•  
  •  
 

Corresponding Author ORCID Identifer

Rehab Mohammed Abdelhalim Mokhtar https://orcid.org/0009-0004-8904-6335

Section

Regular Issue

Abstract

As databases continue to grow and complexity there is an urgent need to achieve more efficient and faster query processing. This paper presents a technical review of methods that work towards optimization of various types of queries. It progresses by focusing on basic approaches like rules-based optimization and cost-based optimization. Based on this basis, the paper presents a synthesized reference architecture for adaptive and learned query optimizers as a generic framework derived from prior system. It explains on how the classical optimizer components can be enhanced through runtime monitoring integration and feedback mechanisms. From this perspective, it becomes possible to explain how learning modules and execution feedback should be added to the standard optimization pipelines instead of requiring any revolutionary changes in the current database systems. Next, various Adaptive Query Processing (AQP) approaches are reviewed to highlight their strengths and limitations. A fundamental aspect to consider with respect to decision-making process is covered in how effective these methods are from a reoptimization point of view, as well as with adaptive operators. The final section discussed the emerging trends in this field, including some current developments and what is expected to come regarding query optimization alongside machine learning integration, and distributed queries handling advancement issues. By synthesizing insights from existing literature, this review equips researchers with a robust understanding of the current state and future potential of query optimization and adaptive processing techniques.

Download

Share

COinS