Build

Namespaces

Partition data within a project using logical namespaces for multi-tenant or multi-environment setups.

Overview

Namespaces partition data inside a single project. Use them to separate datasets, tenants, or environments without creating additional projects. Each namespace is an isolated partition — the same document ID can exist in multiple namespaces without colliding.

Usage

namespaces.py

import os
from kyrodb import KyroDBClient

with KyroDBClient(
    target=os.environ["KYRODB_GRPC_ENDPOINT"],
    api_key=os.environ["KYRODB_API_KEY"],
) as client:
    # Insert into the "products" namespace
    client.insert(
        doc_id=1,
        embedding=[0.1, 0.2, 0.3, 0.4],
        metadata={"name": "Widget A"},
        namespace="products",
    )

    # Insert the same ID into a different namespace
    client.insert(
        doc_id=1,
        embedding=[0.4, 0.5, 0.6, 0.7],
        metadata={"ticket": "T-1234"},
        namespace="support",
    )

    # Search only within "products"
    results = client.search(
        query_embedding=[0.1, 0.2, 0.3, 0.4],
        k=5,
        namespace="products",
    )

Default Namespace

If you omit the namespace parameter, all operations use the default namespace automatically.

Common Patterns

Use Case	Namespace Strategy
Multi-tenant SaaS	One namespace per tenant: `tenant-123`, `tenant-456`
Environment separation	`production`, `staging`, `development`
Data categories	`products`, `support-tickets`, `documentation`
A/B testing	`index-v1`, `index-v2`

Namespace vs. Project

Choose the right isolation level for your use case:

Need	Use
Logical data separation within one app	Namespace
Different embedding dimensions	Separate project
Different distance metrics	Separate project
Different regions	Separate project
Separate API key boundary	Separate project

Next Steps

Python SDK for the full API reference including namespace parameters on every operation.
Projects for when to create new projects vs. namespaces.