A company has about 1 million .csv files on a VM on premises totaling 10 TB now and growing by about 1 TB per week. They must back up the data to AWS daily, and apply custom filters to include only specific source directories. A Direct Connect link is available. Which solution provides automated daily backups with the least operational overhead?

AWS DataSync is the most suitable solution because it's designed for automated, scheduled data transfers between on-premises storage and AWS services like S3. Installing the DataSync agent on the on-premises hypervisor allows it to directly access the VM's file system. DataSync tasks can be configured with filters to include specific directories, meeting the requirement for custom filtering, and can be scheduled for daily execution, ensuring automation and low operational overhead. Using the S3 CopyObject API directly would require custom scripting for filtering, scheduling, and handling large numbers of files and incremental changes, leading to higher operational overhead. AWS Backup is primarily for backing up AWS resources, not directly for on-premises file systems. While Snowball Edge can handle the initial bulk transfer, it doesn't provide an automated daily incremental backup solution for ongoing changes; DataSync is still needed for that.