N.S. Mishin1
1 Bauman Moscow State Technical University (Moscow, Russia)
1 stancuem@yandex.ru

Modern multi-model databases allow working with heterogeneous data using various data models (relational, graph, multidimensional). However, choosing the optimal model for storing specific types of data remains a task that directly affects the performance of the system.

The purpose of the article is to develop a method for determining the optimal data models for each group of queries based on execution time. By experimentally analyzing the performance of relational, graph, and multidimensional models, it is possible to select the optimal model that minimizes query execution time and support costs.

The analysis showed that different data models provide different performance for each group of queries. The constructed polynomial models and clustering of execution times made it possible to select the optimal model for each group of data, minimizing the system response time.

The proposed methodology for selecting a data model in a multi-model database improves system performance by reducing the total query execution time and the costs of maintaining and synchronizing several models.

Mishin N.S. Optimization of the data model of a multi-model DB for different types of queries // Neurocomputers. 2026. V. 28. № 3. P. 5–9. DOI: https://doi.org/10.18127/j19998554-202603-01.

1. Li G., Feng J., Ooi B.C., Wang J., Zhou L. n effective 3-in-1 keyword search method over heterogeneous data source. Information Systems. 2021. V. 36. № 2. P. 248–266. DOI 10.1016/j.is.2008.08.001.
2. Lu J., Holubova I. Multi-model Databases: A New Journey to Handle the Variety of Data. ACM Computing Surveys (CSUR). 2019. V. 52. № 3. P. 1–38. DOI 10.1145/3323214.
3. Zhang C., Lu J., Xu P., Chen Y. Unibench: A benchmark for multi-model database management systems. Performance Evaluation and Benchmarking for the Era of Artificial Intelligence: 10th TPC Technology Conference. TPCTC 2018. P. 7–23. Lecture Notes in Computer Science. V. 11135. Springer International Publishing, 2019. DOI 10.1007/978-3-030-11404-6_2.
4. Yuan G. How the quantum-inspired framework supports keyword searches on multi-model Database. Proceedings of the 29th ACM International Conference on Information & Knowledge Management. ACM, 2020. P. 3257–3260. DOI 10.1145/3340531.3418508.
5. Holubova I., Vavrek M., Scherzinger S. Evolution management in multi-model databases. Data & Knowledge Engineering. 2021. V. 136. Article 101932. DOI 10.1016/j.datak.2021.101932.
6. Liu Z.H., Lu J., Gawlick D., Helskyaho H., Pogossiants G., Wu Z. Multi-model Database Management Systems – A Look Forward. Hete­rogeneous Data Management, Polystores, and Analytics for Healthcare. DMAH Poly 2018. Lecture Notes in Computer Science.
   V. 11470. P. 16–29. Springer, Cham, 2019. DOI 10.1007/978-3-030-14177-6_2.
7. Zhang C., Lu J. Holistic evaluation in multi-model databases benchmarking. Distributed and Parallel Databases. 2021. V. 39. P. 1–33. DOI 10.1007/s10619-019-07279-6.