PosDB: An Architecture Overview

Chernishev, George; Galaktionov, Vyacheslav; Grigorev, Valentin; Klyuchikov, Evgeniy; Smirnov, Kirill

doi:10.1134/s0361768818010024

Cited by 4 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our study of window functions is conducted inside PosDB -a distributed diskbased column-store. A detailed discussion of its architecture can be found in papers [7,8]. Here, we only describe the most important points.…”

Section: Posdb Basicsmentioning

confidence: 99%

“…It is common to implement the segment tree on the base of an array with an implicit tree structure, since the segment tree is always a complete tree. For a complete tree, an array is the most space-efficient representation 7 .…”

Section: Range-based Window Functionsmentioning

confidence: 99%

“…Employing late materialization may allow to speed up processing of such queries and reduce memory footprint, which is very important for this operation. We validate our approach by implementing it inside PosDB [7,8] a distributed disk-based column-store with late materialization -and by comparing it with PostgreSQL.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Implementing Window Functions in a Column-Store with Late Materialization

Mukhaleva¹,

Grigorev²,

Chernishev³

2019

Model and Data Engineering

View full text Add to dashboard Cite

A window function is a generalization of the aggregation operation. Unlike aggregation, the cardinality of its output is always the same as the cardinality of input. That is, the semantics of this operator imply computing values for extra attributes for each row, depending on its context, either expressed by a sliding window or a previously evaluated row. Window functions are a very powerful tool, which is also popular among data analysts and supported by the majority of industrial DBMSes. It allows to gracefully express quite complex use-cases, such as running sums and averages, local maximum and minimum, and different types of ranking. Since they can be expressed without self-joins and correlated subqueries, their evaluation can be performed much more efficiently. In this paper we discuss an implementation of window functions inside a disk-based column-store with late materialization. Late materialization is a technique that aims to keep tuple reconstruction back from individual columns as long as possible. Initially popular in the late 00's, it is rarely considered nowadays. However, in case of window functions it allows to substantially lower memory footprint. Another contribution of this paper is the application of a segment tree to computing RANGE-based window functions.

show abstract

Section: Posdb Basicsmentioning

confidence: 99%

Section: Range-based Window Functionsmentioning

confidence: 99%