es

u
ig
r
Azure Web App Logging

d
o
R
an
Al
 es
u
ig
You get a set of logging features that are available for the Azure Web App.

r
The different types of logging that are available are

d
Application Logging – This captures log messages that are generated by your application code.

o
Web server logging – This records raw HTTP request data.

R
an

Logging
Al
 es
u
ig
Detailed Error Messages – This stores copies of the .htm error pages that would have been sent to the client

r
browser.

d
Deployment logging – These are logs when you publish content to an application.

o
You can also stream logs in real time.

R
an

Logging
Al
Al
an
Azure

R
o
d
r
ig
u
es
 Azure App Configuration

es
u
Central Service Encryption

ig
1
Provides a central service It provides encryption of
for managing application
3 sensitive information at rest

r
settings and feature flags . and in transit.

d
o
Frameworks
Managed

R
Azure App Configuration

It integrates with popular

This is a fully managed
service. 2 frameworks.

4
an
Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Container Instances

es
u
Deployment Public Access

ig
1
Provides a fast and easy Azure Container Instances
way to deploy containers.
3 can also get their own

r
Public IP and DNS name.

d
o
File shares
Infrastructure

R
Azure Container Instances

You can also persist data via

Here you don’t need to
manage the underlying 2 the use of Azure File shares.

4
an
infrastructure.
Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Container Groups

es
u
Collection Deployment

ig
1
This is a collection of The deployment of a
containers that get
3 container group is done via

r
scheduled on the same host a Resource Manager
machine. template or a YAML file.

d
o
File shares
Resources

R
Azure Container Groups

You can also persist data via

The containers then share
the lifecycle , resources, 2 the use of Azure File shares.

4
an
local network and storage
volumes.
Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Blob change feed

es
u
Purpose Records

ig
1
It provides an ordered , The change feed data is
guaranteed, durable,
3 stored in a contained

r
immutable, read-only log of named $blobchangefeed.
changes.

d
o
Record format
Operations

R
Azure Blob change feed

The records are stored in

Audit log of events to Blob
data – Create, Update, 2 Apache Avro format.

4
an
Delete.
Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Blob versioning

es
u
Versions Version ID

ig
1
Allows to maintain previous Each blob gets an initial
versions of a blob.
3 version ID. A new version is

r
set whenever the blob is
updated.

d
o
Feature Enable
Restore

R
Azure Blob versioning

You can enable or disable

You can then restore an
earlier version of the blob. 2 the feature at any point in

4
an
time.
Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Request

es
u
ig
The cost of database operations is measured in terms of Request Units.

r
d
Request Units is a blended measure of CPU, IOPS and memory.

o
R
Request Units are used no matter which API you are using for the Azure
an
Cosmos DB account.
Costing

Al
c
 Request

es
u
ig
The cost of reading a single item (1 KB) is 1 Request Unit.

r
d
Other operations have their own measure of charged request units.

o
R
For each operation, you can see the amount of request units that were used
an
for that operation.
Costing

Al
c
 Provisioned

es
u
ig
1 Here you provision the number of Request Units. You can always increment or decrement the

r
number of request units provisioned.

d
o
2 You are billed on an hourly basis.

R
an
3 You can provision throughput at the container or database level.
Costing

Al
c
 Serverless

es
u
ig
1 You don’t provision any throughput.

r
d
o
2 This is managed by Azure Cosmos DB.

R
an
3 You will be billed based on the Request Units you consume.
Costing

Al
c
 Autoscale

es
u
ig
1 Here the Request Units can automatically scale based o n demand.

r
d
o
2 The demand is checked both at the container and database level.

R
an
3 Great for mission critical workloads
Costing

Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Cosmos DB Partitions

es
u
Logical Partitions Identification

ig
Items in a container are The combination of the

r
divided into subsets called partition key and the item id
logical partitions. helps to uniquely identify the

d
item within the container.

Partition Key Size

o
The partition for the item is Partitions
Azure Cosmos DB Partitions

Each logical partition can grow

decided by the partition key up to 20GB.

R
that is associated with the
item in the container.
an
Limit
Item id
Each item also has an item id. There is no limit when it
This helps to uniquely identify comes to the number of logical
Al

an item in the partition partitions within a container.

c
 Partition

es
u
ig
Choose a property to be a partition key in which the value does not change.

r
d
The property should have a wide range of possible values.

o
R
Once you decide on the partition key for a container, you can’t change it.
an
Partition Key

Al
c
Al
an
Change

R
o
d
r
ig
u
es
 Azure Cosmos DB Change Feed

es
u
Record Changes Records

ig
The records for the change
Here changes to the container
feed are written to another

r
are recorded in the order that
container.
they occur.

d
Feed
Change Sorting

o
Azure Cosmos DB Change Feed

The feed consists of inserts

and updates. Deletes are not
Feed The change feed is sorted in
the order of modification

R
recorded. within each logical partition
key value.
an
Process Change Feed Throughput

You can process the change The same provisioned

feed with the use of Azure throughput can be used to
Al

Functions or a change feed read from the container

processor. containing the changes.
c
Al
Time

an
R
o
d
r
ig
u
es
 Azure Cosmos DB - Time to Live

es
u
Delete items Background task

ig
The deletion process is done as
Here items from the container
a background task that uses

r
can be deleted automatically
the left-over Request Units.
after a certain period of time.

d
Level
Time to For the item

o
Azure Cosmos DB – Time to Live

This can be set at the item

level or the container level.
Live For the TTL to work at the item
level, the setting should also

