FreshDiskANN: A Fast and Accurate Graph-Based ANN Index for Streaming Similarity Search

Approximate nearest neighbor search (ANNS) is a fundamental building block in information retrieval with graph-based indices being the current state-of-the-art and widely used in the industry. Recent advances in graph-based indices have made it possible to index and search billion-point datasets wit...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Singh, Aditi, Subramanya, Suhas Jayaram, Krishnaswamy, Ravishankar, Simhadri, Harsha Vardhan
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Singh, Aditi
Subramanya, Suhas Jayaram
Krishnaswamy, Ravishankar
Simhadri, Harsha Vardhan
description Approximate nearest neighbor search (ANNS) is a fundamental building block in information retrieval with graph-based indices being the current state-of-the-art and widely used in the industry. Recent advances in graph-based indices have made it possible to index and search billion-point datasets with high recall and millisecond-level latency on a single commodity machine with an SSD. However, existing graph algorithms for ANNS support only static indices that cannot reflect real-time changes to the corpus required by many key real-world scenarios (e.g. index of sentences in documents, email, or a news index). To overcome this drawback, the current industry practice for manifesting updates into such indices is to periodically re-build these indices, which can be prohibitively expensive. In this paper, we present the first graph-based ANNS index that reflects corpus updates into the index in real-time without compromising on search performance. Using update rules for this index, we design FreshDiskANN, a system that can index over a billion points on a workstation with an SSD and limited memory, and support thousands of concurrent real-time inserts, deletes and searches per second each, while retaining $>95\%$ 5-recall@5. This represents a 5-10x reduction in the cost of maintaining freshness in indices when compared to existing methods.
doi_str_mv 10.48550/arxiv.2105.09613
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2105_09613</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2105_09613</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-feb68bd49037a7adb9be822e908e071d47f32b2dab33f2663e1e4fd48ac169563</originalsourceid><addsrcrecordid>eNotz71OwzAYhWEvDKhwAUz4BhL8FyfpFgoplaowpBJj9Dn-3Fg0obIDau8eKExHeocjPYTccZaqIsvYA4ST_0oFZ1nKSs3lNXmrA8bhycf3qmmWtKI1xJnCZGnV958BZqTrAMcheYSIP7Fp6GayeKLuI9B2Dgijn_a09aM_QPDzmbYIoR9uyJWDQ8Tb_12QXf28W70k29f1ZlVtE9C5TBwaXRirSiZzyMGa0mAhBJasQJZzq3InhREWjJROaC2Ro3JWFdBzXWZaLsj93-1F1h2DHyGcu19hdxHKb170SqI</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>FreshDiskANN: A Fast and Accurate Graph-Based ANN Index for Streaming Similarity Search</title><source>arXiv.org</source><creator>Singh, Aditi ; Subramanya, Suhas Jayaram ; Krishnaswamy, Ravishankar ; Simhadri, Harsha Vardhan</creator><creatorcontrib>Singh, Aditi ; Subramanya, Suhas Jayaram ; Krishnaswamy, Ravishankar ; Simhadri, Harsha Vardhan</creatorcontrib><description>Approximate nearest neighbor search (ANNS) is a fundamental building block in information retrieval with graph-based indices being the current state-of-the-art and widely used in the industry. Recent advances in graph-based indices have made it possible to index and search billion-point datasets with high recall and millisecond-level latency on a single commodity machine with an SSD. However, existing graph algorithms for ANNS support only static indices that cannot reflect real-time changes to the corpus required by many key real-world scenarios (e.g. index of sentences in documents, email, or a news index). To overcome this drawback, the current industry practice for manifesting updates into such indices is to periodically re-build these indices, which can be prohibitively expensive. In this paper, we present the first graph-based ANNS index that reflects corpus updates into the index in real-time without compromising on search performance. Using update rules for this index, we design FreshDiskANN, a system that can index over a billion points on a workstation with an SSD and limited memory, and support thousands of concurrent real-time inserts, deletes and searches per second each, while retaining $&gt;95\%$ 5-recall@5. This represents a 5-10x reduction in the cost of maintaining freshness in indices when compared to existing methods.