Relative error streaming quantiles

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F21%3A10438016" target="_blank" >RIV/00216208:11320/21:10438016 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1145/3452021.3458323" target="_blank" >https://doi.org/10.1145/3452021.3458323</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3452021.3458323" target="_blank" >10.1145/3452021.3458323</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Relative error streaming quantiles

Popis výsledku v původním jazyce

Approximating ranks, quantiles, and distributions over streaming data is a central task in data analysis and monitoring. Given a stream of n items from a data universe U equipped with a total order, the task is to compute a sketch (data structure) of size poly (log(n), 1/ϵ). Given the sketch and a query item y in U, one should be able to approximate its rank in the stream, i.e., the number of stream elements smaller than or equal to y. Most works to date focused on additive ϵ n error approximation, culminating in the KLL sketch that achieved optimal asymptotic behavior. This paper investigates multiplicative (1+-ϵ)-error approximations to the rank. Practical motivation for multiplicative error stems from demands to understand the tails of distributions, and hence for sketches to be more accurate near extreme values. The most space-efficient algorithms due to prior work store either O(log(ϵ^2 n) / ϵ^2) or O(log^3(ϵ n) / ϵ) universe items. This paper presents a randomized algorithm storing O(log^{1.5} (ϵ n)/ϵ) items, which is within an O(sqrt{log(ϵ n)}) factor of optimal. The algorithm does not require prior knowledge of the stream length and is fully mergeable, rendering it suitable for parallel and distributed computing environments. (C) 2021 ACM.

Název v anglickém jazyce

Relative error streaming quantiles

Popis výsledku anglicky

Approximating ranks, quantiles, and distributions over streaming data is a central task in data analysis and monitoring. Given a stream of n items from a data universe U equipped with a total order, the task is to compute a sketch (data structure) of size poly (log(n), 1/ϵ). Given the sketch and a query item y in U, one should be able to approximate its rank in the stream, i.e., the number of stream elements smaller than or equal to y. Most works to date focused on additive ϵ n error approximation, culminating in the KLL sketch that achieved optimal asymptotic behavior. This paper investigates multiplicative (1+-ϵ)-error approximations to the rank. Practical motivation for multiplicative error stems from demands to understand the tails of distributions, and hence for sketches to be more accurate near extreme values. The most space-efficient algorithms due to prior work store either O(log(ϵ^2 n) / ϵ^2) or O(log^3(ϵ n) / ϵ) universe items. This paper presents a randomized algorithm storing O(log^{1.5} (ϵ n)/ϵ) items, which is within an O(sqrt{log(ϵ n)}) factor of optimal. The algorithm does not require prior knowledge of the stream length and is fully mergeable, rendering it suitable for parallel and distributed computing environments. (C) 2021 ACM.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2021

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems

ISBN

978-1-4503-8381-3

ISSN

—

e-ISSN

—

Počet stran výsledku

13

Strana od-do

96-108

Název nakladatele

Association for Computing Machinery

Místo vydání

Neuveden

Místo konání akce

Virtual (Xi&apos;an, Shaanxi, China)

Datum konání akce

20. 6. 2021

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—