When using cloud-native builds, you are essentially distributing build tasks to various build nodes for execution. These nodes perform build tasks using a specified Docker image as the build environment.

Configuration Method

In the pipeline configuration, you can specify the pipeline.runner.tags property to select different build nodes. By specifying the pipeline.runner.cpus property, you can configure the maximum number of CPU cores to be used. Cloud-native builds calculate the core-hours used by the pipeline by multiplying the actual allocated cores by the pipeline's duration.

The official available build node tags and configurable cpus are as follows:

1. cnb:arch:amd64 represents a CPU server build node with the amd64 architecture.

   • The cpus can be configured in the range of 1 to 64, with a default value of 8.

2. cnb:arch:arm64:v8 represents a CPU server build node with the arm64/v8 architecture.

   • The cpus can be configured in the range of 1 to 16, with a default value of 8.

3. cnb:arch:amd64:gpu represents a GPU server build node with the amd64 architecture.

   • The cpus is fixed at 32.
   • The GPU memory has a maximum of 96GB, operating in shared mode.

4. cnb:arch:amd64:gpu:L20 represents a GPU server build node with the amd64 architecture.

   • The cpus is fixed at 16.
   • The GPU memory has a maximum of 48GB, operating in shared mode.

Example:

main:
  push:
    # Specify execution on an amd64 architecture build node
    - runner:
        tags: cnb:arch:amd64
      stages:
        - name: uname
          script: uname -a
    # Specify execution on an arm64/v8 architecture build node
    - runner:
        tags: cnb:arch:arm64:v8
      stages:
        - name: uname
          script: uname -a

# Start a workspace for remote development with GPU access
$:
  vscode:
    - runner:
        tags: cnb:arch:amd64:gpu
      services:
        - vscode