AWS Import/Export
Developer Guide

Creating Amazon Glacier Import Jobs

An Amazon Glacier import job creates a virtual copy of your device as a single image. The entire image for one device is uploaded to one Amazon Glacier archive in an existing Amazon Glacier vault. One archive holds the contents of one device. The device's file system is never mounted. You send a separate CreateJob request for each device.

If you want the lower cost storage provided by Amazon Glacier, but you want your individual files to be stored as separate objects, first import your files to Amazon S3, then use Amazon S3 Object Lifecycle Management to transition your files to the GLACIER Storage Class. When you transition objects to the GLACIER storage class, Amazon S3 internally uses Amazon Glacier for durable storage at lower cost. Amazon S3 objects that are associated with the GLACIER object class are visible and available only in Amazon S3. For more information, see Object Lifecycle Management.

If you are using AWS Import/Export to upload data to Amazon Glacier, we assume you are familiar with the Amazon Glacier concepts of vaults and archives. To learn more about Amazon Glacier, go to Amazon Glacier.

Amazon Glacier imports complete images of the data on your device. It does not import each file individually. This functionality means that retrieving any of your data from Amazon Glacier requires restoring the complete device image, which can be expensive.

If you want to access your data on a file-by-file basis and store your data in Amazon Glacier, then consider performing an Amazon S3 import instead, and adding a lifecycle policy to your bucket to archive individual objects to Amazon Glacier. For more information on how to create lifecycle policies, see Lifecycle Configuration for a Bucket without Versioning.


You can only perform an Amazon Glacier import from devices of 4 TB in size or smaller.

For more information, see Guidelines and Limitations

Import to Amazon Glacier Process

Import Job Process

1 You copy your data to your storage device. Do not send the only copy of your data. AWS will erase your device. For added security, we strongly recommend using encryption. AWS does not decrypt your data.
2 You create an file that contains the accessKeyId and secretKey associated with your AWS account. For more information, see Save Your Credentials to a File.
3 You create an import manifest file that specifies the Amazon Glacier region and return shipping address. For more information, see Import to Amazon Glacier Manifest File Options.

You initiate an import job by sending a CreateJob request that includes the manifest file. You must submit a separate job request for each device.

Your job expires after 30 days. If you do not send a device, there is no charge.

You can send a CreateJob request using the AWS Import/Export Tool, the AWS Command Line Interface (CLI), the AWS SDK for Java, or the AWS REST API. The easiest method is the AWS Import/Export Web Services Tool. For details, see

Sending a CreateJob Request Using the AWS Import/Export Web Service Tool

Sending a CreateJob Request Using the AWS SDK for Java

Sending a CreateJob Request Using the REST API


AWS Import/Export sends a response that includes a job ID, shipping instructions, and a PDF file.

The PDF file contains job information, shipping instructions, and a barcode.

You must submit a separate job request for each device, and each job request generates a unique barcode.

AWS Import/Export validates the barcode on your storage device before starting the data load. If the barcode is invalid (for example, it doesn't match your device), it's missing, or it's illegible, AWS Import/Export will not perform the data load and we will return your storage device.


You print the PDF. For an example of the PDF, see The Import PDF File.

You cut out the barcode portion of the printed PDF, and tape it to your device. You must send the device to the address indicated in the shipping instructions.


Attach the signature barcode securely to your device, with tape on all four sides. Do not delete, alter, or obscure the signature barcode in any way. If the signature barcode is separated from your device, we cannot validate it and we will return your device without performing the data load.


AWS Import/Export uses the bar code to identify and authenticate your data load.

Once your storage device arrives at AWS, your data load typically begins by the end of the next business day. The time line for loading your data depends on a number of factors, including the availability of an import station, the amount of data to load, and the data transfer rate of your device.

8 Following the import operation, AWS will erase the device, even if the import was not successful. In cases where we are unable to erase the data on the device, we will schedule it for destruction. In this case, our support teams will contact you over the email address specified in the Manifest file.
9 We repack your storage device and ship it to the return address you provided in your manifest file. We do not ship to post office boxes.