Logo
Deploy Now

SaaS Alternative

AWS Step Functions

Stars

22,431

Forks

1,827

Watchers

120

Developer links

Temporal

Powering mission-critical infrastructure at OpenAI, Cursor, Replit, Lovable, Retool, and Snap with over 22,000 GitHub stars, Temporal is the durable execution platform that originated from Uber's Cadence project — built by the creators of AWS SQS, AWS SWF, and Azure Durable Functions with nine years of production-proven reliability. The workflow-as-code model lets developers write business logic in Go, Java, Python, TypeScript, .NET, PHP, or Ruby using native SDKs, while the Temporal Server automatically persists state at every step, replays from failures, retries activities with configurable backoff policies, and manages task queues without developers writing reconciliation logic. Workflows support signals for external event injection, timers for scheduled delays, child workflows for decomposition, and queries for real-time state inspection — all backed by deterministic replay over an event-sourced history that guarantees exactly-once semantics. The Web UI provides visual workflow execution inspection with event timelines, pending activity monitoring, namespace management, and worker health dashboards. Persistence supports PostgreSQL, MySQL, or Apache Cassandra for horizontal scalability, with Elasticsearch or OpenSearch for advanced workflow visibility queries. Multi-cluster replication enables global failover across data centers. The self-hosted stack deploys via Docker Compose with the auto-setup image, PostgreSQL, Web UI, and admin tools — operational within 30 minutes. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Temporal
Temporal
Temporal
Temporal
Temporal

Benefits

  • Automatic Failure Recovery and Retries
  • Workflows automatically survive process crashes, server restarts, and network failures. The Temporal Server persists state at every step and replays from the last checkpoint — zero data loss guaranteed.
  • Native Multi Language SDK Support
  • Write workflows and activities in Go, Java, Python, TypeScript, .NET, PHP, or Ruby using idiomatic native SDKs. No DSL or YAML configuration required — pure code with full IDE support.
  • Production Proven at Global Scale
  • Nine years of production use originating from Uber's Cadence. Powers infrastructure at OpenAI, Cursor, Replit, Snap, and thousands of companies processing billions of workflow executions daily.
  • Permissive MIT Open Source License
  • MIT license allows unrestricted commercial use, modification, and redistribution. No copyleft obligations — deploy in proprietary systems, embed in commercial products, or fork without legal constraints.

Features

  • Durable Workflow Execution
  • Event-sourced workflow history with deterministic replay guarantees exactly-once execution semantics. Workflows survive crashes, restarts, and infrastructure failures automatically.
  • Visual Web Dashboard
  • Browser-based Web UI displays workflow execution timelines, event histories, pending activities, worker status, and namespace management with real-time state inspection.
  • Configurable Retry Policies
  • Per-activity retry policies with exponential backoff, maximum attempts, non-retryable error types, and heartbeat timeouts for fine-grained failure handling control.
  • Flexible Persistence Backends
  • PostgreSQL, MySQL, or Apache Cassandra for workflow state persistence. Elasticsearch or OpenSearch integration enables advanced visibility queries across workflow executions.
  • Signals and Queries
  • External signals inject events into running workflows for human-in-the-loop approvals. Synchronous queries inspect workflow state without affecting execution history.
  • Docker Compose Deployment
  • Official auto-setup Docker image provisions database schemas automatically. Docker Compose stack includes Temporal Server, PostgreSQL, Web UI, and admin tools.