AWS Import/Export FAQs

AWS Import/Export accelerates moving large amounts of data into and out of AWS using portable storage devices for transport. AWS transfers your data directly onto and off of storage devices using Amazon’s high-speed internal network and bypassing the Internet. For significant data sets, AWS Import/Export is often faster than Internet transfer and more cost effective than upgrading your connectivity.

Q. What can I do with AWS Import/Export?

AWS Import/Export makes it easy to quickly transfer large amounts of data into and out of Amazon S3. You can use AWS Import/Export for:

Data Migration – If you have data you need to upload into the AWS cloud for the first time, AWS Import/Export is often much faster than transferring that data via the Internet.
Offsite Backup – Send full or incremental backups to Amazon S3 for reliable and redundant offsite storage.
Direct Data Interchange – If you regularly receive content on portable storage devices from your business associates, you can have them send it directly to AWS for import into your Amazon S3 buckets.
Disaster Recovery – In the event you need to quickly retrieve a large backup stored in Amazon S3, use AWS Import/Export to transfer the data to a portable storage device and deliver it to your site.

Q. How can I get started with AWS Import/Export?

To start using the service, click the “Sign up for AWS Import/Export” button on the AWS Import/Export detail page. You must have an Amazon S3 account to access AWS Import/Export; if you do not already have one, you will be prompted to create one when you begin the AWS Import/Export sign-up process.

Q. When should I consider using AWS Import/Export?

If loading your data over the network would take 7 days or more, then you should consider using AWS Import/Export. To determine the best case for loading your data over the network use this formula:

Number of Days = (Total Bytes)/(Megabits per second * 125 * 1000 * Network Utilization * 60 seconds * 60 minutes * 24 hours)

For example if you have a T1 connection (1.544Mbps) and 1TB (1024 * 1024 * 1024 * 1024 bytes) to move in or out of AWS the theoretical minimum time it would take to load over your network connection at 80% network utilization is 82 days.

Available Internet Connection	Theoretical Min. Number of Days to Transfer 1TB at 80% Network Utilization	When to Consider AWS Import/Export?
T1 (1.544Mbps)	82 days	100GB or more
10Mbps	13 days	600GB or more
T3 (44.736Mbps)	3 days	2TB or more
100Mbps	1 to 2 days	5TB or more
1000Mbps	Less than 1 day	60TB or more

Q. How long does it take to load my data?

AWS typically begins loading your data the business day following receipt of your package. Customers with single drive USB devices with large files have achieved data loading rates over 20 MB/sec. eSATA devices have achieved rates over 40 MB/sec. For more details regarding data loading speeds and cost see the AWS Import/Export Calculator.

Q. Which factors influence data loading speed?

The factors that have the biggest influence on data transfer rates are device speed and average file size. Though other factors such as file system type and number of files per directory may impact the time it takes to load your data. Faster devices and larger files will tend to result in decreased transfer times. For more details regarding data loading speeds and cost see the AWS Import/Export Calculator.

Q. How much data can I load?

The total amount of data you can load using the AWS Import/Export is limited only by the capacity of the devices that you send to AWS. Individual files, which will be loaded as objects in Amazon S3, may range up to 5 gigabytes in size.

Q. How much data can I fit onto my device?

It depends. The amount of data that will fit on your device is determined by the specific exported file set, your formatted device block size, the amount of file system overhead, and actual formatted capacity. For example, a 1TB drive with a raw capacity of 1,000,000,000,000 bytes, when formatted, could be left with only 910,000,000,000 usable bytes. Block size could be an issue for file sets with smaller files. If your device is formatted with 4KB block sizes, but your files are all 1KB, you would have 3KB of unusable space per file. To estimate how many bytes block size could potentially impact, multiply the number of files for export by your formatted block size.

Q. I am using AWS Export. How many devices should I send to AWS?

That depends on the size of your devices and amount of data to export. You can figure out how many jobs you need by using the CREATE EXPORT PLAN command, which includes a manifest file and your device sizes. AWS will respond back with a set of manifest files that splits your export across your devices. The CREATE EXPORT PLAN command incurs Amazon S3 charges for LIST operations.

Q. Can I use AWS Import/Export to transfer data to Amazon EC2 or another AWS product?

Yes, once your data is in Amazon S3, it can be accessed by any AWS product. Data loaded through AWS Import/Export is treated just like data loaded through the network.

