Research on Data-Driven Performance Evaluation Methods for the Internet of Data
Yajie Dou, Jing Sun · 2024
In order to more efficiently test the performance of Internet of Data applications based on Digital Object Architecture (DOA), this paper proposes a data-driven, automated performance testing method for the Internet of Data. This approach separates test data from the testing scripts and implements monitoring of the test data. When new test data is added or existing data is modified, the test scripts automatically read the new data and rapidly update the test cases. To ensure the accuracy of the test results, the method involves sending no-load probe requests to determine the basic performance of the system being tested, using this as a baseline for comparison in actual tests. This paper first introduces the concept of data-driven testing, then integrates it into the performance testing for the Internet of Data, outlining the design and implementation of this data-driven performance testing approach. Finally, the method's reusability, simplicity, efficiency, and stability are demonstrated through the establishment of a testing platform and the analysis of specific test cases.