OpenAI describes scaling Habitat storage on Azure Cosmos DB
Original title: Rapidly scaling online storage to serve over 1 billion ChatGPT users (Part II)
Part II covers multi-tenant Cosmos DB containers after Habitat grew past 500 petabytes for ChatGPT.
OpenAI's engineering post explains how Habitat, its online storage platform, serves more than 1 billion users across 40 regions. Part II focuses on scaling Azure Cosmos DB through co-managed multi-tenant containers after earlier single-tenant and single-container designs. OpenAI describes noisy-neighbor effects, shared logical partition keys and physical-partition throughput limits.
- Habitat must answer in milliseconds for product traffic, OpenAI says.
- The series says Habitat now holds over 500 petabytes.
- Clients could choose keys such as user_id or workspace_id, causing collisions.
- OpenAI says it returned to a multi-tenant container by default after trying single-tenant backing containers.
- Products
- HabitatAzure Cosmos DB
- AI models
- ChatGPT
From the original article
13 min read at openai.com
Learn how OpenAI scaled Habitat with reliable multi-tenancy, global caching, and Azure Cosmos DB to deliver 12× faster reads for ChatGPT.
openai.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.