S3

S3, short for Simple Storage Service, is a scalable object storage service offered by Amazon Web Services. It provides a web-based interface to store and retrieve any amount of data from anywhere on the Internet. Data is stored as objects within named buckets, with each object consisting of a data payload, metadata, and a unique key that identifies it.

S3 is designed for high durability and availability. By default, data is redundantly stored across multiple facilities within an AWS region, helping protect against loss due to facility failures. S3 objects can be configured for cross-region replication to enable data copies in another region.

Storage options include several classes tailored to different access patterns: S3 Standard for general-purpose use; S3 Standard-IA and S3 One Zone-IA for infrequently accessed data; S3 Intelligent-Tiering, which automatically moves objects between tiers based on usage; and archival classes such as S3 Glacier and S3 Glacier Deep Archive for long-term retention with varying retrieval times. S3 Object Lock provides retention settings for compliance, and encryption can be applied at rest (SSE-S3, SSE-KMS, or SSE-C) or via client-side encryption. Access control is managed through IAM, bucket policies, and ACLs, and data transfer can be optimized with features like presigned URLs, multipart uploads, and Transfer Acceleration. Event notifications can trigger workflows through services such as AWS Lambda, SNS, or SQS.

Pricing follows a pay-as-you-go model, with costs determined by storage class, region, data transfer, and API requests. The service underpins many architectures, including backups, data lakes, content delivery, and static website hosting, and has been a core component of AWS since its 2006 introduction.