Soru

Zorluk: OrtaLarge-Scale Data Transfer using Snow Family, DataSync, and Transfer Family

An enterprise needs to migrate 450 TB450\text{ TB} of medical imaging data from an on-premises NFSv3 storage system to an Amazon EFS file system. The migration must be completed within a strict 3-week3\text{-week} window. The enterprise has a dedicated 500 Mbps500\text{ Mbps} internet connection available for this migration. All data must be encrypted at rest throughout the migration process. Which strategy should a solutions architect recommend to meet these requirements?

  1. Order multiple AWS Snowball Edge Storage Optimized devices, specifying an AWS Key Management Service (AWS KMS) customer managed key for encryption. Copy the data to the devices using the NFS interface, ship the devices back to AWS to import the data into an Amazon S3 bucket, and then run an AWS DataSync task to copy the data from the S3 bucket to the Amazon EFS file system.Cevap
  2. B
    Order multiple AWS Snowball Edge Storage Optimized devices, specifying the default AWS-managed KMS key for Amazon S3 (`aws/s3`) for encryption. Copy the data to the devices using the NFS interface, ship the devices back to AWS to import the data into an Amazon S3 bucket, and then run an AWS DataSync task to copy the data from the S3 bucket to the Amazon EFS file system.
  3. C
    Deploy AWS DataSync agents on-premises, configure the agents to connect to the Amazon EFS file system, and run a DataSync task to transfer the data directly from the local NFS file system to the target Amazon EFS file system over the 500 Mbps500\text{ Mbps} internet connection.
  4. D
    Order multiple AWS Snowball Edge Storage Optimized devices, specifying an AWS Key Management Service (AWS KMS) customer managed key for encryption. Copy the data to the devices using the NFS interface, ship the devices back to AWS, and configure the import job to write the data directly to the target Amazon EFS file system.

Cevap

Order multiple AWS Snowball Edge Storage Optimized devices with a customer managed key, copy data using the NFS interface, ship back to AWS for S3 import, and then use AWS DataSync to copy to Amazon EFS.
The correct strategy uses AWS Snowball Edge Storage Optimized devices to physical transfer the data because the network bandwidth (500 Mbps500\text{ Mbps}) is insufficient to transfer 450 TB450\text{ TB} within the 3-week3\text{-week} deadline. A customer managed key is specified since default AWS-managed keys are not supported for Snowball jobs. Because Snowball Edge cannot import directly to Amazon EFS, data is imported into Amazon S3 first, and then AWS DataSync is used to efficiently replicate the data from Amazon S3 to Amazon EFS.

Adım Adım Çözüm

1
Calculate the minimum network transfer time.
At 500 Mbps500\text{ Mbps}, transferring 450 TB450\text{ TB} of data requires at least 7.2×106 seconds7.2 \times 10^6\text{ seconds} (approximately 83.3 days83.3\text{ days}), which exceeds the 3-week3\text{-week} (21-day21\text{-day}) timeline limit.
To verify if online network transfer is a viable option.
2
Select the appropriate migration family and determine destination compatibility.
AWS Snowball Edge is chosen for offline physical transport. Since Snowball Edge cannot import directly into Amazon EFS, Amazon S3 must be used as the initial import destination.
To plan the physical transport and target endpoints based on AWS service capabilities.
3
Verify KMS encryption settings for the Snowball job.
Identify that a customer managed key must be created and specified. Default AWS-managed KMS keys are not supported.
To ensure compliance with the requirement of keeping data encrypted at rest during the physical transfer.
4
Determine the migration step from the intermediate store to the final destination.
Run an AWS DataSync task to copy the imported data from the Amazon S3 bucket to the target Amazon EFS file system.
To move the migrated data from the intermediate S3 staging bucket to the required Amazon EFS production destination.

Anahtar Kavram

Selecting the correct hybrid migration tool based on dataset size, available bandwidth, timeline constraints, target service support, and security requirements.
Bu soruyu puanla