Executing pipeline operations
Apart from creating and deleting pipelines, you likely will need to update your pipelines with new code changes. If changes are made to the pipeline specification itself, such as increasing the number of Pachyderm workers, input repository, glob pattern, or similar, you only need to do it in the YAML or JSON file and update the version of your pipeline spec. However, if the changes are in your code and your code is in your Docker image, you need to rebuild the Docker image. Let's go through each of these use cases.
Updating your pipeline specification
The pipeline specification enables you to control various Pachyderm parameters, such as controlling from which repository your pipeline consumes data, how many workers are started, and how many resources are available to your pipeline. You can also specify your code in the pipeline itself through the stdin field. Such a pipeline can use a basic Docker image that you won't have to ever update...
