Define storage requirements and transfer data
After workloads are selected for migration, analyze what data is required by the workload, and the capacity and performance requirements of the storage subsystem. This evaluation process can significantly delay the process of migrating to AWS, so it is recommended that you begin the process of identifying workload data dependencies early.
You also need to analyze whether data can be migrated to the cloud once, or if it needs to be synchronized with the on-premises storage. In addition to increasing workflow turnaround time, copying large volumes of data between on-premises sites and AWS has cost implications. Moving data into AWS is free, but moving data out is not. For these reasons it is desirable to minimize data transfer back and forth between on-premises sites and AWS.
AWS provides several ways to migrate on-premises data to the cloud that will initially
store the data in Amazon Simple Storage Service
Although tools and semiconductor flows require POSIX-compliant file systems and do not
currently support object storage, S3 can be used as the back-end for creating an AWS-managed,
POSIX-compliant file system using Amazon FSx for
Lustre
The method of transfer largely depends on how much data you need to move. If you have a
relatively small amount of static data and you have a fast, reliable internet connection, you
may be able to use your internet connection. Leveraging AWS Direct Connect allows for
high-bandwidth ingress and egress to AWS. AWS
DataSync
If you have large amounts of library, design, or simulation data that requires an initial
one-time transfer, consider using AWS Snowball
