Skip to content

AWS Snow

Links: 101 AWS SAA Index


  • Snow devices serve two main purpose:

    • Migrate data into and out of AWS.
      • attachments/Pasted image 20220424202243.jpg
    • Collect and process data at edge.
  • AWS Snow family are offline devices to perform data migrations.

Why data migrations with AWS Snow

Sometimes we want to get our data to AWS fast but there are several challenges like - Limited Connectivity - Limited Bandwidth - Higher network cost - Connection stability

If it takes over a week to transfer data to AWS over the network then use AWS Snowball devices.
  • AWS will send you a device via post office, load the data into it and then send it back to AWS.

Data Transfers


  • It is a small portable rugged computing which can withstand harsh environments.
  • The amount of usable data that can be stored in it is 8TBs.
  • It should be used where Snowball Edge cannot be used due to space constraints.
  • We can send it back to AWS offline or connect it to the internet to send data via AWS DataSync.
Snowcone is the only offering in which data can be transferred online using DataSync agent. In both Snowball Edge and Snowmobile data is transferred offline.

Snowball Edge - Storage Optimised

  • Storage optimised version has 80 TB of space.
  • Used for transferring TBs and PBs (few PBs like 5). Although capacity is 80 TBs you can order multiple.
  • 40 vCPUs and 80 GiB of RAM.
  • Provides block storage and S3 compatible object storage.
  • Use Cases: large cloud migrations, disaster recovery
Snowball cannot import to Glacier directly. You must use Amazon S3 first, in combination with an S3 lifecycle policy.
Both snowball edge compute and storage optimised allow for storage clustering.


  • It is used for transferring EBs of data
  • It is literally a truck.
  • Each truck can transfer upto 100PBs of data.
  • The truck is highly secure, temperature controlled, GPS etc.
  • If you are transferring more the 10 PBs then you should consider AWS Snowmobile.

Edge Computing

  • Process data while it is being created at an Edge Location. Edge location is anything that is offline and doesn’t have access to the internet.
  • Use Cases: Preprocess data, Machine learning at Edge, Transcoding media streams
  • Long term discounted price for 1 and 3 years.
  • Can run EC2 Instances & AWS Lambda functions (using AWS loT Greengrass)


  • 2 CPUs, 4 GB of memory, wired or wireless access
  • USB-C power using a cord or the optional battery

Snowball Edge - Compute Optimised

  • 52 vCPUs, 208 GiB of RAM
  • Optional GPU (useful for video processing or machine learning)
  • 42 TB usable storage

AWS OpsHub

  • Historically to deal with the snow family of devices one had to use CLI.
  • AWS OpsHub is something you can download on your desktop to manage the snow devices.

Last updated: 2022-05-06