50It's not selling video games, or operating systems.

10:54It's renting out their compute.

10:56It's renting out their hardware.

10:58It's renting out their infrastructure.

11:00So, as a result of them going, oh,

11:02this was a really good idea.

11:04They threw a tremendous amount of effort

11:07into improving their public cloud offerings

11:11over basically the last decade.

11:13And now we have incredible offerings.

11:15Azure personally is the one that I choose to work with.

11:19So, now we understand the joke when you say there is no cloud,

11:23it's just somebody else's computer.

11:25You know what?

11:25That might be true.

11:27We're really just carving out a piece of Microsoft's computers

11:31and their gigantic data centers about

11:34to actually deploy our code into it.

11:36To deploy our applications, or our systems into it,

11:40and pay them a monthly bill that is charged per hour.

11:46And from there, we actually get the benefits

11:49of running our code, our infrastructure, our systems

11:53on Microsoft's hardware.

11:56So this is understanding the point of the cloud.

11:58The cloud really is somebody else's computer.

12:01In this case, for the rest of this set of videos when we're

12:04talking about the AZ 900, we're running

12:06this on Microsoft's own hardware,

12:09in their own data centers, on their own servers.

12:13Now, what are data centers?

12:15Why do we care about these things?

12:17Why is that such a benefit?

12:18That's what we're going to talk about in the next video,

12:21I hope this has been informative for you,

12:22and I'd like to thank you for viewing.

0:00[AUDIO LOGO]

0:06So why do we care about data centers?

0:08And why is it such a big win for us

0:11to use somebody else's data center?

0:13In this video, we'll talk about what data centers do,

0:17how they're structured, and why we typically

0:20want to put our critical workloads within a data center.

0:23Let's go.

0:25So you've probably seen pictures of data centers

0:27before but never actually realized

0:29that you were looking at a data center.

0:31Look at this.

0:32Google "data center" real quick and just

0:35take a look at some of the Google images that come up.

0:37In fact, right here on the right-hand side,

0:40what you're looking at here is, in fact, a data center.

0:44You see it?

0:44I mean, it's not obvious at first

0:47why this is such a big deal, but it's actually a huge deal.

0:51It tells you right there on the screen,

0:53a data center is a dedicated space

0:56within a building or a group of building

0:59used to house computer systems, blah, blah, blah, blah, blah.

1:01And I'm not going to read this to you because that

1:04would defeat my job.

1:05I'm going to tell you what data centers are all about.

1:08Data centers, at the end of the day,

1:11are really, really important pieces

1:14to keeping our applications online.

1:16Think about it again.

1:17When was the last time amazon.com went down?

1:21When was the last time google.com went down?

1:24When was the last time microsoft.com went down?

1:29They haven't.

1:31They don't go down.

1:32They stay up all the time.

1:34There's never some form of an outage or server or system

1:39failure that wipes out these websites.

1:42Sometimes the internet itself has problems like we

1:47see something like DNS servers get attacked,

1:51and that slows down the ability for us

1:54to find these web servers and actually access the websites

1:57themselves.

1:58But they don't go down.

1:59They don't go down due to outages.

2:02And also, importantly, something we're

2:04going to talk about in the next video,

2:06they don't go down because new features or updates are getting

2:09deployed to the applications.

2:11They stay online.

2:13And there's a combination of things

2:15that make all this happen.

2:17But really important and really at the center of it

2:20are our DCs, or Data Centers.

2:24So why are data centers such a big deal?

2:27Well, at the end of the day, the key thing

2:30that makes data centers so cool is redundancy.

2:35We want two of everything everywhere.

2:39And that does mean it costs twice as much.

2:43Sometimes it means it actually costs more than twice as much

2:46to get two of everything everywhere because if something

2:50were to die or fail on us, we want

2:54to be able to just roll over to the backup piece

2:58and gives us time to fix the initial problem that comes in.

3:02So when we look at things--

3:04you saw those rows and rows and rows of racks.

3:08Those racks are what we call them.

3:10And within these racks, we mount our servers.

3:15I'm going to put it SRV right here,

3:17and I'm going to put SRV right here,

3:19and I'm going to put SRV right here.

3:22Here's our servers.

3:23And then right above that, we'll put in some networking

3:28components right here, some network devices like this.

3:31And all of these servers are going

3:34to have redundant connections to these network

3:38devices like this.

