A company is migrating an on-premises data center to AWS. The data center hosts a storage server that uses NFS and contains 200 GB of data. The migration must not interrupt existing services, and multiple AWS resources must be able to access the data via NFS. Which combination of steps will meet these requirements MOST cost-effectively? (Choose two.)

Creating an Amazon EFS file system is correct because EFS provides a scalable, shared NFS file system that multiple AWS resources can access concurrently, directly fulfilling the requirement for NFS access in the cloud. Installing an AWS DataSync agent and using a DataSync task is correct because DataSync is designed for efficient, secure, and automated data transfer from on-premises to AWS, minimizing service interruption during migration. FSx for Lustre is optimized for high-performance computing workloads, not general-purpose NFS sharing, making it less cost-effective for this scenario. An S3 bucket is object storage and does not support the NFS protocol directly, failing the access requirement. Manually copying data is inefficient, prone to errors, and would likely cause service interruption, failing the non-interruption requirement.