Which AWS architecture and migration strategy will meet these requirements?
Use multiple AWS Snowball appliances to migrate the existing imagery to Amazon S3. Create a 1-Gb AWS Direct Connect connection from the ground station to AWS, and upload new data to Amazon S3 through the Direct Connect connection. Migrate the data distribution website to Amazon EC2 instances. By using Amazon S3 as an origin, have this website serve the data through Amazon CloudFront by creating signed URLs.
Create a 1-Gb Direct Connect connection from the ground station to AWS. Use the AWS Command Line Interface to copy the existing data and upload new data to Amazon S3 over the Direct Connect connection. Migrate the data distribution website to EC2 instances. By using Amazon S3 as an origin, have this website serve the data through CloudFront by creating signed URLs.
Use multiple Snowball appliances to migrate the existing images to Amazon S3. Upload new data by regularly using Snowball appliances to upload data from the network-attached storage. Migrate the data distribution website to EC2 instances. By using Amazon S3 as an origin, have this website serve the data through CloudFront by creating signed URLs.
Use multiple Snowball appliances to migrate the existing images to an Amazon EFS file system. Create a 1-Gb Direct Connect connection from the ground station to AWS, and upload new data by mounting the EFS file system over the Direct Connect connection. Migrate the data distribution website to EC2 instances. By using webservers in EC2 that mount the EFS file system as the origin, have this website serve the data through CloudFront by creating signed URLs.
Explanations:
This option uses AWS Snowball for initial bulk data transfer to Amazon S3 and a Direct Connect connection for ongoing uploads. Storing images in Amazon S3 and serving through CloudFront with signed URLs provides scalable, cost-effective distribution.
While feasible, this option relies solely on Direct Connect for both the initial 2 PB data transfer and ongoing uploads, which is slower and less efficient than Snowball for the initial migration of large data volumes.
This option proposes using Snowball continuously for regular uploads, which is impractical for frequent, ongoing data transfer. Direct Connect is more suitable for daily uploads, making this option inefficient.
Storing images in EFS rather than S3 is less cost-effective and lacks the performance and distribution benefits of using CloudFront with an S3 origin for large-scale, internet-facing access to frequently accessed imagery.
As I see it, the answer is:
Use multiple AWS Snowball appliances to migrate the existing imagery to Amazon S3. Create a 1-Gb AWS Direct Connect connection from the ground station to AWS, and upload new data to Amazon S3 through the Direct Connect connection. Migrate the data distribution website to Amazon EC2 instances. By using Amazon S3 as an origin, have this website serve the data through Amazon CloudFront by creating signed URLs.