Salvato in:
Dettagli Bibliografici
Autori principali: Adams, Philip, Li, Menghao, Zhang, Shi, Tan, Li, Chen, Qi, Li, Mingqin, Li, Zengzhong, Risvik, Knut, Simhadri, Harsha Vardhan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2509.06046
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916938871996416
author Adams, Philip
Li, Menghao
Zhang, Shi
Tan, Li
Chen, Qi
Li, Mingqin
Li, Zengzhong
Risvik, Knut
Simhadri, Harsha Vardhan
author_facet Adams, Philip
Li, Menghao
Zhang, Shi
Tan, Li
Chen, Qi
Li, Mingqin
Li, Zengzhong
Risvik, Knut
Simhadri, Harsha Vardhan
contents We present DISTRIBUTEDANN, a distributed vector search service that makes it possible to search over a single 50 billion vector graph index spread across over a thousand machines that offers 26ms median query latency and processes over 100,000 queries per second. This is 6x more efficient than existing partitioning and routing strategies that route the vector query to a subset of partitions in a scale out vector search system. DISTRIBUTEDANN is built using two well-understood components: a distributed key-value store and an in-memory ANN index. DISTRIBUTEDANN has replaced conventional scale-out architectures for serving the Bing search engine, and we share our experience from making this transition.
format Preprint
id arxiv_https___arxiv_org_abs_2509_06046
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DISTRIBUTEDANN: Efficient Scaling of a Single DISKANN Graph Across Thousands of Computers
Adams, Philip
Li, Menghao
Zhang, Shi
Tan, Li
Chen, Qi
Li, Mingqin
Li, Zengzhong
Risvik, Knut
Simhadri, Harsha Vardhan
Distributed, Parallel, and Cluster Computing
Data Structures and Algorithms
Information Retrieval
E.1; H.3.3
We present DISTRIBUTEDANN, a distributed vector search service that makes it possible to search over a single 50 billion vector graph index spread across over a thousand machines that offers 26ms median query latency and processes over 100,000 queries per second. This is 6x more efficient than existing partitioning and routing strategies that route the vector query to a subset of partitions in a scale out vector search system. DISTRIBUTEDANN is built using two well-understood components: a distributed key-value store and an in-memory ANN index. DISTRIBUTEDANN has replaced conventional scale-out architectures for serving the Bing search engine, and we share our experience from making this transition.
title DISTRIBUTEDANN: Efficient Scaling of a Single DISKANN Graph Across Thousands of Computers
topic Distributed, Parallel, and Cluster Computing
Data Structures and Algorithms
Information Retrieval
E.1; H.3.3
url https://arxiv.org/abs/2509.06046