Application Migration Service (AMS): Disaster & Recovery

Have you ever been in a situation like the following?

Manager:  Hey team, management wants us to test our disaster recovery plan.

Team Member: When do they need it done by?

Manager: They are hoping within the next two weeks if that is possible…

Team Member: We have never done a test before!

Team Member: We don’t even know if we can fail back if we perform one.

If you have found yourself in this situation, this blog will help solve this issue. Let’s get started!

What is Application Migration Service?

Application Migration Service is an AWS service for disaster recovery (DR), backup, and migrations. At a high-level, Application Migration Service “performs block-level replication” via an installed agent that can be deployed onto servers. This agent then replicates to an Amazon EC2 Instance which has an EBS volume connected. This agent copies the associated data from servers to staging area in AWS and replicates in real time.

Why is Application Migration Service Important?

Unlike running a “ad-hoc” disaster recovery scenario like the above, a simple conversation between a manager and team member, and then you can be prepared to have a DR using Application Migration Service at any time. With its robust features Application Migration Service solves these three problems:

Problem 1: How do I afford running my memory/cpu packed host in the cloud?

Solution: Machines feature

Within the migration service you can setup a replication server to be a specific size. If you are running in a DR mode and you can handle not having the same machine size for cost saving reasons, you can map the machine to run at a lower tier until the disaster is over. Then you can simply “recover” back to your original setup.

Problem 2: How do I document my disaster recovery plan?

Solution: Recovery Plans feature

With the ability to define a recovery plan within Application Migration Service, you have your documentation not only available but viewable for any person to see. You could specify steps (order) that must be taken before specific machines are moved over. Example: if your application servers must wait for a database to come online you can add that step.

Problem 3: How can I run a full-blown DR test?

Solutions: Test Mode feature

With full capabilities to test, you can run a test scenario and offload part of your application workload to the newly created environment. You can watch as the resources are spun up and then you can terminate them. With this ability, you don’t even have to have a full “switchover” as a test, as you can see in real-time the DR setup from the console.

Final Thoughts

Using Application Migration Service for disaster recovery is essential to keeping your data secure and it’s important to have the knowledge around AMS when you get that last minute request to do a DR test. My hope is that you can refer back to this blog in a time of need. The good news is that disaster recovery is only one aspect of AMS, you can also use it to migrate services to AWS, stay tuned for my next blog on the topic. If you’re looking for Application Migration Service support, contact us. We’re an AWS Advanced Consulting Partner with a multitude of expertise in AWS technology.