AWS Snowball Edge

Primary Use Cases

• Large-Scale Data Migration: For offline transfer of massive datasets when online transfer over the internet is too slow, unreliable, or expensive. Ideal for migrating data from data centers, shutting down legacy systems, or large-scale analytics data transfer.

• Edge Computing & Storage: Run compute jobs and store data in harsh, remote, or disconnected environments like factory floors, ships, mines, or transportation vehicles.

• Disaster Recovery: Provides a fast way to move large amounts of data to AWS for DR purposes or to quickly restore critical data from S3 to your on-premises environment.

• Content Distribution: Transport large media libraries, scientific data, or other content to partners and customers.

Device Types

AWS Snowball Edge comes in two primary configurations to suit different needs.

| Feature | Snowball Edge Storage Optimized | Snowball Edge Compute Optimized |

| ----------------------- | ---------------------------------------------------------------- | ------------------------------------------------------------------ |

| Primary Function | Large-scale data transfer with some compute capabilities. | Compute-intensive workloads at the edge. |

| Usable HDD Storage | ~80 TB | ~28 TB |

| Usable NVMe SSD Storage| ~1 TB (for EC2 instances) | ~39.5 TB (for block storage and S3-compatible object storage) |

| vCPUs | 40 | 104 |

| Memory | 80 GB | 416 GB |

| GPU Available? | No | Yes (optional, Tesla V100 GPU) |

| Ideal For | Bulk data migration, large-scale data analytics transfer. | Machine learning, full-motion video analysis, IoT data processing. |

Key Features

• On-board Compute: Capable of running EC2 instances and AWS Lambda functions locally, allowing you to pre-process data or run applications at the edge.

• On-board Storage Interfaces:

  • Amazon S3 Compatible Endpoint: Provides an S3-compatible object storage interface, allowing you to use existing AWS SDKs and CLI commands to transfer data.

  • NFS File Interface: Can be configured as a Network File System (NFS) mount point for easy integration with existing file-based applications.

• Clustering: You can cluster 5 to 10 Snowball Edge devices together to create a single, durable, S3-compatible storage pool with increased storage capacity and local redundancy.

• High-Speed Networking: Devices include network ports up to 100 Gbps for fast data transfer on your local network.

Security Features

• Rugged, Tamper-Evident Enclosure: The device is physically hardened and designed to be tamper-resistant.

• End-to-End Encryption: All data written to a Snowball Edge is automatically encrypted with 256-bit encryption keys that you manage using the AWS Key Management Service (KMS).

• Trusted Platform Module (TPM): Provides a hardware root of trust. The TPM chip detects any unauthorized changes to the device, preventing it from being accessed if tampering is detected.

The Snowball Edge Workflow

1. Create a Job in the AWS Console: Plan your job by selecting the device type (Storage or Compute Optimized), the S3 buckets for import/export, and any pre-loaded AMIs for EC2 instances or Lambda functions. AWS prepares and ships the device to you.

2. Receive and Connect: Once the device arrives, connect it to your local network and power it on. The E Ink display on the front provides status information.

3. Unlock the Device: From the AWS Console, get your job manifest file and the 25-character unlock code. Use these credentials with the Snowball Edge client or AWS OpsHub to authenticate and unlock the device.

4. Transfer Data: Use the S3 Adapter/CLI or mount the NFS file share to copy data to the device.

5. Run Compute Jobs: If configured, start your EC2 instances or trigger your Lambda functions to process data locally.

6. Power Off and Return: Once data transfer and compute jobs are complete, power off the device. The E Ink shipping label on the device automatically updates with the correct AWS return address.

7. Data Ingestion: AWS receives the device, inspects it for tampering, and automatically loads your data into the designated S3 bucket. You are notified of the job's completion.

Management Tools

• AWS OpsHub for Snow Family: A graphical user interface (GUI) application that you install on a local computer. It simplifies the process of setting up and managing Snowball Edge devices, allowing you to unlock them, configure services, monitor status, and launch compute instances without using the command line.