GraphQL query performance comparison using MySQL and MongoDB: By conducting Experiments with and without a DataLoader
2022 (English)Independent thesis Basic level (degree of Bachelor), 180 HE credits
Student thesis
Abstract [en]
GraphQL is a query language rising in popularity, causing many transitions from traditional API endpoints to a GraphQL solution. Reflecting upon the positives and the flaws to using GraphQL, a DataLoader for batching queries sent to databases is sold as a solution to the infamous N+1 problem. Experiments were conducted to test how GraphQL response time, with and without DataLoader, changes when paired with MySQL and MongoDB. Along with the experiments, a Literature Review was conducted reflecting over the databases structural differences that could affect the response time for GraphQL. Results suggest that no major differences were to be found, and the explanation for the minor differences could rather be because of the disparity in query optimization instead of architectural differences for MySQL and MongoDB.
Place, publisher, year, edition, pages
2022. , p. 42
Keywords [en]
DataLoader, GraphQL, MongoDB, MySQL, Query, Query Performance
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:hj:diva-58563ISRN: JU-JTH-DTA-1-20220179OAI: oai:DiVA.org:hj-58563DiVA, id: diva2:1700084
External cooperation
Verendus System AB
Subject / course
JTH, Computer Engineering
Supervisors
Examiners
2022-09-302022-09-292025-10-13Bibliographically approved