The Network Administrator Survival Kit

Elastic IP addresses are static Public IP addresses that are allocated to AWS Accounts.You can associate it to an EC2 instance or network interface. You need to manually allocate an Elastic IP address to your account. They are not automatically assigned. By default, you can allocate up to 5 Elastic IP addresses. To allocate more, you need to submit a request to AWS. An Elastic IP Address can easily be disassociated from an EC2 instance or network interface and re-associated with another instance or interface. However, if you were to release it from your account, it would go back to the pool of public IP addresses in AWS that can be assigned to other AWS accounts. There is no guarantee that if you were to allocate a new IP address to your account you would get the same one you had previously. BE CAREFUL and do not just release IP addresses because you are not using them. Disassociating from EC2 instances and network interfaces is fine to do as the Elastic IP addresses still remain in

AWS has many data centres around the world. Amazon has created AWS Regions in different countries and have setup data centres in these locations. There are usually two to three data centres grouped into a particular region. Each region is completely independent of each other and communicate via the Public Internet. A customer would choose to create their Amazon EC2 instances, for example, in a particular region that would be closest to where their clients will be accessing them. Each data centre in a region is called an Availability Zone and are physically isolated from each other. They are isolated enough that should a major catastrophy occur in the location where a particular data centre resides, the other data centres in that region would be unaffected. The only exception would be if a moon sized object hit that region. If that happens, it would just be bad luck. Each Availability Zone in a Region are linked via low-latency private network connections. This allows an infras