September 12th, 2024

DiskANN for Azure Cosmos DB Now in Open Public Preview!

James Codella
Principal Product Manager

We’re excited to announce that the DiskANN vector index is now in open public preview in Azure Cosmos DB for NoSQL! This means that anyone who enrolls an Azure Cosmos DB for NoSQL resource in the Vector Search Public Preview now automatically gets access to the most powerful and scalable vector index on the planet.

What is DiskANN?

DiskANN is a powerful set of algorithms developed at Microsoft Research for low-latency, cost-effective, and highly accurate vector search at scale. It’s an ideal solution for applications requiring fast and efficient search capabilities without compromising on accuracy.

Image DiskANNArch

  • DiskANN is a graph-based indexing and search system that performs fast and accurate approximate nearest neighbor (ANN) search at any-scale.
  • It primarily uses an SSD-based index to scale to an order of magnitude more points compared to in-memory indices, while still retaining high QPS and low latency.
  • Quantized (compressed) vectors are kept in memory, and DiskANN balances interactions between the two to offer low latency and high accuracy.
  • DiskANN is based on a novel graph index called Vamana that is more versatile than existing graph indices by maintaining accuracy despite many insertions, modifications, and deletions, without the need for expensive index rebuilds.

Why use Azure Cosmos DB with DiskANN?

Azure Cosmos DB offers essential capabilities that, combined with DiskANN, make it the most powerful database for cutting-edge AI applications.

  1. Mission-Critical Vector Store: Say goodbye to niche solutions! Our vector store delivers industry-leading similarity search performance, ensuring your critical workloads thrive.
  2. No-ETL Simplicity: Keep data and vectors in harmony. By sidestepping ETL (Extract, Transform, Load) complexities, you’ll streamline your AI application architecture keeping your data in one solution.
  3. Flexible Query Filters: Seamlessly blend vector search with custom query filters to enhance the precision and relevancy of the vector search results.
  4. Quick Start Options: Dive in swiftly and economically with Azure Cosmos DB’s serverless mode. Alternatively, harness dynamic, instant Autoscale capabilities in provisioned throughput mode.
  5. Production-Ready Excellence: With built-in multitenancy (offering five levels), global replication, and industry-leading SLAs, we guarantee up to 99.999% availability.

Enroll in the Vector Search Preview

Developing AI applications with Azure Cosmos DB has never been easier. Here’s how to get started:

Step 1: Setup an Azure Cosmos DB for NoSQL resource in your Azure subscription.

Step 2: In the Azure Portal, navigate to your Azure Cosmos DB for NoSQL resource, select the “Features” tab, and then enroll in the “Vector Search for NoSQL API (preview)”.

Image VSpreview

And that’s it! You’re now setup to perform highly scalable and cost-effective vector search using DiskANN!

Next Steps

About Azure Cosmos DB

Azure Cosmos DB is a fully managed and serverless distributed database for modern app development, with SLA-backed speed and availability, automatic and instant scalability, and support for open-source PostgreSQL, MongoDB, and Apache Cassandra. Try Azure Cosmos DB for free here. To stay in the loop on Azure Cosmos DB updates, follow us on XYouTube, and LinkedIn.

Author

James Codella
Principal Product Manager

0 comments