Use Caching

Learn about caching in Kubeflow Pipelines.

Kubeflow Pipelines support caching to eliminate redundant executions and improve the efficiency of your pipeline runs. This page provides an overview of caching in KFP and how to use it in your pipelines.

Overview

Caching in KFP is a feature that allows you to cache the results of a component execution and reuse them in subsequent runs. When caching is enabled for a component, KFP will reuse the component’s outputs if the component is executed again with the same inputs and parameters (and the output is still available).

Caching is particularly useful when you have components that take a long time to execute or when you have components that are executed multiple times with the same inputs and parameters.

If a task’s results are retrieved from cache, its representation in the UI will be marked with a green “arrow from cloud” icon.

How to use caching

Caching is enabled by default for all components in KFP. You can disable caching for a component by calling .set_caching_options(enable_caching=False) on a task object.

from kfp import dsl

@dsl.component
def say_hello(name: str) -> str:
    hello_text = f'Hello, {name}!'
    print(hello_text)
    return hello_text

@dsl.pipeline
def hello_pipeline(recipient: str = 'World!') -> str:
    hello_task = say_hello(name=recipient)
    hello_task.set_caching_options(False)
    return hello_task.output

You can also enable or disable caching for all components in a pipeline by setting the argument caching when submitting a pipeline for execution. This will override the caching settings for all components in the pipeline.

from kfp.client import Client

client = Client()
client.create_run_from_pipeline_func(
    hello_pipeline,
    enable_caching=True,  # overrides the above disabling of caching
)

Feedback

Was this page helpful?