</description><identifier>DOI: 10.48550/arxiv.2105.09613</identifier><language>eng</language><subject>Computer Science - Information Retrieval</subject><creationdate>2021-05</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,777,882</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2105.09613$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2105.09613$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Singh, Aditi</creatorcontrib><creatorcontrib>Subramanya, Suhas Jayaram</creatorcontrib><creatorcontrib>Krishnaswamy, Ravishankar</creatorcontrib><creatorcontrib>Simhadri, Harsha Vardhan</creatorcontrib><title>FreshDiskANN: A Fast and Accurate Graph-Based ANN Index for Streaming Similarity Search</title><description>Approximate nearest neighbor search (ANNS) is a fundamental building block in information retrieval with graph-based indices being the current state-of-the-art and widely used in the industry. Recent advances in graph-based indices have made it possible to index and search billion-point datasets with high recall and millisecond-level latency on a single commodity machine with an SSD. However, existing graph algorithms for ANNS support only static indices that cannot reflect real-time changes to the corpus required by many key real-world scenarios (e.g. index of sentences in documents, email, or a news index). To overcome this drawback, the current industry practice for manifesting updates into such indices is to periodically re-build these indices, which can be prohibitively expensive. In this paper, we present the first graph-based ANNS index that reflects corpus updates into the index in real-time without compromising on search performance. Using update rules for this index, we design FreshDiskANN, a system that can index over a billion points on a workstation with an SSD and limited memory, and support thousands of concurrent real-time inserts, deletes and searches per second each, while retaining $&gt;95\%$ 5-recall@5. This represents a 5-10x reduction in the cost of maintaining freshness in indices when compared to existing methods.</description><subject>Computer Science - Information Retrieval</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz71OwzAYhWEvDKhwAUz4BhL8FyfpFgoplaowpBJj9Dn-3Fg0obIDau8eKExHeocjPYTccZaqIsvYA4ST_0oFZ1nKSs3lNXmrA8bhycf3qmmWtKI1xJnCZGnV958BZqTrAMcheYSIP7Fp6GayeKLuI9B2Dgijn_a09aM_QPDzmbYIoR9uyJWDQ8Tb_12QXf28W70k29f1ZlVtE9C5TBwaXRirSiZzyMGa0mAhBJasQJZzq3InhREWjJROaC2Ro3JWFdBzXWZaLsj93-1F1h2DHyGcu19hdxHKb170SqI</recordid><startdate>20210520</startdate><enddate>20210520</enddate><creator>Singh, Aditi</creator><creator>Subramanya, Suhas Jayaram</creator><creator>Krishnaswamy, Ravishankar</creator><creator>Simhadri, Harsha Vardhan</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20210520</creationdate><title>FreshDiskANN: A Fast and Accurate Graph-Based ANN Index for Streaming Similarity Search</title><author>Singh, Aditi ; Subramanya, Suhas Jayaram ; Krishnaswamy, Ravishankar ; Simhadri, Harsha Vardhan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-feb68bd49037a7adb9be822e908e071d47f32b2dab33f2663e1e4fd48ac169563</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Computer Science - Information Retrieval</topic><toplevel>online_resources</toplevel><creatorcontrib>Singh, Aditi</creatorcontrib><creatorcontrib>Subramanya, Suhas Jayaram</creatorcontrib><creatorcontrib>Krishnaswamy, Ravishankar</creatorcontrib><creatorcontrib>Simhadri, Harsha Vardhan</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Singh, Aditi</au><au>Subramanya, Suhas Jayaram</au><au>Krishnaswamy, Ravishankar</au><au>Simhadri, Harsha Vardhan</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>FreshDiskANN: A Fast and Accurate Graph-Based ANN Index for Streaming Similarity Search</atitle><date>2021-05-20</date><risdate>2021</risdate><abstract>Approximate nearest neighbor search (ANNS) is a fundamental building block in information retrieval with graph-based indices being the current state-of-the-art and widely used in the industry. Recent advances in graph-based indices have made it possible to index and search billion-point datasets with high recall and millisecond-level latency on a single commodity machine with an SSD. However, existing graph algorithms for ANNS support only static indices that cannot reflect real-time changes to the corpus required by many key real-world scenarios (e.g. index of sentences in documents, email, or a news index). To overcome this drawback, the current industry practice for manifesting updates into such indices is to periodically re-build these indices, which can be prohibitively expensive. In this paper, we present the first graph-based ANNS index that reflects corpus updates into the index in real-time without compromising on search performance. Using update rules for this index, we design FreshDiskANN, a system that can index over a billion points on a workstation with an SSD and limited memory, and support thousands of concurrent real-time inserts, deletes and searches per second each, while retaining $&gt;95\%$ 5-recall@5. This represents a 5-10x reduction in the cost of maintaining freshness in indices when compared to existing methods.</abstract><doi>10.48550/arxiv.2105.09613</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2105.09613
ispartof
issn
language eng
recordid cdi_arxiv_primary_2105_09613
source arXiv.org
subjects Computer Science - Information Retrieval
title FreshDiskANN: A Fast and Accurate Graph-Based ANN Index for Streaming Similarity Search
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-20T19%3A03%3A35IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=FreshDiskANN:%20A%20Fast%20and%20Accurate%20Graph-Based%20ANN%20Index%20for%20Streaming%20Similarity%20Search&rft.au=Singh,%20Aditi&rft.date=2021-05-20&rft_id=info:doi/10.48550/arxiv.2105.09613&rft_dat=%3Carxiv_GOX%3E2105_09613%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true