↑ Top

Billing

Q. How will I be charged and billed for my use of AWS Import/Export?

AWS Import/Export let’s you pay only for the resources that you use. Pricing is per device and per data-loading-hour. Partial data-loading-hours are billed as full hours. Data transferred between AWS Import/Export and Amazon S3 is free of charge (i.e., $0.00 per GB). Usage for Amazon S3 is billed separately from AWS Import/Export.

For AWS Import/Export pricing information, please visit the pricing section on the AWS Import/Export detail page. For Amazon S3 pricing information, please visit the pricing section of the Amazon S3 detail page.

Q. What charges am I responsible for?

You are responsible for purchasing your storage device(s) and the shipping charges to AWS. AWS will pay reasonable return shipping charges, but you will be responsible for any return shipping expenses that AWS determines to be unreasonable or excessive.

Q. When will I be billed for my use of AWS Import/Export?

The device fee is charged when your package arrives at AWS. The data-loading-hour billing starts once AWS Import/Export begins transferring data to or from your storage device. Billing ends when the data load is completed. If there are any interruptions to the data load, you will not be charged during the period of interruption.

↑ Top

Using the Service

Q. Do I need any special software?

No, but the AWS Import/Export resources include an example Java application for generating signatures. You should either create your own signature tool or download the Java Standard Edition 2, 1.4.2 or newer to run this example. For more details refer to the AWS Import/Export documentation.

Q. Do jobs expire?

Yes, if AWS has not received your storage device within 30 days of creating the job, it automatically expires. AWS will not accept packages for expired jobs. To resubmit, simply create a new job and update the SIGNATURE file, hard drive label, packing slip, and JOBID on the ship-to label.

Q. Can a job have more than one storage device?

No, each job is associated with exactly one manifest file and one customer device. If your data spans multiple devices, you must create multiple jobs.

Q. Can I enclose multiple jobs in the same package?

No, each job must be sent in a separate package.

Q. Can my job span multiple partitions?

No, each job has exactly one partition.

Q. I’m not in the US; can I use AWS Import/Export?

No, the beta only supports US-based buckets, and the ship-from and ship-to address must be located within the US.

Q. What happens if I change my AWS keys before a pending job completes?

The signature on your storage device will be invalid. As a result the job will not load your objects and your device will be returned. You’ll need to create a new job and generate a valid signature for your storage device.

Q. Can I cancel a job?

Yes, you can cancel a job anytime before we finish the data load. If AWS has not yet received your package, then we will deny delivery of the package. If AWS has already received your package you will be charged the per device fee. If the data load has started you will be charged for all data- loading-hours prior to processing the cancel job command. To cancel a job, simply send an e-mail to awsimportexport@amazon.com with the subject line “CANCEL JOB” and with the referenced JOBID as the first part of the e-mail body.

Q. Can I undo a job?

No, there is no option to restore your bucket to its previous state. If you want to delete the loaded data, you’ll need to write a program to traverse your import log and issue a DELETE operation on each object.

Q. Can you import my exported data?

Yes, but the exported files won’t map back to the same location in Amazon S3. If this is important to your use case, please contact AWS with the details at awsimportexport@amazon.com

↑ Top

Using Amazon Import

Q. How do I send my data to AWS?

To use import, start by loading your data onto your storage device, such as an external USB or eSATA hard drive. Then create a simple manifest file that includes your Amazon S3 bucket, AWS Access Key Id, and return shipping address. Next send an e-mail to awsimportexport@amazon.com with the subject line “CREATE JOB” and the manifest file attached.

After we validate and store your manifest file, a job is created. We return a JOBID in a “RE: CREATE JOB” success e-mail. An example JOBID is X452D. Following this, you generate a signature file named “SIGNATURE” in the root directory of your storage device by signing the manifest file and JOBID with your AWS Secret Access Key.

Next, label your media with the JOBID and fill out the information on the packing slip. Pack your media and ship it to AWS with the JOBID clearly labeled. For example:

JOBID X452D

WA 98144

After we load your data to Amazon S3, we send your device to the Return Shipping Address specified in your manifest file via standard ground shipping.

For more information about the manifest file format, SIGNATURE file, or the process to create the AWS Import/Export job, please review the AWS Import/Export Developer Guide.

Q. Can my data be loaded into more than one bucket?

No, data loads are restricted to one bucket per job.

Q. How are my files mapped to Amazon S3 objects?

The job’s manifest file specifies how to map files to key names and can be used to set the following object attributes: Content-Disposition, Content-Type and x-amz-acl.

