We’ve integrated DiskANN, a state-of-the-art vector indexing algorithm, into Azure Cosmos DB NoSQL, a state-of-the-art cloud-native operational database. Learn how we overcame the systems and algorithmic challenges of this integration to achieve <20ms query latency at the 10 million scale, while supporting scale-out to billions of vectors via automatic partitioning.
Magdalen Manohar is a Senior Researcher in Microsoft Azure. Her researcher interests include all things vector databases, especially the design of efficient parallel and concurrent vector indexing algorithms. She is develops the open source ParlayANN library, a benchmark suite of high-performance vector indexing algorithms. She holds a PhD in Computer Science from Carnegie Mellon University.
