Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Comprehensive Survey on Vector Database: Storage and Retrieval Technique, Challenge

Le Ma, Ran Zhang, Yikun Han, Shirui Yu, Zaitian Wang, Zhiyuan Ning, Jinghan Zhang, Ping Xu, Pengjiang Li, Wei Ju, Chong Chen, Dongjie Wang, Kunpeng Liu, Pengyang Wang, Pengfei Wang, Yanjie Fu, Chunjiang Liu, Yuanchun Zhou, Chang-Tien Lu

TL;DR

The paper surveys vector databases, focusing on storage and retrieval of high-dimensional embeddings and their integration with large language models. It catalogs storage techniques (sharding, partitioning, caching, replication) and search methods (NNS and ANNS across hash-, tree-, graph-, and quantization-based families), and benchmarks leading systems. It discusses key challenges in indexing, data heterogeneity, distributed processing, and security, and highlights opportunities for LLM–VDB synergy through retrieval-augmented generation, semantic caching, and memory frameworks. The work also outlines a practical framework and identifies open problems and trends, guiding researchers and practitioners in building scalable, AI-enabled vector databases.

Abstract

Vector databases (VDBs) have emerged to manage high-dimensional data that exceed the capabilities of traditional database management systems, and are now tightly integrated with large language models as well as widely applied in modern artificial intelligence systems. Although relatively few studies describe existing or introduce new vector database architectures, the core technologies underlying VDBs, such as approximate nearest neighbor search, have been extensively studied and are well documented in the literature. In this work, we present a comprehensive review of the relevant algorithms to provide a general understanding of this booming research area. Specifically, we first provide a review of storage and retrieval techniques in VDBs, with detailed design principles and technological evolution. Then, we conduct an in-depth comparison of several advanced VDB solutions with their strengths, limitations, and typical application scenarios. Finally, we also outline emerging opportunities for coupling VDBs with large language models, including open research problems and trends, such as novel indexing strategies. This survey aims to serve as a practical resource, enabling readers to quickly gain an overall understanding of the current knowledge landscape in this rapidly developing area.

A Comprehensive Survey on Vector Database: Storage and Retrieval Technique, Challenge

TL;DR

The paper surveys vector databases, focusing on storage and retrieval of high-dimensional embeddings and their integration with large language models. It catalogs storage techniques (sharding, partitioning, caching, replication) and search methods (NNS and ANNS across hash-, tree-, graph-, and quantization-based families), and benchmarks leading systems. It discusses key challenges in indexing, data heterogeneity, distributed processing, and security, and highlights opportunities for LLM–VDB synergy through retrieval-augmented generation, semantic caching, and memory frameworks. The work also outlines a practical framework and identifies open problems and trends, guiding researchers and practitioners in building scalable, AI-enabled vector databases.

Abstract

Vector databases (VDBs) have emerged to manage high-dimensional data that exceed the capabilities of traditional database management systems, and are now tightly integrated with large language models as well as widely applied in modern artificial intelligence systems. Although relatively few studies describe existing or introduce new vector database architectures, the core technologies underlying VDBs, such as approximate nearest neighbor search, have been extensively studied and are well documented in the literature. In this work, we present a comprehensive review of the relevant algorithms to provide a general understanding of this booming research area. Specifically, we first provide a review of storage and retrieval techniques in VDBs, with detailed design principles and technological evolution. Then, we conduct an in-depth comparison of several advanced VDB solutions with their strengths, limitations, and typical application scenarios. Finally, we also outline emerging opportunities for coupling VDBs with large language models, including open research problems and trends, such as novel indexing strategies. This survey aims to serve as a practical resource, enabling readers to quickly gain an overall understanding of the current knowledge landscape in this rapidly developing area.
Paper Structure (30 sections, 44 equations, 9 figures, 4 tables)

This paper contains 30 sections, 44 equations, 9 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Storage
  3. Sharding
  4. Partitioning
  5. Caching
  6. Replication
  7. Search
  8. Nearest Neighbor Search
  9. Brute Force Approach
  10. Tree-Based Approach
  11. Approximate Nearest Neighbor Search
  12. Hash-Based Approach
  13. Tree-Based Approach
  14. Graph-Based Approach
  15. Quantization-Based Approach
  16. ...and 15 more sections

Figures (9)

  • Figure 1: Framework overview of this survey structure covering Storage Techniques, Search Techniques, Database Comparison, Challenges, and the Synergy of Large Language Models (LLMs) with VDBs. Each section represents a fundamental facet of the operation and integration of modern VDBs within advanced AI technologies.
  • Figure 2: Taxonomy of Vector Database (VDB) Storage and Search Technologies.
  • Figure 3: The process of approximate nearest neighbor search based on hash approach
  • Figure 4: The process of performing nearest neighbor search on a small-world network.
  • Figure 5: Codewords and Voronoi Regions in a Two-Dimensional Space
  • ...and 4 more figures