Database Optimization for Low-Latency Analytics with Adaptive Indexing

Abstract

Data are rapidly produced in various fields such as finance, e-commerce, healthcare, and the Internet of Things (IoT), resulting in a constantly growing need for real-time analytics that can scale up. Traditional relational database management systems (RDBMSs) are often unable to keep up with high-speed processing requirements due to the limitations of ACID properties. In this report, auto optimized stream is introduced as a self-adaptive optimization framework to enhance real-time analytical pipelines powered by PostgreSQL, Kafka, and Flink in terms of both scalability and performance. The major difference between the proposed framework and the static optimization methods, which are used by Oracle, is that it is a continuous process where live query workloads and data ingestion are monitored regularly, thus allowing for the optimization of indexing strategies, partitioning schemes, and stream processing configurations. The framework employs rule-based decision-making to improve query execution, ingestion throughput, and resource utilization across the board. The adaptive method, when compared to static optimization methods, allows lower query latency, better stream stability, and higher CPU efficiency while causing minimal overhead and keeping full tuning action traceability. These findings not only prove the practical viability of adaptive database systems for real-time analytics but also lay down a steppingstone for future systems.

Received: 8 September 2025 | Revised: 24 November 2025 | Accepted: 12 January 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in "vega-datasets" at https://github.com/vega/vega/blob/main/docs/data/seattle-weather.csv.

Author Contribution Statement

Lakmali Karunarathne: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing — original draft, Writing — review & editing, Visualization, Supervision, Project administration. Swathi Ganesan: Investigation. Kavindu Karunarathne: Methodology, Software, Formal analysis, Resources, Writing — original draft, Writing — review & editing, Visualization. Nalinda Somasiri: Investigation.