R
be defined at the container
level.
an
Deletion Process Benefit

Items will be deleted based on Here the process of deletion is

the Time to live that is set in managed by Azure Cosmos DB.
Al

seconds.
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Active Directory

es
u
What is it Identities

ig
1
This is a cloud-based You can manage identities
identity and access
3 such as users, groups and

r
management service. even applications.

d
o
Features
Usage

R
Azure Active Directory

Manage security aspects

This can be used as an
identity provider for Azure, 2 when it comes to your

4
an
identities.
Microsoft 365 and other
SaaS products as well.
Al
c
 Azure

es
u
ig
1 Free User and group

r
management.

d
o
2 Premium P1 Dynamic groups, hybrid
identities, self-service

R
password reset for on-
premise users.
an
License

3 Premium P2 Azure AD Identity

protection and
Al

Privileged Identity
Management.
c
Al
an
R
Dynamic

o
d
r
ig
u
es
 es
u
ig
Dynamic thresholds

r
Here Azure Monitor uses machine learning to check the historical behavior of metrics.

d
Based on the historical data, it can then identify patterns and anomalies that could indicate possible issues.

o
R

Azure Monitor
an
Al
 es
u
ig
Sensitivity

r
High – Here the alert rule will be triggered even for the smallest deviation.

d
Medium – Here you have more balanced thresholds and fewer alerts will be generated.

o
Low – Here alerts will only triggered on large deviations.

Azure Monitor
an
Al
Al
an
R
o
d
Application

r
ig
u
es
 Application Insights

es
u
Monitoring Applications

ig
This works for applications
This provides the feature of
hosted in Azure, on-premises

r
application performance
environments, or other cloud
management and monitoring
platforms.

d
of live web applications.

Aspects Application Integration

o
Here you can see aspects such Insights It has Integration with the
Visual Studio IDE.

R
as detecting performance
issues or any other issues.
Application Insights

an
Support Users

There is support for .NET, You can also see how users
Node.js, Java and Python. interact with your application.
Al
c
 es
u
ig
How does it work

r
You can install a small instrumentation package (SDK) for your application. Or use the Application Insights agent.

d
You can instrument web applications, background components and JavaScript in web pages.

o
The telemetry data sent by Application Insights has very little impact on the performance of your application.

Application Insights
an
Al
Al
an
R
o
d
Application

r
ig
u
es
 es
u
ig
Users, Sessions and Events

r
Users – Here you can see how many people have used your application and its features.

d
Session – You can see sessions of user activity. This includes certain pages and features of the application.

o
Events – This gives a view of how often certain pages and features have been used in the application.

Application Insights
an
Al
 es
u
ig
Funnels – Here you can have multiple stages like a pipeline. And then you can see how users are progressing

r
through your application as an entire process.

d
Cohorts – This is a set of users, sessions, events or operations that have something in common. It helps to analyze

o
a particular set of users or events.

R
Impact – Here you can see how load times and other aspects of your application impact the conversion rate for

Application Insights
your application.
an
Al
 es
u
ig
Retention – Here you can analyze how many users return back to your application.

r
User flows – This can help in answering useful questions such as

d
1. What do users click on a page within the application

o
2. Where are the places within the application that users churn the most from the site.

Application Insights
3. Are there places in the application where the users repeat the same action over and over again.
an
Al
Al
an
R
o
d
Application

r
ig
u
es
 Availability Tests

es
u
Tests Endpoint

ig
1
You can define tests that You can setup tests against
can monitor the availability
3 an HTTP or HTTPS endpoint

r
and responsiveness of an that’s accessible over the
application. public Internet.

d
o
API
Web Requests

R
You can also test your REST
Availability Tests

This feature can send web

requests to your web 2 API’s as well.

4
an
application from different
points across the world.
Al
c
Al
an
Azure

R
o
d
r
ig
u
es
 Azure Cache for Redis

es
u
Data store Advantage

ig
1
This is an in-memory data The advantage of server
store that is based on the
3 memory is that you can

r
Redis software. write and read data quickly.

d
o
For the application
Server memory

R
It helps to provide low
2
Azure Cache for Redis

Here you can store latency to your data and

frequently accessed data in
4
an
high throughput.
server memory.
Al
c
Al
an
R
Azure

o
d
r
ig
u
es
 es
u
ig
You can enable caching for your endpoint routes.

r
When delivering large files, Azure Front Door request the file from the origin in chunks of 8 MB.

d
When a chunk of data is received, the data chunk is cached and immediately sent to the user.

o
R
an

Azure Front Door

Al
 es
u
ig
You have different query string behaviors when it comes to caching

r
Ignore query strings – Here the query strings from the requestor is passed to the origin for the first request and

d
then the asset is cached. All further requests ignore the query strings until the expiry of the cached asset.

o
Cache every unique URL – Here each URL is treated as unique and cached separately.

R
Specify cache key query string – You can decide to include/exclude specified parameters to determine what gets
cached.
an

Azure Front Door

Al
Al
an
Topic

R
o
d
r
ig
u
es
 es
u
ig
With the help of filters, subscribers can decide which messages they want to receive.

r
This can be defined with the use of rules.

d
Each rule has a filter condition.

o
The message will contain a property called RuleName that shows the matching rule.

R
an

Topic filters
Al
 es
u
ig
Boolean filter – This is either the TrueFilter or FalseFilter. This can cause all messages to be sent to the

r
subscription or for the subscription to receive no messages.

d
SQL filters – Here SQL like language can be used to evaluate the messages user-defined or system properties.

o
Correlation filters – Here also conditions can be matched against the messages’ user or system defined

R
properties.

Correlation filters are more efficient that SQL filters.

an

Topic filters
Al