Performance Evaluation of Query Response Time in The Document Stored NoSQL Database

Abstract

NoSQL Database is one solution that can be used to store large amounts of data and have an excellent performance during the reading and writing process. Analysis of data with large volumes increasingly feels the benefits. Database exploration is one of the activities carried out in data analysis. Information on estimated response time queries can help in planning database exploration. This study aims to evaluate the response time of each query for the process of creating, read, update, delete (CRUD) on a document stored, NoSQL Database. MongoDB, ArangoDB, and CouchDB were selected for use in experiments. Experiments are carried out by processing queries for repeated create, read, update, and delete commands with different quantities. The experimental results in the study revealed that MongoDB had the lowest average query response time for the read process (0.017 seconds), update (25,358 seconds) and delete (0.055 seconds) compared to ArangoDB and CouchDB. However, for the creation process, the ArangoDB response time is the smallest, which is 28,493 seconds compared to MongoDB and CouchDB.