Temporal
By Temporal Technologies
Temporal is an open-source workflow orchestration platform for writing durable, fault-tolerant business logic as ordinary code, guaranteeing that long-running processes complete correctly even through crashes, restarts, and infrastructure…
Definition
Temporal is an open-source workflow orchestration platform for writing durable, fault-tolerant business logic as ordinary code, guaranteeing that long-running processes complete correctly even through crashes, restarts, and infrastructure failures.
Overview
Temporal grew out of a workflow engine originally built at Uber and was later reworked into an independent open-source project and company, Temporal Technologies. Its core idea is 'durable execution': developers write workflows as regular functions in languages like Go, Java, TypeScript, Python, or .NET, and the Temporal service transparently persists each step's progress, so that if a worker process crashes mid-execution, the workflow resumes exactly where it left off rather than restarting or losing state. A Temporal application is split into workflows, which define the ordered sequence of steps, and activities, which perform the actual side-effecting work such as calling an external API, writing to a database, or sending a notification. The Temporal server tracks event history for every workflow execution, enabling automatic retries, timeouts, signals, and long-running processes that can span days or months without the developer manually building retry logic or state machines. This makes Temporal well suited to problems that traditional message queues or cron jobs handle poorly — multi-step order fulfillment, payment processing sagas, infrastructure provisioning pipelines, and other workflows where correctness and visibility over time matter more than raw throughput. It is often compared to simpler event-driven systems like Kafka, though Temporal focuses on orchestrating stateful business processes rather than just streaming events between services, and is commonly deployed alongside Docker and Kubernetes for the worker and server components. Teams building event-driven backends may also look at a course like Apache Kafka & Messaging to understand how the two approaches complement each other.
Key Features
- Durable execution model that automatically persists workflow state and progress
- Automatic retries, timeouts, and error handling defined in code rather than infrastructure config
- Workflows written in ordinary code (Go, Java, TypeScript, Python, .NET) rather than YAML or DSLs
- Built-in support for long-running processes spanning days, weeks, or months
- Signals and queries for interacting with running workflows from external systems
- Full visibility and replay of workflow execution history for debugging
- Horizontally scalable worker model decoupled from the orchestrating Temporal service
- Available as self-hosted open source or a managed Temporal Cloud offering