3:39It's going to look like a lot of spaghetti all over the place,

3:42but it's going to make sense to the network engineers who

3:44are installing it.

3:45So if one of these servers goes down right here,

3:49I've got two other servers waiting in line

3:52to pick them up.

3:53If one of these cables fails right here,

3:55I've got three other cables plugged

3:57into all of these network devices going on right here.

4:01If this network device goes down,

4:03I've got this network device standing in line

4:06to pick up the traffic and redirect it everywhere else.

4:09And if this entire rack goes down,

4:12good thing I've got an entire facility of racks

4:15absolutely everywhere to pick up all

4:18of these services that are running

4:21on these servers or these network devices.

4:23But wait, there's more.

4:24These servers and these network devices need electricity.

4:28So I'll run electricity to the rack like this.

4:32And this will come from my power company

4:37A. Is that redundant though?

4:39What if power company A goes down?

4:42Well, guess what?

4:43That's why I'm going to have power company B

4:47step in and deliver electricity to my facility

4:51and run that to the rack too.

4:53So all of these servers and these network devices

4:55will have two power plugs.

4:57But even further than that, even more

5:00than that is, at the bottom of these racks,

5:02I'll put in two batteries right here, two battery backups.

5:07These are called Uninterruptible Power Supplies, UPSs.

5:12So on the bottom of these racks, there will be batteries.

5:15So power company B will feed with this battery,

5:18and power company A will feed these batteries.

5:21And that will distribute the electricity securely

5:24and in a proper load-balanced way to these servers

5:27and these network devices.

5:28And if power company A goes down,

5:31it'll flip over to power company B.

5:32And if that company goes down, then it'll

5:34flip on to the batteries until we have enough time

5:37to restore power or migrate our services to other racks.

5:42And the crazy part about all of that is this

5:46is all done automatically.

5:49Oh, but wait, we're still not done.

5:51We still have the internet to deal with.

5:53We've got these network devices.

5:55So guess what?

5:56I will have at least two service providers

5:59and two interconnections, at least two,

6:02coming in to each one of my data centers

6:05to provide redundant internet.

6:07So if internet access goes down because ISP A is under attack

6:11or something or somebody, the new guy,

6:14cut the fiber or something, we just roll over

6:16to ISP B. Again, all of that is handled automatically.

6:20Oh, but wait, there's still more on top of that.

6:24These things that we have throughout this entire data

6:27center, they are hot.

6:30This is generating all sorts of heat

6:33and blowing heat all over the place.

6:35So we'll actually design the data centers to have a cool air

6:40lane where cool air gets sucked in by the servers

6:43and it spits hot air out the back.

6:46But we'll install fans such that it actually sucks the hot air

6:50out, returns it to the HVAC units,

6:53and then blows the cold air back in.

6:55And you guess what?

6:56Those HVAC units, they'll be redundant too.

7:00We'll have multiple HVAC units up and running

7:03to keep the cold air blasting in all the time

7:08and keep the air circulating in such a way

7:10to keep this hardware up and running.

7:12In each one of these servers, we're

7:14going to have multiple CPUs.

7:16We're going to have multiple hard drives.

7:19In fact, we're going to have storage racks of just nothing

7:22but hard drives using software-defined storage

7:27or SANs.

7:28This is Storage Area Networks.

7:30We'll talk about how storage works when the time comes.

7:33But for now, just know we're going

7:35to have dedicated storage appliances

7:37for humongous amounts of redundancy with inner storage.

7:40And we're going to talk-- again, I can't stress this enough.

7:43There is a very sophisticated way

7:46that Azure handles all of this redundancy.

7:49At the end of the day, deploying a data center-- as hopefully,

7:54this is starting to register in your mind--

7:56deploying a single data center is an expensive feat

8:00because we have to have very good servers.

8:04And we have to have multiple of them in case one of them

8:07goes down or has a component failure.

8:09We have to have very good, very fast networking.

8:13We have to have very good electricity.

8:15And we have to have very good internet connections.

8:18We have to have very good HVAC flows.

8:21Managing power levels and temperatures and bandwidth

8:27and actual hardware component failures

8:30is an extraordinary undertaking to managing data centers.

8:34And there are many, many, many, many, many systems engineers

8:39and network engineers whose sole job

8:42is to show up to these data centers

8:44and monitor every single aspect of this

8:47to keep these things online.

