Data Migration from On-Prem Tape Archive

0

Hi,

Can I migrate data on tape archive on-prem to AWS using AWS Datasync? This is a complete migration, NOT extending your storage. The amount of data is 10TB with max 500KB file size, bandwidth 500Mbps. I know there is the option of Snowball but on this case I want to know more about Datasync. Once agent is installed on the on-prem VM, can it discover and read the tape storage with NFS protocol or further configuration are required to present the tape storage to the Datasync Agent?

Thank you,

Amanuel.

Aman
asked a year ago274 views
1 Answer
0

Amanuel,

If I understand your question correctly it comes in two parts.

  1. Can you use DataSync to migrate data from your network to AWS S3, and probably one of the Glacier categories of storage.
  2. Can you use DataSync to discover and read the tape storage.

For question #1, the answer is YES. you can definitely setup an AWS DataSync agent on your network and attach to a NAS or File Server as a Mounted File System. Data transferred to this File system will be automagically transferred to the Selected AWS Destination, to include a variety of S3 storage classes, EFS, FSx for windows or FSx for Lustre.

For question #2, the answer is NO. For a Tape Backup system, you will need some software or utility to transform the data from the Tape backup to files, and then transfer the file data to the DataSync Mounted File System discussed in Question #1.

I hope this helps Cam

Amendment to respond to 3rd question. Aman, To address your additional question - as in most IT questions - it ‘depends’.

10 TB is not THAT large, and your network bandwidth would accommodate direct transmission in a matter of hours during off-peak usage, so both options are viable in your use-case.

AWS Snowball (and related services) provides a mechanism to migrate large quantities of data to AWS, especially when your ability to network to AWS is bandwidth constrained.

If the intent is to migrate the tape data to AWS S3, but you are NOT seeking to replicate a Tape Library - just a data archive, then it makes sense to approach this with a DataSync Agent as a Mounted File on a dedicated server network connected to AWS. To accomplish this, you will need an extraction utility to convert the Tape Library data into files.

If you use AWS Snowball, you will still require an extraction utility to convert the Tape Library data into files on the AWS Snowball Appliance. Your Result? A Collection of File-Objects in S3, probably utilizing one of the Glacier or Deep Archive S3 Services, depending upon business requirements.

IF the intent is to replace the On-Prem Tape Library with an AWS Virtual Tape Library in AWS S3, then you would use Storage Gateway Tape Gateway, and it has a utility to transform existing Tape Libraries into its AWS S3 backed Virtual Tape Library. You can convert this to an AWS Snowball Appliance with Storage Gateway, OR you can do this direct to a Storage Gateway Tape Gateway, Networked to AWS S3.
Your Result? An AWS S3 backed Virtual Tape Libraries leveraging Storage Gateway Tape Gateway, probably utilizing one of the Glacier or Deep Archive S3 Services as the backend storage.

Again, i hope this makes sense and that it helps?

answered a year ago
  • Great answer Cambeeler! For reason stated in the second question, does it make it preferable to use Snowball? For this case we have 10TB of data stored in tape library 500KB MAX file size can be transferred over 1Gbps utilizing 70-80% off-peak hours.

    Thank you.

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions