Cost-Effective, Low Latency Vector Search with Azure Cosmos DB

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Upreti, Nitish, Simhadri, Harsha Vardhan, Sundar, Hari Sudan, Sundaram, Krishnan, Boshra, Samer, Perumalswamy, Balachandar, Atri, Shivam, Chisholm, Martin, Singh, Revti Raman, Yang, Greg, Hass, Tamara, Dudhey, Nitesh, Pattipaka, Subramanyam, Hildebrand, Mark, Manohar, Magdalen, Moffitt, Jack, Xu, Haiyang, Datha, Naren, Gupta, Suryansh, Krishnaswamy, Ravishankar, Gupta, Prashant, Sahu, Abhishek, Varada, Hemeswari, Barthwal, Sudhanshu, Mor, Ritika, Codella, James, Cooper, Shaun, Pilch, Kevin, Moreno, Simon, Kataria, Aayush, Kulkarni, Santosh, Deshpande, Neil, Sagare, Amar, Billa, Dinesh, Fu, Zishan, Vishal, Vipul
Format:	Preprint
Veröffentlicht:	2025
Schlagworte:	Databases Information Retrieval H.3.3
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

_version_	1866913969119166464
author	Upreti, Nitish Simhadri, Harsha Vardhan Sundar, Hari Sudan Sundaram, Krishnan Boshra, Samer Perumalswamy, Balachandar Atri, Shivam Chisholm, Martin Singh, Revti Raman Yang, Greg Hass, Tamara Dudhey, Nitesh Pattipaka, Subramanyam Hildebrand, Mark Manohar, Magdalen Moffitt, Jack Xu, Haiyang Datha, Naren Gupta, Suryansh Krishnaswamy, Ravishankar Gupta, Prashant Sahu, Abhishek Varada, Hemeswari Barthwal, Sudhanshu Mor, Ritika Codella, James Cooper, Shaun Pilch, Kevin Moreno, Simon Kataria, Aayush Kulkarni, Santosh Deshpande, Neil Sagare, Amar Billa, Dinesh Fu, Zishan Vishal, Vipul
author_facet	Upreti, Nitish Simhadri, Harsha Vardhan Sundar, Hari Sudan Sundaram, Krishnan Boshra, Samer Perumalswamy, Balachandar Atri, Shivam Chisholm, Martin Singh, Revti Raman Yang, Greg Hass, Tamara Dudhey, Nitesh Pattipaka, Subramanyam Hildebrand, Mark Manohar, Magdalen Moffitt, Jack Xu, Haiyang Datha, Naren Gupta, Suryansh Krishnaswamy, Ravishankar Gupta, Prashant Sahu, Abhishek Varada, Hemeswari Barthwal, Sudhanshu Mor, Ritika Codella, James Cooper, Shaun Pilch, Kevin Moreno, Simon Kataria, Aayush Kulkarni, Santosh Deshpande, Neil Sagare, Amar Billa, Dinesh Fu, Zishan Vishal, Vipul
contents	Vector indexing enables semantic search over diverse corpora and has become an important interface to databases for both users and AI agents. Efficient vector search requires deep optimizations in database systems. This has motivated a new class of specialized vector databases that optimize for vector search quality and cost. Instead, we argue that a scalable, high-performance, and cost-efficient vector search system can be built inside a cloud-native operational database like Azure Cosmos DB while leveraging the benefits of a distributed database such as high availability, durability, and scale. We do this by deeply integrating DiskANN, a state-of-the-art vector indexing library, inside Azure Cosmos DB NoSQL. This system uses a single vector index per partition stored in existing index trees, and kept in sync with underlying data. It supports < 20ms query latency over an index spanning 10 million vectors, has stable recall over updates, and offers approximately 43x and 12x lower query cost compared to Pinecone and Zilliz serverless enterprise products. It also scales out to billions of vectors via automatic partitioning. This convergent design presents a point in favor of integrating vector indices into operational databases in the context of recent debates on specialized vector databases, and offers a template for vector indexing in other databases.
format	Preprint
id	arxiv_https___arxiv_org_abs_2505_05885
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	Cost-Effective, Low Latency Vector Search with Azure Cosmos DB Upreti, Nitish Simhadri, Harsha Vardhan Sundar, Hari Sudan Sundaram, Krishnan Boshra, Samer Perumalswamy, Balachandar Atri, Shivam Chisholm, Martin Singh, Revti Raman Yang, Greg Hass, Tamara Dudhey, Nitesh Pattipaka, Subramanyam Hildebrand, Mark Manohar, Magdalen Moffitt, Jack Xu, Haiyang Datha, Naren Gupta, Suryansh Krishnaswamy, Ravishankar Gupta, Prashant Sahu, Abhishek Varada, Hemeswari Barthwal, Sudhanshu Mor, Ritika Codella, James Cooper, Shaun Pilch, Kevin Moreno, Simon Kataria, Aayush Kulkarni, Santosh Deshpande, Neil Sagare, Amar Billa, Dinesh Fu, Zishan Vishal, Vipul Databases Information Retrieval H.3.3 Vector indexing enables semantic search over diverse corpora and has become an important interface to databases for both users and AI agents. Efficient vector search requires deep optimizations in database systems. This has motivated a new class of specialized vector databases that optimize for vector search quality and cost. Instead, we argue that a scalable, high-performance, and cost-efficient vector search system can be built inside a cloud-native operational database like Azure Cosmos DB while leveraging the benefits of a distributed database such as high availability, durability, and scale. We do this by deeply integrating DiskANN, a state-of-the-art vector indexing library, inside Azure Cosmos DB NoSQL. This system uses a single vector index per partition stored in existing index trees, and kept in sync with underlying data. It supports < 20ms query latency over an index spanning 10 million vectors, has stable recall over updates, and offers approximately 43x and 12x lower query cost compared to Pinecone and Zilliz serverless enterprise products. It also scales out to billions of vectors via automatic partitioning. This convergent design presents a point in favor of integrating vector indices into operational databases in the context of recent debates on specialized vector databases, and offers a template for vector indexing in other databases.
title	Cost-Effective, Low Latency Vector Search with Azure Cosmos DB
topic	Databases Information Retrieval H.3.3
url	https://arxiv.org/abs/2505.05885