8:50So you as a customer or as a systems administrator

8:55or engineer or developer who's thinking,

8:58I need to make sure that when I deploy my application or when I

9:04deploy my infrastructure like my file share or my Active

9:09Directory domain controller, when I deploy these things,

9:12I need to know that they simply will not go down

9:17because they're deployed in state-of-the-art data centers

9:22with redundancy absolutely everywhere.

9:26So when you go to Microsoft Azure and you're like,

9:30I'm going to deploy a virtual machine or a web

9:33hosting application, you are deploying it

9:36somewhere on one of these servers that

9:40has multiple hard drives, multiple CPUs, multiple RAM

9:44chips, multiple network interface

9:46cards, multiple uplinks into the network,

9:49multiple power connections into batteries,

9:52and to multiple power companies.

9:54Multiple ISPs give you a exit point out of the network.

9:59Multiple HVAC units keep this thing running cold.

10:04Do you see what I mean, how data centers

10:06are an absolutely critical part to deploying applications

10:12to keep them online?

10:13Ultimately, your application or your infrastructure

10:17is your money.

10:20It is your income stream.

10:23A lot of applications sell widgets online,

10:26but a lot of applications are also

10:28something like in-house built to help

10:30manage our processes online.

10:32Think about how the government works.

10:33The government's not really out there selling products,

10:36are they?

10:37No, but they have applications that they use.

10:40And they need those applications to be online

10:44or else they stop working.

10:46And when they stop working, they stop generating productivity.

10:51They stop efficiently using their dollars.

10:54It costs them money to be down.

10:56So this is why we use data centers.

11:00And this is why the cloud is really renting out

11:03already-existing data centers by the largest technical providers

11:08that can possibly build data centers.

11:10So now that we understand data centers

11:12and we understand how they keep applications

11:15online via availability and redundancy, now

11:22we'll talk about the other piece of how

11:25these applications stay online when we roll out new features.

11:30That's the art of DevOps, and that

11:32is a really powerful thing.

11:34I hope this has been informative for you,

11:36and I'd like to thank you for viewing.

0:00[AUDIO LOGO]

0:06Probably the biggest thing to come out of the cloud,

0:09aside from organizing the way that we bill and pay

0:12for our infrastructure services, is DevOps.

0:16DevOps is the advent or the theory

0:19or the methodology of changing an application without really

0:25breaking anything.

0:26Think about it for a second.

0:28When was the last time google.com went down?

0:31When was the last time amazon.com went down?

0:34These websites change all the time.

0:37They get new features added to them behind the scenes,

0:40or they roll out new products.

0:41How does that really work, though?

0:43How do we change this application

0:45without actually taking it down?

0:47This is the magic of DevOps.

0:48And in this video, we're going to talk about,

0:51at a very high level, how DevOps works

0:54and how it was enabled by the cloud.

0:57Let's go.

0:58So maybe you've heard about this thing, maybe not.

1:01It's a big deal that's really found its prominence, it's

1:05really found its feet in popularity within, I'd say,

1:09the last six years, although it's been

1:11a thing for over a decade now.

1:13It's called DevOps.

1:16And the idea here is-- well, let's actually

1:19start with the scenario.

1:20Think about it again.

1:21We actually mentioned this in the previous video.

1:24We had amazon.com.

1:25We had google.com.

1:27We had microsoft.com.

1:30I'm just going to put MSFT right here to shorten it up.

1:33These sites have never gone down,

1:35or they haven't gone down for a really, really long time.

1:39But the crazy thing about these sites is that they change.

1:43In fact, let's clear the screen and take a look

1:45at some of these sites.

1:47Well, I'm on google.com right now.

1:48You saw that.

1:49If I go to amazon.com right now, this site

1:52is changing all the time, not just the stuff

1:55that it's advertising in front of you,

1:58but the actual blocks of what they're putting here

2:00themselves.

2:01And how did it know to populate this data right here?

2:06So if I'm drawing right here and I say like, explore more

2:08from across this store, or if I scroll up,

2:11how did it know that I like Mountain Dew so much?

2:15Or how does it know that I'm big into tech?

2:18Or how does it know that these are

2:19the books that I'm interested in and recommended for me?

2:22I'm looking at graphics cards and tablets and Raspberry Pis.

2:25How did it know that these are all of the things

2:29that I'm particularly interested in?

2:31Look at these glasses.

2:32I remember buying those glasses.

2:34How did it know that?

2:35The crazy thing about all of this is that behind the scenes,

2:40there are tons of miniature applications

2:44that work together to make all of this work.

