@TobiasTernent

GitHub: Tobias-Ternent

Thursday, 23 June 2016

AWS storage

AWS offers 3 main ways for data storage:

S3 
Simple Storage Service (S3) - data is held in the cloud, in a "bucket". Inside a bucket, files and sub-directories can be created. The point of this type of storage is to write once, read may times perhaps from different services (EC2 instance, the Internet, etc). They can also be deleted easily. Data are automatically backed up and have a 99.9% uptime.

EBS 
Elastic Block Storage (EBS) - these are volumes that can be added onto EC2 servers, think of them like hard drives. They can be scaled up/down, added, removed, backed-up as needed per instance. 

Glacier 
The Glacier is for remote "off-site" backups. Write once, read (almost) never. This should be for long-term archival of data, e.g. that was present in S3 for a long time, and will never normally be used again, save for restoration of backup data. Glacier prices are cheaper than S3 for storage, but have retrieval costs.

No comments:

Post a Comment