Scaling is a technique to scale web applications in conditions when there is a higher load or a lesser load, and to keep the performance or the user response time consistent.

With Azure Web Apps, Mobile Apps, API Apps, and Logic apps, scaling can be done in four ways, which are as follows:

• Scale up: To increase the resources such as CPU, memory, and other attributes based on the pricing tier of the underlying VM
• Scale down: To decrease the resources such as CPU, memory, and other attributes based on the pricing tier of the underlying VM
• Scale out: To balance the load by diverting the traffic of incoming requests to multiple VMs
• Scale in: To reduce the number of VMs if the load weighs down

Scaling can be done manually or scheduled, or can be done based on the usage of CPU, memory, or disk space. Scaling can be done easily by opening the Web App blade in the Azure portal and selecting the scale up and scale out options.