2:49The idea here is called microservices.

2:54And it's a big part of a bigger picture topic that is DevOps.

3:01What we're really talking about here

3:03when we talk about DevOps is we're talking about,

3:07how can we change code?

3:11How can we troubleshoot issues, fix bugs, as well as roll out

3:18and introduce new features into a production environment

3:23without taking the production environment offline?

3:27For the longest time, when we created applications,

3:31if I were to create an amazon.com in the most

3:34traditional possible sense, we would

3:37have a web server or a web what we call the front end.

3:42This is what end users, people out there in the world

3:46interact with.

3:47So on this front end, we'll allow people to shop.

3:51They add things into their cart, and then they

3:56go to purchase the actual item itself.

3:59There's a bunch of things that have to take place

4:01if you really think about this.

4:03It has to charge a credit card.

4:05So this needs to go to a credit card processor.

4:09And the credit card processor has to tell the application,

4:13was it good?

4:14Or was it bad?

4:15Did it actually successfully charge the card?

4:18Then once it's gotten a successful charge,

4:21we then need to send this out into logistics

4:25for shipping and handling.

4:27We need someone in the warehouse to package this up and ship it

4:31out.

4:31And underneath the hood, there is typically,

4:35in a traditional sense, a back-end application

4:40that is up and running.

4:41The front end communicates to the back end.

4:44When somebody clicks the Buy button, that

4:46tells the back-end server, OK, you

4:48need to charge this credit card.

4:49And if it succeeds, process the shipping and handling

4:53and let me, the front end, know when you're done.

4:56That way, I can tell the front-end user,

4:59hey, we were able to successfully charge

5:01this credit card.

5:02Then, most importantly, we need to store

5:05a record of all of this transaction stuff

5:08in a database.

5:10That way, if our front-end user ever

5:12wants to look at their order history,

5:14we can pull it out of the database

5:16and show it back to them on their screen.

5:19And they can look at the status of their order

5:21or what they ordered or all sorts of details like that.

5:25This is a traditional three-tier app, a three-tier application.

5:32There's a front end.

5:33There's a back-end app, sometimes known as middleware.

5:37And then there is a database that actually stores

5:41all of the historical data.

5:42This three-tier app, we used to actually bundle this all up

5:47into one gigantic application.

5:49And we called it a monolithic application.

5:54There are still monolithic applications

5:57absolutely deployed today.

5:59Typically, they're very small applications.

6:02They can run on top of one server

6:05and have all of this running on the exact same server

6:08with no real issues about it at all.

6:12This would be great for an intranet website

6:16that we would use inside of our company.

6:18And only a handful of users are using it at any given time.

6:23This is the ideal with that.

6:25But with amazon.com or microsoft.com or google.com,

6:29where we are literally processing

6:32hundreds, if not thousands, if not millions of requests

6:37per second, well, we have to get a little more

6:40flexible with these types of things, don't we?

6:43So what they did is, first of all, they do have front end.

6:48Of course, that's how we access amazon.com.

6:50I like to pick on Amazon a lot because it's

6:53the app that we know.

6:55But where does Amazon get its data?

6:58Where does Amazon get the information

7:01to fill in all of these boxes recommended for you

7:05and buy again and what's popular in your area?

7:09You should think of each one of these things

7:13as its own separate application right here.

7:17So I'm going to draw this right here.

7:20Instead of one back-end application,

7:23you should think of these as a cluster of applications

7:27that all serve one important purpose.

7:30And that's to get recent orders.

7:34That's to get what is currently in my shopping cart.

7:37That is to get what is popular in my area.

7:41That's to get my suggestions based on the categories

7:45of things that I shop for.

7:47You should also consider the everything

7:49that you do on this website is being

7:52recorded in an application.

7:55What did you click on?

7:56How did you navigate from one thing to the next thing?

8:00And if you clicked on this thing that was suggested for you,

8:03that might mean you're more interested in things like that.

8:07So maybe we have a machine learning algorithm

8:10that is running, that is learning,

8:13what are the types of things that you shop for?

8:16And that tells the suggestion application to suggest

8:20different kinds of products.

8:23This is the idea of microservices.

8:27We can take a huge application and break it down

8:31into smaller pieces.

8:33And, yeah, you should absolutely think that each one of these

8:37is effectively running on its own server with its own network

8:43components, with its own storage and everything like that.

8:47Why is this such a big deal?

