This thesis addresses the challenges of manual performance testing for the
Symbiot platform, which is used at Iskraemeco to manage smart metering
systems. The existing process is time-consuming, labor-intensive, involves
numerous repetitive steps, and is highly prone to human error. Due to large
volumes of data, multiple virtual environments, and complex communication
processes, manual test execution is difficult to replicate, hindering reliable
comparison of results across different runs.
In this work, I developed a comprehensive approach to automating the key
steps of performance testing. The solution includes the graphical application
Symbiot Installer for automated platform deployment on virtual machines,
the Mass Importer tool for input data generation, and an SQL script to
check and resolve index fragmentation. Several components are integrated
into pipeline agents, enabling fully automated, repeatable, and time-efficient
execution of the setup procedures.
The results confirm that automation significantly reduces test preparation
time, minimizes the risk of errors, and simplifies performance testing for less
experienced employees. This solution represents a major step toward higher
efficiency, transparency, and reliability in the testing process.
|