Here is a mapping of server hardware lifecycle to AWS EC2:
1. Start : On AWS - ec2-start-instances
Note: This is different then instantiating an instance for the the first time (from an AMI). The command for this is ec2-run-instances.
2. Reboot : On AWS - ec2-reboot-instances
(If a Linux/UNIX instance does not cleanly shut down within four minutes, Amazon EC2 will perform a hard reboot)
Note: Instance reboots are reboots of your virtual instance, and are equivalent to an operating system reboot. You can also reboot your instance from the operating system of your instance. More here: http://serverfault.com/questions/350581/differences-between-reboot-and-ec2-reboot-instances
3. Hard kill : ec2-stop-instances
4.AWS option : ec2-terminate-instances http://docs.aws.amazon.com/AWSEC2/latest/CommandLineReference/ApiReference-cmd-TerminateInstances.html
Note: This would be like throwing away a machine but keeping the hard drive.
Tagging instances can be done at anytime using the ec2-create-tags command.