Cluster Namespace and Object Naming

In YDB, all schema objects are organized into a hierarchical namespace called the cluster schema. This structure determines how databases, directories, and other objects are arranged and named, providing a clear and scalable way to manage data.

Cluster Schema

The cluster schema forms a tree-like hierarchy where:

• The root represents the top level of the namespace.
• At the first level are databases.
• Inside each database, you can create nested directories to build a custom hierarchy.
• Schema objects—such as tables, views, and indexes—can reside either at the database root or within any directory.

Naming Rules for Schema Objects

Every schema object in YDB has a name. In YQL expressions, object names are used as identifiers, which can be unquoted or enclosed in backticks (`). For details on identifier syntax, see Keywords and identifiers.

Valid names for schema objects must meet the following criteria:

• Allowed characters:
  • Latin letters (A–Z, a–z)
  • Digits (0–9)
  • Special characters: ., -, and _
• Maximum length: 255 characters
• Placement restrictions: Objects cannot be created in directories with names that start with a dot (for example, .sys, .metadata, .sys_health)