Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Survey of Learned Indexes for the Multi-dimensional Space

Abdullah Al-Mamun, Hao Wu, Qiyang He, Jianguo Wang, Walid G. Aref

TL;DR

This survey presents a taxonomy that classifies and categorizes both learned one- and multi-dimensional indexes, and surveys the existing literature on learned indexes according to this taxonomy with an emphasis on learned multi-dimensional index structures.

Abstract

A recent research trend involves treating database index structures as Machine Learning (ML) models. In this domain, single or multiple ML models are trained to learn the mapping from keys to positions inside a data set. This class of indexes is known as "Learned Indexes." Learned indexes have demonstrated improved search performance and reduced space requirements for one-dimensional data. The concept of one-dimensional learned indexes has naturally been extended to multi-dimensional (e.g., spatial) data, leading to the development of "Learned Multi-dimensional Indexes". This survey focuses on learned multi-dimensional index structures. Specifically, it reviews the current state of this research area, explains the core concepts behind each proposed method, and classifies these methods based on several well-defined criteria. We present a taxonomy that classifies and categorizes each learned multi-dimensional index, and survey the existing literature on learned multi-dimensional indexes according to this taxonomy. Additionally, we present a timeline to illustrate the evolution of research on learned indexes. Finally, we highlight several open challenges and future research directions in this emerging and highly active field.

A Survey of Learned Indexes for the Multi-dimensional Space

TL;DR

This survey presents a taxonomy that classifies and categorizes both learned one- and multi-dimensional indexes, and surveys the existing literature on learned indexes according to this taxonomy with an emphasis on learned multi-dimensional index structures.

Abstract

A recent research trend involves treating database index structures as Machine Learning (ML) models. In this domain, single or multiple ML models are trained to learn the mapping from keys to positions inside a data set. This class of indexes is known as "Learned Indexes." Learned indexes have demonstrated improved search performance and reduced space requirements for one-dimensional data. The concept of one-dimensional learned indexes has naturally been extended to multi-dimensional (e.g., spatial) data, leading to the development of "Learned Multi-dimensional Indexes". This survey focuses on learned multi-dimensional index structures. Specifically, it reviews the current state of this research area, explains the core concepts behind each proposed method, and classifies these methods based on several well-defined criteria. We present a taxonomy that classifies and categorizes each learned multi-dimensional index, and survey the existing literature on learned multi-dimensional indexes according to this taxonomy. Additionally, we present a timeline to illustrate the evolution of research on learned indexes. Finally, we highlight several open challenges and future research directions in this emerging and highly active field.
Paper Structure (105 sections, 4 figures, 4 tables)

This paper contains 105 sections, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Surveys and Tutorials
  3. Evolution of Learned Indexes
  4. Indexing the Learned Models
  5. Learned Immutable Indexes
  6. The One-dimensional Case
  7. Pure Learned Indexes
  8. RMI kraska2018case
  9. Hybrid Learned Indexes
  10. RS kipf2020radixspline
  11. The Multi-dimensional Case
  12. Pure Learned Indexes in the Projected Space
  13. SageDB-LMI kraska2019sagedb
  14. ZM-index wang2019learned
  15. ML-index davitkova2020ml
  16. ...and 90 more sections

Figures (4)

  • Figure 1: Spectrum of learned indexes
  • Figure 2: Taxonomy of learned indexes. The end of a branch indicates that there are no papers in that category as of the time this article has been written. An asterisk (*) symbol is used if the index natively supports concurrency. The hybrid learned indexes are categorized based on their underlying traditional data structures.
  • Figure 3: Illustration of the taxonomy criteria
  • Figure 4: Timeline of the evolution of learned indexes. Lines connecting between the various learned multi-dimensional indexes reflect dependence of these indexes on earlier work.