Q. What happens if a key already exists?

Just like a standard Amazon S3 PUT, if a new object uses the same key as an existing object, the AWS Import/Export overwrites the old object with the new object.

Q. Can I write to someone else’s bucket? And who pays for it?

Yes, as long as your AWS account has write access to the bucket. The AWS account specified in the manifest file will pay all AWS Import/Export charges, while the bucket owner will pay the applicable Amazon S3 request and storage fees.

Q. Can I write to a requestor pays bucket?

No, requestor pays buckets are not supported at this time. You’ll need to load your data to a non-requestor pays bucket and then either make that bucket a requestor pays bucket or copy the data to the requestor pays bucket.

Q. Do I get a checksum or any kind of receipt on what was loaded into Amazon S3?

Yes, AWS Import/Export saves an import log report to your bucket. This report contains per file information including the date and time of the upload, Amazon S3 key, MD5 checksum, and number of bytes.

Q. I just loaded over a million objects into Amazon S3, how do I manage those objects?

Amazon S3 users create custom tools to manage millions of objects. We recommend reviewing the Best Practices for Using Amazon S3 article before creating your own tools.

Q. What happens when a file is larger than 5GB?

AWS Import/Export rejects all files larger than 5GB. We will provide the names of those files in the import log report.

Q. Do I need to backup my data before sending it to AWS?

Yes, you should have another copy of your data before shipping it to AWS. Your device and data could be damaged or lost in transit.

↑ Top

Using Amazon Export

Q. How do I get my data from AWS?

To use export, start by creating a manifest file that includes your Amazon S3 buckets, AWS Access Key Id, and return shipping address. If all of your data won’t fit on one device, or you’re unsure, you should issue a CREATE EXPORT PLAN command. This command takes a manifest file and allows you to specify information such as size about one or more storage devices. AWS will return a set of manifest files that splits your job up among those devices. Next send an e-mail to awsimportexport@amazon.com for each device you’re ready to send with the subject line “CREATE JOB” and the corresponding manifest file attached.

After AWS validates and stores your manifest file, a job is created. AWS returns a JOBID in a “RE: CREATE JOB” success e-mail. An example JOBID is X452D. Following this, you generate a signature file named “SIGNATURE” in the root directory of your storage device by signing the manifest file and JOBID with your AWS Secret Access Key.

Next, label your media with the JOBID and fill out the information on the packing slip. Pack your media and ship it to AWS with the JOBID clearly labeled. For example:

JOBID X452D

WA 98144

After AWS finishes transferring the data to your storage device, the device is sent to the Return Shipping Address specified in your manifest file via standard ground shipping. For more information about the manifest file format, SIGNATURE file, or the process to create the AWS Import/Export job, please review the AWS Import/Export Developer Guide.

Q. Can I retrieve data from more than one bucket?

Yes, you can specify more than one bucket in your manifest file. However you must have read permissions on each bucket and each object.

Q. How are my Amazon S3 objects mapped to files?

Each key is copied to your device in a directory tree that starts with the bucket’s name. For example, if the key was “images/chewie.jpg” and the bucket was “starwars”, then the object would be saved to /starwars/images/chewie.jpg. Meta data associated with each object is not copied to your storage device.

Q. What about my user metadata, ACLs, or headers?

At this time export does not include this data, however if this is important for your use case, please contact AWS with the details at awsimportexport@amazon.com.

Q. Do I get a checksum or any kind of receipt on what was copied to my storage device?

Yes, AWS saves an export log report to your specified log bucket. The report contains per file information which includes the date and time of the transfer, location on your storage device, MD5 checksum, and number of bytes.

Q. What happens if a key name doesn’t map to a valid filename?

Your file will be written to the recovery directory. Your log file will contain a redirect status code.

Q. Can I specify the name of the recovery directory?

Yes, you can specify the name of the recovery directory in your manifest file. The default name is “EXPORT-RECOVERY”.

Q. Can AWS format or erase my device before starting the export?

No, AWS only copies additional information to your device.

Q. What happens if a file already exists on my device?

The existing file will be overwritten with the new file. However before performing the export, AWS checks all keys for case sensitive name collisions on case insensitive file systems and saves duplicates to the recovery directory. For example, on FAT32 “gumby.txt” and “GUMBY.TXT” refer to the same file.

Q. What happens if my device is full?

An error will be flagged in your export log for each attempted key. Those objects will not be written to your storage device. You’ll need to download those files over the network or create a new job.

