DATABASES FOR GENERATIVE MODELS: ARCHITECTURES, CHALLENGES, AND PROSPECTS

Kunitsyn V.; Luchnikov V.; Ilyin E.; Kartashov A.

doi:doi:10.58168/PPSPM2025_108-120

Home / Conferences / PROBLEMS AND PROSPECTS IN SYSTEMS AND PROCESS MODELING / PROBLEMS AND PROSPECTS IN SYSTEMS AND PROCESS MODELING : Materials of the All-Russian Scientific and Practical Conference, Voronezh, October 14, 2025

DATABASES FOR GENERATIVE MODELS: ARCHITECTURES, CHALLENGES, AND PROSPECTS

Submit manuscript

To cite

DATABASES FOR GENERATIVE MODELS: ARCHITECTURES, CHALLENGES, AND PROSPECTS

Section: SECTION 1. SYSTEMS ANALYSIS, MANAGEMENT, INFORMATION PROCESSING

Proceedings: PROBLEMS AND PROSPECTS IN SYSTEMS AND PROCESS MODELING : MATERIALS OF THE ALL-RUSSIAN SCIENTIFIC AND PRACTICAL CONFERENCE, VORONEZH, OCTOBER 14, 2025

UDC 004.9

Kunitsyn V. ¹

Luchnikov V. ²

Ilyin E. ³

Kartashov A. ⁴

Author and publication information

Authors:

1. Voronezh State University of Forestry and Technologies named after G.F. Morozov
Russian Federation

2. Voronezh State University of Forestry and Technologies named after G.F. Morozov
Russian Federation

3. Voronezh State University of Forestry and Technologies named after G.F. Morozov
Russian Federation

4. Voronezh State University of Forestry and Technologies named after G.F. Morozov
Russian Federation

Type:

Сonference article

DOI:

https://doi.org/10.58168/PPSPM2025_108-120

Pages:

from 108 to 120

Published:

15.01.2026

Subject area:

UDC 004.9

Language:

Russian

Keywords:

generative models; databases; vector databases; embeddings; semantic search; RAG; indexing; machine learning; data security; federated learning

Abstract and keywords

Abstract (English):
The rise of generative artificial intelligence—from large language models to diffusion and adversarial architectures—has laid bare the strategic role of data. Models create new content, but the quality of their outputs is determined by how training and auxiliary datasets are collected, stored, indexed, and delivered. Traditional relational DBMSs and widely used NoSQL stores have provided the foundation for many projects; however, the specifics of generative systems have shifted the focus: there is a need to store vector representations (embeddings), to perform similarity-based (semantic) retrieval, and to support high-throughput reads during inference. This monograph systematizes database architectures applicable to generative scenarios, reviews indexing methods and integration with models, and analyzes risks and future directions.

Keywords:
generative models; databases; vector databases; embeddings; semantic search; RAG; indexing; machine learning; data security; federated learning

References

1. Garcia-Molina, H.; Ullman, J.; Widom, J. Database Systems: The Complete Book. – 2nd ed. – Pearson, 2023. – 1248 p.

2. Stonebraker, M.; Hellerstein, J. M. Readings in Database Systems (5th ed.). – MIT Press, 2024. – 620 p.

3. Pinecone. Vector Database for Machine Learning. – Rezhimdostupa: https://www.pinecone.io/, svobodnyy. – Dataobrascheniya: 10.09.2025.

4. Weaviate. Open-SourceVectorDatabase. – Rezhim dostupa: https://weaviate.io/, svobodnyy. – Data obrascheniya: 09.09.2025.

5. Milvus. Open-SourceVectorDatabaseforAI. – Rezhim dostupa: https://milvus.io/, svobodnyy. – Data obrascheniya: 12.09.2025.

6. FAISS (FacebookAISimilaritySearch). – Rezhim dostupa: https://github.com/facebookresearch/faiss, svobodnyy. – Data obrascheniya: 15.09.2025.

7. TimescaleInc. TimescaleDBDocumentation. – Rezhim dostupa: https://docs.timescale.com/, svobodnyy. – Data obrascheniya: 15.09.2025.

8. InfluxData. InfluxDBDocumentation – Rezhim dostupa: https://docs.influxdata.com/, svobodnyy. – Data obrascheniya: 15.09.2025.

9. OpenAI. Retrieval-AugmentedGeneration: BestPractices. – Rezhim dostupa: https://platform.openai.com/, svobodnyy. – Data obrascheniya: 15.09.2025.

Submit manuscript

To cite

Citations:

Confirmation

Регистрация