Wait — Folders Aren’t Really Folders in Amazon S3 bucket?

Ajinkya Rathod
3 min readOct 29, 2024

--

Amazon S3 (Simple Storage Service) offers a scalable, durable storage solution designed for storing objects of any size, from a few bytes to terabytes of data. Think of it as a limitless vault, ready to securely hold all your data

amazon s3 cloud service — s3 bucket

Amazon S3 is a cloud storage service, while a S3 bucket is a container for storing objects in S3

Why is S3 an “Object” Storage?

Amazon S3 doesn’t use real folders — it’s a flat storage system where each object has a unique key, and the ‘folder’ structure is just an illusion created by slashes (/) in the key names!

Yes, you read that right-no folders! S3 stands out because of its flat storage model. This means S3 doesn’t use a traditional folder or block storage system.

  • No Hierarchical Structure: Unlike typical file systems (which have folders) or block storage (which can be mounted), S3’s storage is flat
  • Every Object Has a Unique Key: In S3, each object is assigned a unique key, often appearing like a file path, such as /folder_name/image.jpeg. While it may look like folder_name is a real directory, it's just a visual aid. S3 doesn't use actual folders or directories.
  • Using Delimiters for Visual Structure: The forward slash (/) acts as a delimiter to create the appearance of folders. For example, folder_name/image.jpeg looks like an image inside a folder, but it's stored as a single, unique key.

Important Facts About S3

S3 brings several unique advantages, from global naming requirements to flexible replication options.

Bucket Naming Rules

  • Globally Unique: Each bucket name in S3 must be globally unique, so once a name is claimed, it’s unavailable across all of AWS.
  • Character Requirements: Bucket names must be between 3–63 characters, all lowercase, with no underscores. Names must begin with a lowercase letter or number.

Region-Specific Storage with Cross-Region Replication

  • Single Region by Default: By default, each S3 bucket exists in a specific AWS region.
  • Cross-Region Replication: For added resilience, S3 allows you to set up cross-region replication.
  • Example: If you create a bucket in us-west-1 (United States), you can configure it to replicate to eu-central-1 (Germany). This setup allows automatic replication of objects to a backup location, protecting against regional failures.

S3 Storage Limits: Soft and Hard

  • Soft Limit (Adjustable): Each account has a default limit of 100 buckets. This limit is adjustable upon request to AWS, so if you need more, you can increase it.
  • Hard Limit (Fixed): The maximum number of buckets per AWS account is 1,000. This limit cannot be exceeded, even with a request.

--

--

No responses yet