Replacing my home backup server with Amazon's S3
Not too long ago, Amazon released their Simple Storage Service (or "S3" for short). It provides a hosted storage platform which developers can build all sorts of applications on top of. Smugmug, a popular photo sharing web site, is using it to store and host pictures.
Context & Ripple Effects
Amazon introduced S3 as a grid-storage web service in March 2006, with availability documented the following day in Amazon’s S3 launch coverage. This account turns that developer-facing service into a personal infrastructure choice.
SmugMug’s use of S3 to store and serve photos shows the same hosted-storage model operating for a photo-sharing service as well as for an individual’s backup needs.
First-order effects
- The server owner shifts backup-storage responsibility from a home machine to Amazon’s S3 service, eliminating the need to keep that local system running for that purpose.
- SmugMug and similar web applications can treat S3 as hosted storage and image delivery infrastructure rather than building equivalent storage capacity themselves.
Second-order effects
- Amazon’s S3 competes with self-managed servers for storage workloads, making operational convenience part of the choice alongside raw storage capacity.
- Developers gain a common hosted-storage layer on which to build applications, reducing the amount of storage infrastructure each service must operate independently.
Third-order effects
- If personal and application storage both move onto services such as S3, storage becomes a network-delivered utility and ownership of the underlying machines concentrates with the service operator.
The trend: S3 is an early example of storage shifting from customer-operated hardware to programmable hosted infrastructure.