8:49Because now instead of deploying all

8:53of this on one gigantic server, then the entire server

8:57comes under load, and we have to expand

9:00that entire server, which causes us to take that server down.

9:04Now, if the suggestion app, which

9:08was located right here-- if the suggestion app becomes really,

9:12really, really popular, we can now

9:14just take one tiny little component offline

9:17and expand it to be even bigger.

9:20But even better than that, with the advent of DevOps,

9:26we can now add redundancy in place across all of my apps.

9:33And, very interestingly, I can introduce new features

9:37to one of these apps and see how it works.

9:41If it's working well, I'll roll that feature

9:44into its little buddy application right here.

9:47And then I'll roll that feature into this little buddy

9:50application right here.

9:52So because of DevOps, I now have the ability

9:56to introduce new features into a small subset

10:01of my underlying application and then introduce that

10:06into the broader subset of the application

10:11until it's in full-blown production.

10:14And I can perform maintenance fixes

10:16on each one of these microservices

10:19without ever taking down the main production

10:22application itself.

10:23What DevOps is, to define DevOps,

10:27it has given developers a way to automatically--

10:33this is important-- automatically introduce

10:38and test new code into an application

10:43without taking the application down because now we

10:49have the ability to introduce our code into a developer

10:53environment, where only they can see it.

10:56And then they can run all sorts of automated or manual tests

11:01against it.

11:01And then they can move that into an actual test environment,

11:06where we allow specific end users to test out new features

11:10or fixes.

11:11Then we can roll that into a staging environment, where

11:16we might allow some real-world production traffic to access

11:22that staging environment just to see

11:25how it works when we actually put a real production

11:28load on it.

11:29And then when we're finally ready

11:31and we know that it's going to work,

11:32we can roll it into the production environment

11:35again because we're really just impacting

11:38a microservice and not the entire application itself.

11:45So this is now giving developers a operational standard

11:50and a management standard for deploying code

11:54into production environments in such a way

11:57that it doesn't tear everything down.

11:59So why did the cloud enable this?

12:03Why did the cloud make all this possible?

12:06Because at the end of the day, the big thing about the cloud

12:10is called scale.

12:13We can, with the click of a button or even automatically,

12:19deploy new servers automatically.

12:24We can also, automatically or with a click of a button,

12:27reduce the number of servers that we have.

12:30Beyond that, with a click of a button or automatically,

12:34I could just as easily take my servers

12:37and make them even bigger servers.

12:40We can give them more resources because the cloud does all

12:45of these operations virtually.

12:49So what I'm saying here is when we talk about scale,

12:53if I have, say, a web server and I'm serving web content right

12:59here with these two servers.

13:00And all of a sudden, I'm getting a huge influx of traffic

13:04because the stuff that I'm selling on my website

13:07becomes really popular, I can scale out

13:12or scale horizontally.

13:14And what that means is with a few clicks of a button

13:16or automatically--

13:18we're going to talk about this a lot more, automatically--

13:21I can simply allow the cloud to add more servers

13:26with the exact same code, just literally copy and paste

13:29the servers themselves and load balance

13:32the traffic across all of these servers

13:35automatically or manually if I want that type of control.

13:40And the exact same thing goes.

13:42When the traffic dies off, I can just scale them back in.

13:46And remember that whole billing thing.

13:48I only get billed for what I use.

13:50That's exactly right.

13:51I would only get billed whenever I've got an influx of traffic,

13:56and I scale out.

13:57During this period, I would get billed more.

13:59When the traffic fades off and I tear down these servers,

14:03I pay less.

14:06So I literally only get billed for what I use.

14:08But then if I just decide, you know what?

14:11The real solution is I just need bigger servers.

14:14With a click of a button, I can scale them up.

14:17That's right.

14:18Instead of scaling horizontally, we can scale vertically.

14:22And when we say scale a server up,

14:24that means take it from two CPUs to four CPUs or 8 gigs of RAM

14:32to 32 gigs of RAM.

14:35Again, this stuff is as easy as clicking a couple of buttons

14:39or letting the code handle it itself.

14:42It can automatically react to current events

14:45and current workloads and scale my servers

14:48on an as-needed basis.

14:50Yes, we'll talk about how Azure specifically does

14:53all of these things coming up.

14:55But the big thing I wanted you to take away right

14:57now is understanding where DevOps came into play.

