Auto Scalability

Auto Scalability means scaling in/out or up/down automatically or dynamically as per the conditions or we can configure depending upon the forecasted demand for our business.

Consider the scenario of deploying our website in AWS Amazon EC2 Instance and our website generally experience more traffic on weekends than that of on weekdays. So, we can configure to add more servers for our site to handle the increased traffic on weekends and remove them on weekdays. we pay for the additional servers only on weekends thereby reducing the costs of usage.

In case of sudden surge or non-forecasted load on servers, we can use dynamic scaling for our website 

e.g. Whenever CPU load becomes greater than 70%, scale out and add two more servers. If CPU load is less than 30%, then scale in by removing superfluous servers.

Similarly, we can auto-scale depending on requirements by keeping other constraints like above.

Vertical scaling (increasing capacity of same existing resource) cannot be done without stopping the instance.


Go to Next Post - Cloud Service Models

Go to Previous Post - Characteristics of Cloud Computing

Characteristics of Cloud Computing

These are the main characteristics for cloud computing:

1) On Demand & Self Serviced

Customer should be able to provision the services without human interaction on cloud service provider side. Launch or terminate the usage of resources at any time without the necessity of manual intervention from service provider.

2) Measured Service

The quantity of resources or services that are being utilized should be measured and be transparent to both sides. Customers pay only for what they use.

3) Resource Pooling / Multi Tenancy

All resources are pooled to serve multiple consumers using multi-tenant model. Physical & Virtual resources are assigned and reassigned as per consumer's demand.

Virtualization on top of physical resources of service provider is key for achieving this multi-tenancy.

4) Elasticity / Scalability

Capabilities or Configurations should be flexible to increase or decrease as per demand (preferably automatically). Adding or removing capacity whenever needed.

Resources can be scaled vertically or horizontally.

  • Vertical Scalability
Increasing or decreasing capacity of the same resource.

  • Horizontal Scalability

Adding or removing more resources of required type.


Go to Next Post - Auto Scalability


Go to Previous Post - AWS Post number0