Hardware Information

Q. What are the storage devices requirements?

AWS Import/Export supports storage devices that can be mounted on Red Hat Linux. Your device must have:

USB 2.0 or eSATA connector
FAT32, ext2, ext3, or NTFS file system
Power supply that supports 120Volts at 60Hz with a standard US plug
Consumes less than 2,000 Watts
Weigh less than 50 pounds
Fit within 8 rack units of a standard 19-inch rack, or 14 inches high by 19 inches wide by 36 inches deep.
Device capacity is less than or equal to 2TB.

AWS can support larger and heavier devices upon request. If you have a device that does not meet these requirements, please contact us with the specifics. Make sure to include your AWS account(s) and the number of terabytes to load.

Q. Do I need to include my USB/eSATA cable , power cable and power supply?

Yes, your package must include your connector cables, power cables and power supply. If any of these are missing, we will return your package unprocessed. You will be charged the per device fee, but not for any data-loading-hours.

Q. How do I determine if my power supply supports 120Volts at 60Hz?

If your device works in a standard US wall socket you’re good to go. If you have concerns, the power supply typically will have a label that specifies voltage and frequency. For example, your power supply might specify, Input: 100-240V~ 50-60Hz. This label indicates that the power supply supports voltages from 100 to 240 volts at frequencies between 50 and 60Hz. If you have additional questions, please contact your device manufacturer.

↑ Top

Security

Q. How is my device authenticated?

The signature file on your storage device is created with your AWS Secret Access Key. This signature is used to securely identify and authenticate your device.

Q. Where does the data load occur?

During the beta your package is sent to one of our Seattle mailrooms. After your package is processed it is transported to one of our data centers. Once at the data center it’s attached to an import station for a data load over our internal high-speed network.

Q. Can I use a courier service?

Yes, for inbound shipment you can use a courier service. However at this time all outbound shipments are standard UPS ground.

Q. Can AWS wipe my device?

Yes, if you set the eraseDevice field to Yes in your manifest file, after completing the data load all writable blocks on your storage device will be overwritten with zeroes. You will need to repartition and format your device after the wipe. Data-loading-hours will be charged during this process.

Q. Who handles my package and device?

AWS uses 3rd party contractors to process and transport your package to the data center. Once your package is in the data center, it is handled by an Amazon employee. All 3rd party contractors and Amazon employees have gone through extensive background checks.

Q. Does AWS provide a chain of custody for my package?

Yes, we have a chain of custody in place for inbound and outbound packages between the mailroom and data center.

Q. What happens if my device is lost or damaged in transit?

If your device is lost in transit, you’ll need to follow up with the shipping carrier. If AWS Import/Export is unable to process your device, because it was damaged in transit, AWS will return your device to the Return Shipping address specified in your manifest file. You will need to contact your shipping carrier regarding the damage to your device.

Q. What happens if AWS loses or damages my device?

Although AWS has a number of internal controls and procedures to prevent loss, damage or disclosure of your data, AWS is not responsible for damages associated with loss or inadvertent disclosure of data, or the loss, damage or destruction of your storage device. You should always retain a back-up copy of your data.

Q. Do I need to encrypt my data?

There is no requirement to encrypt your data. However AWS generally recommends encrypting any sensitive data sent to AWS on your storage device or stored in Amazon S3. You should determine whether your use case warrants encryption.

Q. Can AWS Import/Export encrypt my exported data?

No, if you want your exported data encrypted, then you’ll need to save it to Amazon S3 in an encrypted format prior to export. If this is important to your use case, please contact AWS with the details at awsimportexport@amazon.com.

↑ Top

3rd Party Data Loads

Q. Can I load data on behalf of another AWS account or have data sent to me using AWS Import/Export?

Yes, provided that the AWS account specified in the manifest file is signed up for AWS Import/Export, and the signature on the device is produced using that account’s credentials.

Q. What is the best way to load data for another account?

If your agreement with the other account holder does not provide you access to their AWS credentials, you should have them produce the signature for the storage device using their keys. If your agreement provides you access to their AWS credentials, you can produce both the manifest and signature yourself.

AWS recommends loading data directly into the intended target bucket rather than using staging buckets.

↑ Top

Infrastructure Services

Virtual Private Cloud

Amazon Virtual Private Cloud

Payments & Billing

On-Demand Workforce

Amazon Mechanical Turk

Alexa Web Services

Merchant Services

Amazon Fulfillment Web Service (Amazon FWS)