How can you combine Capacity Reservations with Reserved Instances to achieve both availability and cost optimization?
Example
Suppose you need 5 EC2 instances continuously for 1 year.
You can:
- Create a Capacity Reservation for 5 instances → ensures the capacity is available in the required AZ.
- Purchase Reserved Instances matching those instances → reduces the effective EC2 compute cost.
What happens to your billing if you provision a Capacity Reservation but leave the instance unlaunched or stopped?
If you create an EC2 Capacity Reservation but don't launch an instance—or the instance is stopped you still pay for the reserved capacity at the equivalent On-Demand rate.
AWS starts billing as soon as the Capacity Reservation is provisioned and continues until you cancel/it expires.
Is Dedicated Host a cheaper version of Spot?
What is block storage?
What does "attach EBS to EC2" mean?
Attaching an EBS volume to an EC2 instance means virtually plugging a hard drive into a computer.
Think of your EC2 instance as a desktop computer tower, and the EBS volume as an external USB hard drive or an internal SSD. When you "attach" it via the AWS console or command line, you are telling AWS to run a virtual cable from that specific storage drive into the virtual drive slot of your server.
Once attached, the EC2 instance's operating system will immediately detect that a new piece of
storage hardware has been plugged in.