After a two-year break, on 18 May 2022, the 4th Data Science Day Jena took place. The Friedrich-Schiller University is hosting this mini-conference format annually.
Xceptance presented this year a closer look at the data that is collected and processed during load and performance testing. René Schwietzke, Managing Director of Xceptance, talked about the challenges to capture the right data as well as translate the collected data into meaningful results.
A load test simulates millions of user interactions with a website and therefore is capturing huge amounts of data points. These have to be transformed into a few numbers to make the result of the test easy to communicate but still preserve important details. The talk started with typical business requirements and expectations of the target groups of a load test. It showed the data XLT captures and the dimensions which later drive the data reduction. A few example data series demonstrated the challenges behind the data reduction and what numbers are finally used to satisfy the requirements.
An example load test result illustrated the talk with real data. That example test run created about 17,500 data rows per second which contain about 293,000 data points. The entire test result consists of 3.2 billion data points. This massive data set is turned into a consumable report by XLT in less than six minutes.
For everyone with an interest in data science, this presentation also offers ideas for research in regard to unsolved data challenges. There might be even some Master’s and Bachelor’s theses topics waiting for you.
You can find a recording of the presentation below (courtesy of Thüringer Universitäts- und Landesbibliothek Jena).
This is the accompanying slide deck. It is a Reveal.js-based. You can navigate with the spacebar and the arrow keys.