Skip to main content

​
Benchmark performance impact

The performance impact denotes an improvement or regression in performance of a benchmark. It is calculated by comparing the benchmark time measurement of the head commit with the time measurement of the base commit. impact=speed−baseSpeedbaseSpeedimpact = \frac{speed - baseSpeed}{baseSpeed} A negative performance metric means that the benchmark is slower than the previous commit. The closer its value is to -1, the slower it is. −1<impact<0-1 \lt impact \lt 0 A positive performance metric means that the benchmark is faster than the previous commit. Its value can go up to +∞+\infty to denote massive speed improvements. 0<impact<+∞0 \lt impact \lt +\infty Naturally, when the benchmark is as fast as the previous commit, the performance metric is 00.

​
Regression threshold

On the settings page of a project, you can set a threshold for a regression to be considered a regression. By default, this value is set to 10% (which is equivalent to 0.1). The value can be set from 0% to 50% by an admin of the project. More information about setting the threshold can be found in the customization documentation. Individual benchmarks can have their own custom regression thresholds that override the project-level setting. See the customization documentation for details on configuring per-benchmark thresholds.

​
Commit performance impact

To get the overall performance impact of a commit, we aggregate all the benchmark impacts. n being the number of benchmarks.

​
Regression threshold exceeded

If there is a regression above the threshold, the overall commit impact will be the biggest regression impact. commitImpact=min⁡0≤i<nimpact[i]commitImpact = \min_{0\leq i\lt n} impact[i]
For example, with impacts: [0.1, 0, -0.3] and a threshold of 0.25, the overall commit impact will be -0.3.

​
Improvement threshold exceeded

If there is an improvement above the threshold, the overall commit impact will be the maximum improvement impact. commitImpact=max⁡0≤i<nimpact[i]commitImpact = \max_{0\leq i\lt n} impact[i]
For example, with impacts: [0.1, 0.3, -0.2] and a threshold of 0.25, the overall commit impact will be 0.3.

​
No threshold exceeded

Finally, in the remaining cases, a geometric mean is calculated from all the benchmark performance impacts. commitImpact=(∏i=0n−11+impact[i])1n−1commitImpact = \left(\prod_{i=0}^{n-1} 1+impact[i]\right)^{\frac{1}{n}} - 1
For example, with impacts: [0.1, 0.3, -0.2] and a threshold of 0.5, the overall commit impact will be approximately 0.0459.
A geometric mean will give more relevant results than an arithmetic mean for this kind of measure as it will be less sensitive to outliers.

​
Baseline report selection

To create a performance impact, we need to compare the execution speed of the benchmarks against a baseline of those benchmarks’ execution speed. Depending on the context of the run, the baseline report can be different.

​
Pull Request

When triggering a CodSpeed run on a pull request between a head branch and a base branch, the baseline report will be the report of the latest commit of the base branch with a CodSpeed run.
Checked-out commit of a Pull Request in GitHub ActionsBy default, when using the action/checkout in GitHub Actions, the checked-out commit of a pull request will be the merge commit of the pull request. This means that GitHub will create a merge commit M between the base branch and the head branch, and use it as the checked-out commit.
Pull Request Base In this case, the report will be between the HEAD commit C of the base branch and the merge commit M. The report will thus only include the changes of commits B1 and B2.
Every time a new commit is pushed on the feat-branch, the baseline report will be updated to the latest commit of the base branch with a CodSpeed run.
This pull request base selection algorithm will only work on pull_request event for GitHub Actions and merge_request_event for GitLab CI. If you run CodSpeed on a different event, the branch base selection algorithm defined below will be used instead.

​
Branch

When triggering a CodSpeed run following a push on a branch, the baseline report will be the report of the closest commit of the branch. Branch Base
In this example, a report already exists for the commit B of the main branch. A new commit C is pushed on the main branch. The baseline report for C will be the report of the commit B.

​
Performance impact Gauge

The performance impact gauge is a visual representation of the performance impact, displayed in multiple places of the CodSpeed UI. Some examples of the gauge with their corresponding impact values:
Gauge -0.75
Gauge -0.2
Gauge 0
Gauge 0.3
Gauge 1.5
To make it easier to spot regressions and improvements, the mapping between the actual performance impact value and the gauge is not linear.

​
Next Steps

Set Up Automated Performance Checks

Configure GitHub branch protection to automatically block performance regressions

Debug Performance Issues

Use flame graphs and profiling data to identify and fix bottlenecks

Customize Regression Thresholds

Fine-tune regression sensitivity for your project’s requirements

Create Comprehensive Benchmarks

Build a robust benchmark suite to catch performance issues early