15:00So the ability to dynamically adjust

15:04the number of servers I have at any given moment, hopefully,

15:07when I start talking about these clusters of servers

15:11that I may deploy as microservices,

15:14hopefully, it's starting to make more sense how the cloud made

15:18this a thing because now when I'm

15:21talking about my suggestions app right

15:24here, if my suggestions app becomes more and more and more

15:27popular, I can just add more of them

15:30and load balance across this portion right here.

15:34And this still minimizes my cost because I didn't

15:37have to scale everything else.

15:40I only had to scale the piece that really matters.

15:43And even cooler is as I'm starting

15:45to ship new code into it, I can ship my new code

15:49into just this instance of the suggestion app right here,

15:52test it out, and then roll it into more of these right here.

15:59The actual process right here--

16:01this is just for your trivia.

16:03If you ever find yourself in this world,

16:05this is just for your trivia.

16:07The actual process of shipping code

16:10into one of these tiny little environments

16:12before you roll it out into the rest, this

16:14is actually known as a canary deployment,

16:18just so you know, just for your own good measure.

16:20That way, you can actually tell someone,

16:22I know what a canary deployment is.

16:24It's when I have basically four of the exact same instance

16:28running, and I ship new code into just one of them

16:31just to see how it does.

16:33And if anything goes wrong, I can just

16:35roll the code back or take it out of the load balancer.

16:38That's the idea here.

16:39So because the cloud enables dynamic scaling, up, down,

16:46in, or out of each of our servers,

16:48I can break my application into a bunch of tiny little servers

16:53and scale each of those instances as needed

16:56or ship new features into one of those little subsets

17:01and see how it does and then slowly

17:03roll the code into the actual production feature set itself.

17:07So DevOps is the actual practice of leveraging

17:13the elasticity of the cloud, the flexibility

17:17and scalability of the cloud to meet our demand.

17:21We leverage that feature set to break our application

17:26into smaller pieces and then introduce new code to it safely

17:31and securely in phases.

17:34And because of this, we don't have

17:36to take down the entire application every time

17:39we make a change.

17:41This has been understanding DevOps.

17:43And this has been understanding how the cloud brought us

17:47to this point, to this entire new style of managing

17:50and running and deploying applications.

17:52I hope this has been informative for you.

17:53And I'd like to thank you for viewing.

0:06So now that you have an Azure subscription, what can you do?

0:10I'd encourage you right now to just start clicking around, see

0:14what we can do to explore the Azure resources,

0:18how is it laid out, can we change

0:20any of the settings or the way it looks and feels

0:23from the Azure portal?

0:25In this video, we're going to take a quick look

0:27and encourage you to start your exploration.

0:30Let's go.

0:31So you've landed here in Azure, portal.azure.com,

0:35you can get signed in right now.

0:37I want you to explore some of the things that are in here.

0:40I want you to actually see if you

0:42can adjust this to be the way that you want it to.

0:45First of all, notice there's a hamburger menu over here

0:48in the top left corner.

0:49If you give it a click, it'll show

0:51you some quick links to other popular resources

0:55that they find out.

0:56What you're actually seeing right here

0:58is this piece will run color, right, let's do this.

1:01This piece right here, this is all kind of pinned,

1:05you're always going to have create a resource, home,

1:08dashboard and a list of every service,

1:11Azure offers right here.

1:13You're definitely going to want to click on all services just

1:17to explore what you can do to get started.

1:19There's all resources right here.

1:21So you may be thinking, what's the difference between services

1:24and resources.

1:25Well, services are the offers, resources

1:30are things you have already deployed.

1:34So that's a big thing to understand right now.

1:36This is looking at what can I shop for versus what do I have.

1:40Then the stuff that's under that this is your favorites list.

1:44This is stuff that by default comes pinned for you,

1:48and these are really popular resources that

1:51are used in Azure nearly every single day by most

1:56Azure developers and a lot of systems administrators too,

1:59you have the ability to change these favorites.

2:03When you start to explore things that you

2:05want to actually favor it.

2:07How do you adjust that list?

2:08Look, if I just hover over one of these things

2:11like virtual machine scale sets right here,

2:13do you see, right there, it's kind of tricky to see

2:16but right, oh, there was a star in the top right corner, here,

2:20let's try and get it, right there where my mouse is,

2:22you see this star, that's how you add it to your favorites.

2:25If you want to remove something from your favorites,

2:27pick on something like SQL databases right there

2:29and see how the stars already clicked,

2:31you can unfavoured it like that by just clicking the button.

