What are we using?

Stencil utilizes the k6 benchmarking tool to conduct load tests and evaluate application performance. This page provides an overview of the key components involved in benchmarking with Stencil, with a specific focus on the boilerplate setup used for load testing.

k6 Configuration

The core configuration for our benchmarking setup is defined in the k6config.js file. This file includes parameters essential for shaping the testing environment:

• Virtual Users (VUs): Specifies the number of virtual users simulated during the test. In our configuration, we set this to 5 (vus: 5).

• Duration: Sets the duration of the test. Here, we've configured it to run for 30 seconds (duration: '30s').

• Thresholds: Defines performance thresholds to assess the success of the test. For example, we've set a criterion where the 95th percentile response time should be below 500 milliseconds (thresholds: { http_req_duration: ['p(95)<500'] }).

• Tags: Provides metadata to categorize and label the test, aiding in result analysis. In our case, we've tagged the test with the environment as 'staging' (tags: { environment: 'staging' }).

Load Test Boilerplate

To facilitate the load testing process, Stencil employs a versatile boilerplate encapsulated in the common.js file. This boilerplate contains essential functions for conducting HTTP requests, checking responses and much more.

The subsequent documentation will provide further details on the configuration options in k6config.js and guide users on effectively utilizing the boilerplate for their specific performance testing needs with Stencil.