Quick Definition: According to Amazon Web Service, AWS Snowball is a petabyte-scale data transport service that uses secure devices to transfer large amounts of data into and out of the AWS cloud. Snowball addresses challenges like high network costs, long transfer times, and security concerns to migrate data as efficiently as possible.
As many organizations are defining their cloud adoption strategies, data migration is a key consideration that requires special attention. This is because on-premise data transfer/migration becomes very easy when the right strategies and processes are in place. Failure to take this all-important readiness assessment step may result in some technical risks or challenges as the project progresses.
If AWS Snowball forms part of your data migration strategies, do not forget to follow strictly the migration guidelines as specified by AWS User Guide. This practice will greatly help your delivery team to complete different data migration tasks by following the project plan. It also helps checkmate against project risks that can impact your delivery timelines. The migration should be carried out in an orderly manner with proper documentation at every stage.
Key Data Transfer Considerations to Make Before Moving Data to Cloud
More often than not, organizations that want to move their workloads to the cloud especially AWS, overlook certain tasks that ought to be considered in the readiness assessment stage. In this article, we shall have a look at those tasks that are mostly overlooked in the planning phase. They include:
1. Confirm your on-premises system specifications. That is to say, make sure that your on-premise workstation meets the minimum requirement for the amount of data transfer you want to migrate. A 16-core processor, 16 MB of RAM, an RJ45/SPF+ network connection or higher will be ideal.
2. Determine the nature of data or the data type to be migrated to the cloud.
Make sure that AWS Snowball can handle such data. A cloud migration model can help provide you with certain specific details that you may need to understand the data you want to move. Doing so helps to eliminate risks during the data transfer proper.
3. Answer the question, what amount of data to be migrated? This is because it is more cost-effective to use AWS Snowball to transfer a large amount of data (tens of terabytes) instead of a few Gigabytes. It’s an important consideration to make so that your company can enjoy the economy of scale that AWS provides.
Other AWS Snowball Data Transfer Considerations:-
4. Download and install Snowball client on the workstation your organization wants to use for the migration. After a successful installation of the agent, perform a data transfer test on the workstation. This is necessary because it will help you have a better understanding of the data transfer methods and time requirements. The time requirements include the estimated efforts in terms of the amount of time required to complete the data transfer and the amount of time the Snowball device might be in the organization’s location. These efforts estimations are necessary because of cost implications. Organizations can incur additional costs 10 days after taking delivery of the Snowball device. So, do not order a Snowball device during the holiday season.
5. Dedicated Administrative IAM user account for the data transfer activity. Using AWS management console, create a dedicated administrative IAM user account that you will use to order, track, and manage Snowball Import/Export jobs as well as return the Snowball device to AWS.
6. In naming your S3 bucket, do follow strictly the “Object Key Naming convention”. This is key.
7. Track each stage of the Snowball Import/Export job. This you can do by the use of Amazon SNS. This is a good way to constantly update the entire team about the shipping status. It also provides an up to date information about the data import into Amazon S3. It is also important to specify the shipping address for the Snowball device as well as key account owners and their contacts.
8. Manage your change request processes properly to avoid unnecessary delays and associated costs. Set a clear timeline for the receipt of the device, loading of data and shipment back to AWS to avoid extra payments.
9. Make sure that you do not delete your company source data until the data transfer has been moved to Amazon S3 buckets successfully.
Types of AWS Snowball
Let’s have a look at different types of AWS Snowball available today. This is to help us in arriving at the right decision as to whether we really need a particular snow family type or not. They are as follows:
Snowball – for Petabyte data migration to Amazon S3 bucket.
Key Features are:
- 50 – 80 TB storage capacity and globally available.
- 10 GB Link (Copper & Fibre)
- Requires an S3 agent client
- Rain and dust resistant
- Rugged 8,5 G Impact case
Snowball Edge – for compute and storage for hybrid/edge workloads. The data is transferred to the Amazon S3 bucket.
Key Features are:
- Two (2) separate IP addresses (S3-compatible endpoint & file interface (NFS)).
- 100 TB local storage
- Faster data transfer (10GBase-T, 10/25GB SFP28, and 40GB QSFP+ Copper, and Optical networking
- Ruggedized and rack-mountable.
- Can run Lambda Function and includes local compute equivqlent to an Amazon EC2 m4.4xLarge instance – 16vCPU/32 GB RAM
Snowmobile – for Exabyte data migration.
Having a great project plan that takes into account all the key considerations highlighted above will help your project team deliver data migration to AWS via Snowball without technical risks and unnecessary challenges.