The Temporal Server is a micro-service orchestration platform. It can be thought of as a layer in your technology stack that exists between your compute servers and your executable source code. It is completely open source and can be run in a variety of cloud or on-premise environments. A single Server instance can be used for many different use cases simultaneously, or be tied to a single application.
- Read the system architecture page to get a view of the server's topology.
- Read the Server versions and dependencies page to see which dependencies are supported with the version of the Server you are using.
How and where you run the Server depends entirely on your use-case, intended environment, and hosting infrastructure.
Choose from the list below to get started:
- Namespaces: Create basic logical separations within a Temporal cluster, or use Global Namespaces to enable failover across multi-cluster instances.
- Security: Learn how to keep your self-hosted instance secure with encryption and pluggable auth protocols.
- Experimental features:
- Archival: Want to back up your Workflow event history? You can with a cloud provider of your choice.
- Multi-cluster Replication: Increase availability by asynchronously replicating workflows from active cluster to other passive clusters.
- Enhanced Workflow search: Integrating Elasticsearch enables the use of customizable search attributes and complex search queries.