Apache DolphinScheduler
dolphinscheduler.apache.org
- Free plan
- Open source
- Linux
- Web
- API
- Self-hosted
What Apache DolphinScheduler is
Apache DolphinScheduler is a free, open-source platform for designing and orchestrating data workflows. Teams can build visual DAG workflows with drag-and-drop tools or define them with Python, YAML, or the Open API. The project lists more than 30 built-in task types, including Shell, SQL, Spark, Flink, Hive, Kubernetes, and HTTP. Workflows can depend on other projects or workflows, pause and resume, and run batch backfills serially or in parallel. Its decentralized multi-master and multi-worker design is described as increasing scheduling capacity with cluster size. Documented deployment options include standalone, cluster, and Kubernetes. Standalone is intended for quick evaluation rather than production, is recommended for fewer than 20 workflows, and uses a default in-memory H2 database that is cleared when the service stops or restarts. Security documentation lists password, LDAP, Casdoor SSO, and OAuth2 authentication, plus administrator-managed access controls. The security model assumes a secure, trustworthy network and does not address internal network cross-infiltration. Documentation, quick starts, and a community mailing list are available.
Who it suits
It may suit teams that need self-hosted workflow orchestration with visual design or workflow-as-code options. Teams evaluating standalone deployment should account for its workflow guidance and in-memory database behavior.
Good to know, and what to watch
- Free, open-source workflow orchestration.
- Supports visual workflows and Python, YAML, or Open API.
- Lists more than 30 built-in task types.
- Provides standalone, cluster, and Kubernetes deployment options.
- Supports several authentication methods.
- Standalone is recommended for fewer than 20 workflows.
- Standalone's default H2 database clears on stop or restart.
- Security model assumes a secure, trustworthy network.
Plans
Every plan and priceApache DolphinScheduler
Open source
Open source
The verdict
DolphinScheduler offers visual and code-defined workflow creation, a broad set of task types, and multiple deployment options. For production planning, note that standalone is positioned for evaluation and review the documented security boundary.
What the maker publishes
- Access controls
- The security guide describes administrator-managed users, tenants, queues, tokens, worker groups, and authorization for resources and data sources. source
- API
- The product supports workflow coding and integration through its Open API, and the maker provides a Python API for defining workflows as code. source
- Authentication
- The security model documentation lists password, LDAP, Casdoor SSO, and OAuth2 authorization as authentication methods. source
- Authentication
- Supported login methods are password authentication, LDAP, Casdoor SSO and OAuth2 authorization. source
- Authorization
- The security center controls queues, tenants, users, alarm groups, worker groups, tokens, projects, resources and data-source permissions. source
- Backfill
- Complement runs support executing a workflow over a selected date range in serial or parallel mode. source
- Community support
- The project directs users and developers to GitHub issues plus users and dev mailing lists for support and discussion. source
- Concurrency
- The official site claims support for millions of data tasks with high concurrency, throughput and low latency. source
- Dependency management
- Workflows can express dependencies across projects and workflows, including dependent and sub-process task nodes. source
- Deployment
- The maker documents standalone, cluster, and Kubernetes deployment; standalone is intended for quick evaluation rather than production. source
- Documentation depth
- Documentation covers installation, architecture, configuration, security, API usage and individual Python task classes with code and YAML examples. source
- Execution controls
- Users can kill, pause and resume task or workflow execution. source
- Extensibility
- The platform supports custom task types and plugin families for tasks, data sources, storage, alerts and scheduling. source
- Founded
- 2017 source
- High availability
- The architecture uses decentralized multi-master and multi-worker services so masters and workers can be scaled independently. source
- Independent project footprint
- The public repository page shows about 14.5 thousand stars and 5.1 thousand forks. source
- Latest release
- GitHub lists version 3.4.3 as the latest release, published on September 6, 2026. source
- License
- The source repositories identify the software as licensed under Apache License 2.0. source
- Limit: code review workflow
- An independent review says UI-authored DAGs are easy to build but harder to review in pull requests than code-defined pipelines. source
- Limit: failure recovery incident
- A published incident report reproduced a master-crash scenario where restart caused previously unscheduled tasks to run together and potentially exhaust CPU and memory. source
- Limit: operational complexity
- Independent catalog coverage rates operational complexity as high because production deployments require multiple services and infrastructure components. source
- Limit: scaling and maintenance
- A field report describes painful configuration sprawl, many manual operations, and significant infrastructure management when moving to Kubernetes. source
- Maker founded
- The Apache Software Foundation says it was formed in 1999. source
- Maker headquarters
- The Apache Software Foundation lists its mailing address in Wilmington, Delaware, United States. source
Checked against the maker's pages on 15 Oct 2026.