Hera

Hera makes Python code easy to orchestrate on Argo Workflows through native Python integrations. It lets you construct and submit your Workflows entirely in Python.

See the Quick Start guide to start using Hera to orchestrate your Argo Workflows!

The Argo was constructed by the shipwright Argus,
and its crew were specially protected by the goddess Hera.

PyPI stats

Repo information

Explore the code

Hera at a glance

Steps diamond

from hera.workflows import Steps, Workflow, script


@script()
def echo(message: str):
    print(message)


with Workflow(
    generate_name="single-script-",
    entrypoint="steps",
) as w:
    with Steps(name="steps") as s:
        echo(name="A", arguments={"message": "I'm a step"})
        with s.parallel():
            echo(name="B", arguments={"message": "We're steps"})
            echo(name="C", arguments={"message": "in parallel!"})
        echo(name="D", arguments={"message": "I'm another step!"})

w.create()

DAG diamond

from hera.workflows import DAG, Workflow, script


@script()
def echo(message: str):
    print(message)


with Workflow(
    generate_name="dag-diamond-",
    entrypoint="diamond",
) as w:
    with DAG(name="diamond"):
        A = echo(name="A", arguments={"message": "A"})
        B = echo(name="B", arguments={"message": "B"})
        C = echo(name="C", arguments={"message": "C"})
        D = echo(name="D", arguments={"message": "D"})
        A >> [B, C] >> D

w.create()

See the examples for a collection of Argo workflow construction and submission via Hera!

Requirements

Hera requires an Argo server to be deployed to a Kubernetes cluster. Currently, Hera assumes that the Argo server sits behind an authentication layer that can authenticate workflow submission requests by using the Bearer token on the request. To learn how to deploy Argo to your own Kubernetes cluster you can follow the Argo Workflows guide!

Another option for workflow submission without the authentication layer is using port forwarding to your Argo server deployment and submitting workflows to localhost:2746 (2746 is the default, but you are free to change it). Please refer to the documentation of Argo Workflows to see the command for port forward!

> Note Since the deprecation of tokens being automatically created for ServiceAccounts and Argo using Bearer tokens > in place, it is necessary to use --auth=server and/or --auth=client when setting up Argo Workflows on Kubernetes > v1.24+ in order for hera to communicate to the Argo Server.

Authenticating in Hera

There are a few ways to authenticate in Hera - read more in the authentication walk through - for now, with the argo cli tool installed, this example will get you up and running:

from hera.workflows import Workflow, Container
from hera.shared import global_config
from hera.auth import ArgoCLITokenGenerator

global_config.host = "http://localhost:2746"
global_config.token = ArgoCLITokenGenerator

with Workflow(generate_name="local-test-", entrypoint="c") as w:
    Container(name="c", image="docker/whalesay", command=["cowsay", "hello"])

w.create()

Installation

> Note Hera went through a name change - from hera-workflows to hera. This is reflected in the published Python > package. If you'd like to install versions prior to 5.0.0, you should do pip install hera-workflows<5. Hera > currently publishes releases to both hera and hera-workflows for backwards compatibility purposes.

Optional dependencies

yaml

• Install via hera[yaml]
• PyYAML is required for the yaml output format, which is accessible via hera.workflows.Workflow.to_yaml(*args, **kwargs). This enables GitOps practices and easier debugging.

cli

• Install via hera[cli]. The [cli] option installs the extra dependency Cappa required for the CLI
• The CLI aims to enable GitOps practices, easier debugging, and a more seamless experience with Argo Workflows.
• The CLI is an experimental feature and subject to change! At the moment it only supports generating YAML files from workflows via hera generate yaml. See hera generate yaml --help for more information.

experimental

• Install via hera[experimental]. The [experimental] option adds dependencies required for experimental features that have not yet graduated into stable features.

Presentations

• KubeCon/ArgoCon NA 2024 - Data Science Workflows Made Easy: Python-Powered Argo for Your Organization
• KubeCon/ArgoCon EU 2024 - Orchestrating Python Functions Natively in Argo Using Hera
• CNCF TAG App-Delivery @ KubeCon NA 2023 - Automating the Deployment of Data Workloads to Kubernetes with ArgoCD, Argo Workflows, and Hera
• KubeCon/ArgoCon NA 2023 - How to Train an LLM with Argo Workflows and Hera
  • Featured code
• KubeCon/ArgoCon EU 2023 - Scaling gene therapy with Argo Workflows and Hera
• DoKC Town Hall #2 - Unsticking ourselves from Glue - Migrating PayIt's Data Pipelines to Argo Workflows and Hera
• Argo Workflows and Events Community Meeting 15 June 2022 - Hera project update
• Argo Workflows and Events Community Meeting 20 Oct 2021 - Hera introductory presentation

Blogs

• How To Get the Most out of Hera for Data Science
• Data Validation with Great Expectations and Argo Workflows
• Hera introduction and motivation
• Dyno is scaling gene therapy research with cloud-native tools like Argo Workflows and Hera

Contributing

See the contributing guide!

Hera Emeritus Maintainers

These emeritus maintainers dedicated a part of their career to Hera and reviewed code, triaged bugs and pushed the project forward over a substantial period of time. Their contribution is greatly appreciated:

• Flaviu Vadan

License

Hera is licensed under Apache 2.0. See License for details.