Azure API Management (APIM) Part 2

Video Activity
Join over 3 million cybersecurity professionals advancing their career
Sign up with

Already have an account? Sign In »

14 hours 28 minutes
Video Transcription
Hello, Siberians. Welcome to lessen six points to off Model six. Off this constituted is a trees were one Microsoft as your tactic design
here the leading objectives they'll be covering in this video.
We'll start by introducing you to the concept off visions as it concerns a P I am
wouldn't cover the concept of visions.
We'll try to explain the differences between visions and vision so you know when to use, which
would talk about policies when it comes to a P. I am what policies can be used for on the different options that are available to us
on Finally would talk about architectural decisions off a P. I am especially in relation to scalability, use, cases, availability and pricing. Let's get into this.
So whenever I hate behinds, I in production and they have been used by developers,
we eventually needs to make changes to those AP eyes on. It could be that we have had in a new operation at the back and service now supports. Or maybe we're retiring the legacy operation that's no longer supported.
Whatever reason east for having to make changes to our happy highs,
it is best not to disrupt collars off. Ah, happy when making changes.
It's also useful to release changes in a controlled way. Maybe even let's develop us know about the changes we've made.
We can do this in azure AP a management using something called Divisions.
So this is our vision works
you can see on the screen. We have an existent revision off a P I that develop as a using
at the moment.
First we had a new revision. In the back end,
we'll make the changes that we wants to make to this new vision. Maybe adding a new post operation
would test the changes, make sure everything is working fine.
And when we're ready to go life, we simply make a switch
and our vision that will create that. We make it to the cover Entravision On off course. We can go back if we need to.
Its importance than not this
which revisions providers juice went to deploy and when they wants to go life.
So let's talk about visions
with versions. There are times when it is impractical, tohave all the colors of our baby Hi to use exactly the same vision.
So maybe we have different years that they're subscribing to. Maybe there's a free tier where we don't make all oppressions available. And maybe there's another tier where people can pay for under we make more precious available. That could be the reason.
Or maybe we want our users to have the flexibility to adopt new releases of our Happy Ice Wendy. Already, we don't just want to force it on them,
And this is where visions can help us
with visions. We can publish multiple visions off our happy high at the same time,
and this is how vision works
again. We have an existent version of I a P high that developers are using at the moment.
What we do is we add a new vision that supports, knew our positions or maybe knew a way to use our service
with them. Published this new vision
US A new vision off on existent AP Hi
within a p I management
on developers can opt in two statues in the New vision. Wendy already so it is important to note the following when it comes to differences between visions and visions,
which revision
providers choose went to deploy when we go life and we only have one
in production
with visions consumers choose went to adopt so we can have multiple versions in productions, and then the consumers can choose when they want to adopt a particular vision.
Let's talk about a P I am policies for a minute
policies I powerful capability off AP ahem. That allows publisher to change the behavior off the A P I to configuration with doubt right in court
on their two men types of policies.
They're inbound policies that we can use to modify the request that is coming from the client before the request is sent to the back in service. And all of this happens within the get way components that we discussed in the previous lesson.
For example,
we can set an impound policies of the strict incoming requests to specified I P addresses.
So if the request is not coming from one of allowed I P addresses, it simply dropped by to get way
all we can specify an impound policy to save request coming from the client, we can had its limits to them the amount off a loud cause within the time frame.
This can help to maybe put text from things like D does attack right? The distributed denial of service attacks by Harding rates limits.
And then we have our bomb policies that we can use to modify responses that I received from the back and service on. We can make modifications before their sense to the client. For example, we can have an hour. One policy to hide said in http head. Is
that a comment from the responses from the back in service?
And we do not want this editors to go to the requesting client so we can use our bomb policies. Toe I did these edits auto the lips. They said this and I'll be showing you in the demo.
Finally, let's talk about some key architectural decisions off the A P I am service. Let's start by talking about use case. First of all,
so we earlier mentioned that they're different price in tears off a Pia. Him Andi have different use cases. For example, are we sure you The demo using the developer tear, which supports a lot off the features and functionality, is that premium tier support, but do not use is in production. Please.
It has no s Ellis apart. So it's for non production use cases for development is cases off evaluation Use cases.
Consumption use cases is mainly around scale, right? So whenever want maybe more of a more flexibility in pricing paper use basis and also in terms off when not locked down to the maximum amount off unit
won't get to that in a minute.
And we have the basics here, which is entry level production use cases. Stand out with this medium volume production. Just cases and premium for I've volume or enterprise production use cases.
Let's look at scalability.
So when it comes to scalability, you can see that
the Marty demand you move to the right. When it comes to the tear the mosque kill, you have, for example, if you look at the cash on this is talking about cash per unit. So number one, let's live different talk about units to begin, which can see that the developers say you can only have one unit, so you can't really skill, but you can see that basic. You can have two unit
standard. You can have up to four unit and premium. You can have up to 10 units per vision, and if you need more you could re quite make it requires to Microsoft support, and they could
increased that for you.
But you can say that the amount of cash that's available unit also varies. Also, you can see the estimated maximum true put, but you need varies depending on the tier
that you're using.
Finally, let's discuss pricing and especially when it comes to price. And you can see that for the developer, basic standard and premium would pay per hour. And that's per unit
on When it comes to S L. A. You can see the S and the state. But when it comes to
pricing for consumption, he considers its pecol. So we attach Paco and the 1st 1 million cause that will make that free.
I hope that you've got in some useful information from this lessons on a p A management. Thanks very much for watching on. I'll see you in the next lesson
Up Next
AZ-301 Microsoft Azure Architect Design

This AZ-301 training covers the skills that are measured in the Microsoft Azure Architect Design certification exam. Learn strategies to plan for the exam, target your areas of study, and gain hands-on experience to prepare for the real world.

Instructed By