2:33Bring it out over here, and now the SQL database is gone.

2:36I'm going to put it back because I like those SQL

2:38databases quite a bit.

2:39Now, there's also the actual view or template or theme

2:42itself, how can you change the appearance or what

2:46is actually on your dashboard whenever you come to Azure?

2:49Click on the little gear, right here,

2:51kind of in the top center over these menus, click on the gear,

2:55and then you see appearance in startup views.

2:57If you give that a click, look at this,

2:59now, we have dark mode.

3:02Thank you.

3:03Let's click on the theme of dark, and apply add, my eyes,

3:07they're so grateful to have dark mode in Azure like this.

3:11So with this set, now on dark mode,

3:13you can click the X right over here

3:16to close what we like to call the blades.

3:19So I'll close this right here, and we'll

3:21take a look at a little bit more of these blade things

3:23that we're talking about.

3:25Right here, I see virtual machines

3:27but if you don't see virtual machines,

3:28you can click on the hamburger menu,

3:30and I can pretty much promise you,

3:32you'll see virtual machines right here

3:34on the left hand side.

3:35Click on Virtual machines, and then this

3:37takes you into the landing zone of where

3:40you would view and manage all of your virtual machines.

3:43Now, I'm not going to deploy one right now.

3:45But I do want you to see and understand a big navigation

3:49point of Azure.

3:50When I click on Create, I'll choose Azure Virtual Machine

3:54by default, and what it does is it brings up the first blade.

3:59Azure, as you start to navigate through many different menus

4:03in Azure like this, what it'll do

4:05is it'll bring up a new menu like so.

4:09And then you'll navigate through that menu,

4:11and then it'll bring up another new menu like so.

4:14So you'll start here with menu one,

4:17and then you'll find yourself in menu two,

4:20and then you'll find yourself in menu three, each of these

4:24are known as blades.

4:26And within each of these blades, you'll

4:29see the X to closed blade that you're currently working on.

4:33This also will be seen as a navigation path right here.

4:37So as you open up more blades, you'll

4:40see these as navigation paths right here at the top.

4:43And you'll be able to quickly jump to any of them

4:46as you go along.

4:48So because we haven't talked about things like resource

4:51groups or anything like that, I'm

4:53not going to show you how to deploy a virtual machine,

4:56because that's jumping way ahead.

4:59But I do want you to see an instance of blades

5:02so give me one second while I pull up

5:05a diploid virtual machine, and you

5:07can see how the navigation path really looks.

5:10So here, I've just quickly deployed

5:12a virtual machine running in the Central US right now,

5:16look at this, notice when I start

5:18clicking on some of the virtual machine templates here,

5:21right here, I'm in home and then I'm in the Overview page

5:24this is one really long string to say I'm in the Overview

5:27page of the virtual machine.

5:29But if I start clicking into things like networking,

5:32this will bring up the networking blade.

5:35Then if I want to look at the exact interface,

5:38the actual resource that was created

5:40on this virtual machine, I can click on that,

5:43and this takes me into yet another blade one more time.

5:47Notice, if I click X right here, see,

5:49I'm in the network interface, if I click the X right here,

5:52it jumps me back to the previous screen where

5:55I was in networking, and if I click X right here,

5:58now, I'm all the way back to this deployment resource

6:01that I was originally in.

6:03So this is just to understand how the blades actually work,

6:08and how there is kind of a natural progression of clicking

6:11around in Azure when you've already

6:14deployed some resources.

6:15So again, I really, just to kind of start

6:18to wrap your head around what is in Azure because we're

6:22going to talk about a lot of this stuff, don't worry,

6:24there's going to be a lot coming up,

6:26I just wanted you to see all of the different categories

6:30under All Services that exist within Azure

6:33as well as how to customize your Azure look

6:36and feel with a theme and how to navigate around Azure.

6:41So get to know what are in these categories;

6:44compute, networking, storage, web, containers, databases,

6:49analytics, AI, machine learning, resources for IoT,

6:53look at this, mixed reality, integration is personally

6:57my favorite section, and even identity and security are

7:01crucial parts to understand when it comes to working

7:04with and administering Azure.

7:06So this has been how to get started

7:09with Azure, understanding the point of the cloud,

7:12and how we can actually kick start our Azure journey.

7:15I hope this has been informative for you

7:16and I'd like to thank you for viewing.