Continuous integration
What is Argo Workflows? An introduction to Argo Project's container-native workflow engine
Argo Workflows is an open source, container-native workflow engine for orchestrating parallel jobs on Kubernetes. It is implemented as a Kubernetes Custom Resource Definition (CRD), so workflows are defined and run as Kubernetes resources. Each step in a workflow is a container.
- Tool
- Argo Workflows
- Made by
- Argo Project
- Official site
- argoproj.github.io/workflows
- Category
- Continuous integration
Who makes Argo Workflows
Argo Workflows is developed by the Argo Project. The official website is linked below this article.
Official website: argoproj.github.io/workflows
What Argo Workflows is for
Argo Workflows orchestrates jobs on Kubernetes. A workflow describes a set of container-based steps and how they relate to each other, and the engine runs them on the cluster. It is built for highly parallel jobs, where many containers run at the same time.
It can be used for compute-intensive work such as machine learning or data processing. It can also run CI/CD pipelines natively on Kubernetes, which is the reason it appears in the continuous integration category on Deployelm.
Who Argo Workflows is for
The vendor does not state a target audience. From its design, it suits teams that already run containers on Kubernetes and want to express multi-step jobs as Kubernetes resources.
It is designed for containers rather than legacy VM and server-based environments. Teams whose build and processing jobs still depend on that older model are not the intended fit.
Notable capabilities
The capabilities below come from the project's own description of the engine. They centre on how workflows are defined, how steps are structured, and where the engine runs.
- Container-native steps: each step in a workflow is defined as a container.
- Kubernetes CRD implementation: workflows are managed as custom resources within a cluster.
- Task sequences: a workflow can be modeled as a sequence of tasks.
- Dependency graphs: a workflow can instead be modeled as a DAG, where steps declare their dependencies.
- Both DAG-based and step-based workflows are supported, so the structure can match the job.
- Parallel orchestration: the engine is built to run highly parallel jobs on Kubernetes.
- Cloud agnostic: it runs on any Kubernetes cluster.
- Native CI/CD pipelines: pipelines can run directly on Kubernetes.
How Argo Workflows fits continuous integration
In continuous integration, Argo Workflows covers the job-execution layer. Build, test and processing steps are written as containers and ordered as sequences or DAGs. The engine runs them on a Kubernetes cluster.
Within Deployelm's wider niche of CI/CD and deployment platforms, it is a general workflow engine that can run CI/CD pipelines, rather than a tool limited to one pipeline style. Because it runs on any Kubernetes cluster, it does not tie a team to a particular cloud.
Frequently asked questions
What is Argo Workflows used for?
It orchestrates parallel jobs on Kubernetes. Typical uses include compute-intensive jobs such as machine learning or data processing, and CI/CD pipelines run natively on Kubernetes.
Who develops Argo Workflows?
Argo Workflows is developed by the Argo Project. It is an open source project.
Does Argo Workflows support DAGs?
Yes. Workflows can be modeled as sequences of tasks or as dependency graphs (DAGs), and both DAG-based and step-based workflows are supported.
Does Argo Workflows run outside Kubernetes?
It is a Kubernetes-based engine implemented as a CRD, and it runs on any Kubernetes cluster. It is designed for containers rather than legacy VM and server-based environments.
Deployelm is independent and is not affiliated with Argo Project. Product names belong to their owners. Features change; confirm details on the official website.