Skip to main content

Run a stage or step multiple times using a matrix

A matrix enables you to run the same Stage or Step multiple times with different parameters. Matrix strategies eliminate the need to copy the same stage or step with different inputs for each variation. Matrix strategies also make your Pipelines more readable, clean, and easy to maintain. You can easily define matrix strategies to support workflows such as:

  • A Run Step that load-tests a UI feature in 4 different browsers and 3 different platforms.
  • A Build Stage that builds artifacts for 10 different JDK versions.
  • A Deploy Stage that deploys 3 different services to 4 different environments.

Before you begin

You can apply matrix strategies to both CI and CD workflows. This topic assumes that you are familiar with the following:

Important Notes

  • There is no limit on the number of dimensions you can include in a matrix or the number of looping strategies you define in a Pipeline.
  • You should avoid complex looping scenarios unless you clearly understand the resources that your scenario will require. See Best Practices for Looping Strategies.

Add a Matrix Strategy to a Stage or Step

  1. In the Pipeline Studio, go to the Advanced tab of the Stage or Step where you want to apply the Looping strategy.
  2. Under Looping Strategies, select Matrix. You can also use a Loop strategy to iterate through a simple list. See Looping Strategies Overview: Matrix, Repeat, and Parallelism.
  3. Enter the YAML definition for your strategy as illustrated in the following examples.

CI Example: Run an App in [browser] on [os]

Suppose you have a Pipeline that builds an app in Go. You want to test the app on three different platforms and three different browsers. In the Stage where you test the app, you can define a matrix like this. 

browser: [chrome, safari, firefox ]
os: [ macos, windows, linux ]
maxConcurrency: 3

In this example, os and browser are user-defined tags. You can specify any tag in a matrix:  jdk, platform, node-version, and so on.

You can use the maxConcurrency keyword to run multiple jobs concurrently. In this case, the effective matrix has 9 combinations. With maxConcurrency set to 3, the Pipeline runs 3 Build Stages concurrently and load-balances the combinations between them.

Excluding Combinations from a Matrix

You can use the exclude keyword to exclude certain combinations from being run. Suppose you don’t want to run the app in Safari on Windows. In this case, you can exclude this combination from the run matrix: 

browser: [chrome, safari, firefox ]
os: [ macos, windows, linux ]
- browser: safari
os: windows
maxConcurrency: 3 # test the app across 3 Stages running concurrently

You can also exclude all combinations based on just one value. If you want to exclude all combinations with MacOS, for example, you can do the following: 

browser: [chrome, safari, firefox ]
os: [ macos, windows, linux ]
- os: macos
maxConcurrency: 4 # test the app across 4 Stages running concurrently

CD Matrix Example: Deploy [service] to [environment]

You can easily set up a Deploy Stage to deploy multiple services to multiple environments by defining a matrix like this:

service: [ svc1, svc2, svc3 ]
environment: [ env1, env2 ]
- service: svc1
environment: env1
maxConcurrency: 2

Simple List Example: Build App [items]

You can also use the repeat and items keywords to iterate through a simple list. Suppose you want to build a Java app for multiple JDKs. Under Looping Strategy, select For Loop and enter the following:

items: [ "18", "17", "16", "15", "14", "13", "12", "11", "10", "9" ]
maxConcurrency: 2

Note that this example is simply an alternative way of defining a one-dimensional matrix with the items keyword as the key. You can define the same basic strategy like this:

 jdk: [ "18", "17", "16", "15", "14", "13", "12", "11", "10", "9" ]
maxConcurrency: 2

Using Matrix Variables in Your Pipeline

You can reference matrix values in your Stages and Steps using <+matrix.tag>. Here are some examples.

Given the CI example above, you can enter the following in a Run Step to output the current run:

echo “Testing app in <+matrix.browser> on <+matrix.os>”

Suppose you have a matrix in a Stage and another in a member Step. The Stage matrix has tags browser and os. The Step matrix has tags browser and os. You can reference both sets of tags from within the Step like this:

echo "stage values (parent):"  
echo "Current service for stage: <+stage.matrix.browser>"
echo "Current os for stage: <+stage.matrix.os>"
echo "step values (local):"
echo "Current browser for step: <+matrix.browser>"
echo "Current os for step: <+matrix.os>"

Given the CD example above, you can go to the Service tab of the Deploy Stage and specify the service using <+matrix.service>.

The following variables are also supported:

  • <+strategy.iteration> — The current iteration.
  • <+strategy.iterations> — The total number of iterations.
  • <+repeat.item> — The value of the item when iterating through a list using the repeat and items keywords.

YAML Pipeline Example

The following example illustrates how you can define matrix strategies in a pipeline.


name: matrix-example-2
identifier: matrixexample2
projectIdentifier: myproject
orgIdentifier: myorg
tags: {}
- stage:
name: echoMatrixSettings
identifier: echoMatrixSettings
description: ""
type: Custom
- step:
type: ShellScript
name: echo
identifier: echo
shell: Bash
onDelegate: true
type: Inline
script: |-
echo "iteration index = <+strategy.iteration>"
echo "total iterations = <+strategy.iterations>"
echo "stage values (parent):"
echo "Current version for stage: <+stage.matrix.service>"
echo "Current environment for stage: <+stage.matrix.environment>"
echo "step values (local):"
echo "Current item (version): <+repeat.item>"
environmentVariables: []
outputVariables: []
executionTarget: {}
timeout: 10m
failureStrategies: []
- "18"
- "17"
- "16"
- "15"
- "14"
- "13"
- "12"
- "11"
- "10"
- "9"
maxConcurrency: 2
tags: {}
- svc1
- svc2
- svc3
- env1
- env2
- service: svc1
environment: env1
maxConcurrency: 2

See also