Cost Effective

No up-front cost
We pay for what we consume

Secure

Service provider takes care of physical security
Allows very granular control on the security options on services

Scalability

Process of adding or removing resources
Increasing Size (Vertical Scalability)
Increasing Amount (Horizontal Scalability)

Elasticity

Auto Scaling (adding/ removing) of resource dynamically based on demand

Agility

The ability to quickly provision resources as per need
When using on premise resources this process could take up to months

Fault Tolerance

The ability of resources to remain up and running during component and service failure
This is achieved by maintaining multiple copies of the resources

Disaster Recovery

Creating copies of the required resources in two separate Azure Regions and setting up replication between them to so that they are up to date and then setting up an DNS between these two regions so that the traffic to the resource can be automatically redirected to the working region
This is an type of Fault Tolerance that is provided for dealing with major disruption to services that can be caused by Natural Disasters

High Availability

It is the metric that measures the uptime of the resource
Downtime can be caused due to planned maintenance or due to resource failure
Cloud providers at a minimum try to aim for four 9s availability