On the Costs and Benefits of Learned Indexing for Dynamic
High-Dimensional Data

D 2025

On the Costs and Benefits of Learned Indexing for Dynamic High-Dimensional Data

SLANINÁKOVÁ, Terézia; Jaroslav OĽHA; David PROCHÁZKA; Matej ANTOL; Vlastislav DOHNAL et. al.

Základní údaje

Originální název

On the Costs and Benefits of Learned Indexing for Dynamic High-Dimensional Data

Autoři

SLANINÁKOVÁ, Terézia; Jaroslav OĽHA; David PROCHÁZKA; Matej ANTOL a Vlastislav DOHNAL

Vydání

1. vyd. Cham, Big Data Analytics and Knowledge Discovery 27th International Conference, DaWaK 2025, Bangkok, Thailand, August 25–27, 2025, Proceedings, od s. 251-258, 8 s. 2025

Nakladatel

Springer Cham

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10200 1.2 Computer and information sciences

Stát vydavatele

Švýcarsko

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

tištěná verze "print"

Odkazy

URL

Organizační jednotka

Fakulta informatiky

ISBN

978-3-032-02214-1

DOI

http://dx.doi.org/10.1007/978-3-032-02215-8_20

Klíčová slova anglicky

Learned indexing; Dynamization; Dynamic datasets; kNN search ; ANN search

Štítky

ANN search, DISA, dynamic data, Dynamic datasets, Dynamization, index structure, kNN search, learned index, learned indexing, LMI, nearest-neighbors query, performance evaluation, scalability, similarity search

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 21. 8. 2025 14:23, doc. RNDr. Vlastislav Dohnal, Ph.D.

Anotace

V originále

One of the main challenges within the growing research area of learned indexing is the lack of adaptability to dynamically expanding datasets . This paper explores the dynamization of a static learned index for complex data through operations such as node splitting and broadening, enabling efficient adaptation to new data. Furthermore, we evaluate the trade-offs between static and dynamic approaches by introducing an amortized cost model to assess query performance in tandem with the build costs of the index structure, enabling experimental determination of when a dynamic learned index outperforms its static counterpart. We apply the dynamization method to a static learned index and demonstrate that its superior scaling quickly surpasses the static implementation in terms of overall costs as the database grows.

Návaznosti

GF23-07040K, projekt VaV

Název: Naučené indexy pro podobností hledání

Investor: Grantová agentura ČR, Naučené indexy pro podobností hledání, Lead agentura

LM2018131, projekt VaV

Název: Česká národní infrastruktura pro biologická data (Akronym: ELIXIR-CZ)

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, Česká národní infrastruktura pro biologická data

LM2018140, projekt VaV

Název: e-Infrastruktura CZ (Akronym: e-INFRA CZ)

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, e-Infrastruktura CZ

MUNI/A/1638/2024, interní kód MU

Název: Umělá inteligence a správa komplexních rozsáhlých dat

Investor: Masarykova univerzita, Umělá inteligence a správa komplexních rozsáhlých dat

Citovat

SLANINÁKOVÁ, Terézia; Jaroslav OĽHA; David PROCHÁZKA; Matej ANTOL a Vlastislav DOHNAL. On the Costs and Benefits of Learned Indexing for Dynamic High-Dimensional Data. In Carson K. Leung, Anton Dignös, Gabriele Kotsis, A. Min Tjoa, Ismail Khalil. Big Data Analytics and Knowledge Discovery 27th International Conference, DaWaK 2025, Bangkok, Thailand, August 25–27, 2025, Proceedings. 1. vyd. Cham: Springer Cham, 2025, s. 251-258. ISBN 978-3-032-02214-1. Dostupné z: https://dx.doi.org/10.1007/978-3-032-02215-8_20.

@inproceedings{2512838,
   author = {Slanináková, Terézia and Oľha, Jaroslav and Procházka, David and Antol, Matej and Dohnal, Vlastislav},
   address = {Cham},
   booktitle = {Big Data Analytics and Knowledge Discovery 27th International Conference, DaWaK 2025, Bangkok, Thailand, August 25–27, 2025, Proceedings},
   doi = {http://dx.doi.org/10.1007/978-3-032-02215-8_20},
   edition = {1},
   editor = {Carson K. Leung, Anton Dignös, Gabriele Kotsis, A. Min Tjoa, Ismail Khalil},
   keywords = {Learned indexing; Dynamization; Dynamic datasets; kNN search ; ANN search},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Cham},
   isbn = {978-3-032-02214-1},
   pages = {251-258},
   publisher = {Springer Cham},
   title = {On the Costs and Benefits of Learned Indexing for Dynamic High-Dimensional Data},
   url = {https://link.springer.com/chapter/10.1007/978-3-032-02215-8_20},
   year = {2025}
}

TY  - CONF
ID  - 2512838
AU  - Slanináková, Terézia - Oľha, Jaroslav - Procházka, David - Antol, Matej - Dohnal, Vlastislav
PY  - 2025
TI  - On the Costs and Benefits of Learned Indexing for Dynamic High-Dimensional Data
PB  - Springer Cham
CY  - Cham
SN  - 9783032022141
KW  - Learned indexing
KW  - Dynamization
KW  - Dynamic datasets
KW  - kNN search
KW  - ANN search
UR  - https://link.springer.com/chapter/10.1007/978-3-032-02215-8_20
N2  - One of the main challenges within the growing research area of learned indexing is the lack of adaptability to dynamically expanding datasets . This paper explores the dynamization of a static learned index for complex data through operations such as node splitting and broadening, enabling efficient adaptation to new data. Furthermore, we evaluate the trade-offs between static and dynamic approaches by introducing an amortized cost model to assess query performance in tandem with the build costs of the index structure, enabling experimental determination of when a dynamic learned index outperforms its static counterpart. We apply the dynamization method to a static learned index and demonstrate that its superior scaling quickly surpasses the static implementation in terms of overall costs as the database grows.
ER  -

SLANINÁKOVÁ, Terézia; Jaroslav OĽHA; David PROCHÁZKA; Matej ANTOL a Vlastislav DOHNAL. On the Costs and Benefits of Learned Indexing for Dynamic High-Dimensional Data. In Carson K. Leung, Anton Dignös, Gabriele Kotsis, A. Min Tjoa, Ismail Khalil. \textit{Big Data Analytics and Knowledge Discovery 27th International Conference, DaWaK 2025, Bangkok, Thailand, August 25–27, 2025, Proceedings}. 1. vyd. Cham: Springer Cham, 2025, s.~251-258. ISBN~978-3-032-02214-1. Dostupné z: https://dx.doi.org/10.1007/978-3-032-02215-8\_{}20.

Přehled o publikaci