I've never really looked at AWS Application Migration Service in a way of improving. There's always room for improvement, but I think we can improve the automation. It's already automated, but some of the migration tools require manual intervention. For example, if the agent fails, I need to push a few commands to keep pushing the migration. I suggest using AI and machine learning to predict and resolve compatibility issues during migration. I think that would actually help AWS. I've also faced challenges in migrating legacy systems to modern cloud architecture.

The schema conversion is not included in the document management services itself. Complex formations require additional steps. Furthermore, I noticed that performance can vary with the source and target load and network latency.

There are some features in the AWS Cloud Migration Factory solution that could be integrated directly into AWS Application Migration Service. Built-in refactoring features, dependency mapping, and cost optimization are some suggestions for improvements. Improved reporting & dashboards would also be beneficial.

Considering some of the cases we have, I think it is important to have more logs, and more details would be great because we have just logged on the client's side, but there weren't many details on the cloud. More logging during the migration would be great.

Live Migration has some issues with target setups. In the next release, Live Migration should reduce the disc size.

I do not see any improvements required for the CloudEndure.

We would like to have a disaster recovery feature included in this solution.

One drawback to using CloudEndure is that the default is to give one small, lightweight server, which is created in the cloud. From there, we need to switch over to the actual target server. My customers do not want to pay for both the lightweight instance and target instance when they only use one of them.