Key points are not available for this paper at this time.
In the burgeoning realm of big data, document-oriented NoSQL databases stand out for their flexibility and scalability. This paper delves into the optimization of these databases, specifically through the lens of parallel processing techniques. A comparative study was conducted against the traditional non-parallel approaches, where marked performance enhancements were observed. For instance, the execution time for retrieving movies of a specific year decreased by over 80% when parallel processing was applied, plummeting from 1.578765 seconds to a brisk 0.300000 seconds. Memory usage and CPU utilization were meticulously recorded, revealing up to a 70% reduction in peak memory consumption in certain queries, and a moderate fluctuation in CPU usage between 49.25% to 75.2%. This indicates not only improved efficiency but also a prudent utilization of system capacity, without overtaxing resources. However, the study identified scenarios, such as highly complex queries, where the gains from parallel processing were less pronounced, suggesting a marginal improvement in CPU utilization. While the findings advocate for the adoption of parallel processing in handling intensive data retrieval tasks, it is recommended that future research should further scrutinize the scalability thresholds and explore alternative parallelization strategies to fortify the efficacy of document-oriented NoSQL databases.
Inas Ismael Imran (Mon,) studied this question.