The vast quantities of data generated today, referred to as Big Data, offer significant opportunities and applications for researchers across various fields. Even so, managing Big Data requires addressing significant demands on time and computational resources, making the optimization of database management system (DBMS) performance essential. Numerous studies have employed online analytical processing (OLAP) to compare the efficiency of relational and non-relational DBMS, aiming to identify the most effective systems for handling extensive data workloads. By covering a decade of research, this survey captures the variety of benchmarks used for performance comparison. The main comparative articles were selected from three primary academic platforms: Springer Nature Link, Scopus, and Web of Science, filtering studies published between 2013 and 2024. Out of a total of 3,285 papers, 41 relevant studies were identified using full-text examination and backward snowballing technique. This survey investigates (1) the databases examined across diverse architectures, (2) the methodologies employed, and (3) synthesizes the performance results to provide a comprehensive understanding of database efficiency.