cadence-workflow/cadence
Cadence is a distributed, scalable, durable, and highly available orchestration engine to execute asynchronous long-running business logic in a scalable and resilient way. observed · 2026-08-28
Health v2 · maintenance only
96/100
- Activity 99
- Release rhythm 90
- Longevity 100
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 28.5
- age_days: 3480
- days_rel: 64
- days_push: 7
- n_releases_24m: 15
Adoption not part of the score
9417 stars · 909 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded
Cadence is an open-source, CNCF-hosted durable execution and workflow orchestration engine originally developed at Uber. It lets developers write fault-tolerant, stateful, long-running workflows as plain code in Go, Java, or Python, with the backend handling retries, timers, event sourcing, and state recovery across failures.
Use cases
- orchestrate long-running multi-step business processes as code
- replace glue code over message queues, cron jobs, and state tables
- run recurring scheduled workflows with overlap policies and backfill
- implement saga-style compensation and automatic retries for unreliable downstream calls
- build event-driven applications that wait durably for signals or timers
- self-host a workflow engine in your own VPC without vendor lock-in
When to choose
- your processes span multiple steps with dependencies, retries, and long waits (minutes to years)
- you need durable state that survives worker and server restarts without managing queues or state machines
- you want an open-source, self-hosted alternative to Temporal or managed workflow services
- you need visibility into running workflows via a web UI, CLI, or API
- you're coordinating microservices where exactly-once semantics and partial-failure handling matter
When to avoid
- you only need simple fire-and-forget async task delivery - a message queue is simpler
- you want to store large business payloads or use it as a database (payloads should stay small)
- your team can't follow workflow determinism constraints (no direct time, network, or threading in workflow code)
- you need a lightweight in-process scheduler rather than a distributed orchestration platform
Facets
service · maturity stable
workflow-automation scheduling microservices message-queue developer-tools microservices backend developer-tools go jvm self-hosted cross-platform durable-execution workflow-engine orchestration event-sourcing saga-pattern cncf temporal-alternative fault-tolerance automation docker kubernetes
10 sources
- readme: https://github.com/cadence-workflow/cadence · fetched 2026-08-28 · cadfb3073dd8
- homepage: https://cadenceworkflow.io · fetched 2026-08-29 · ce2487ee5c98
- site_page: https://cadenceworkflow.io/docs/get-started · fetched 2026-08-29 · 0e5303b5c53f
- site_page: https://cadenceworkflow.io/docs/concepts · fetched 2026-08-29 · 4e4ac68da895
- site_page: https://cadenceworkflow.io/docs/concepts/schedules · fetched 2026-08-29 · e2f4d03afae4
- site_page: https://cadenceworkflow.io/docs/concepts/workflow-engine · fetched 2026-08-29 · aa929b454d92
- site_page: https://cadenceworkflow.io/docs/concepts/open-source-workflow-engine · fetched 2026-08-29 · 44c8c35d60b1
- site_page: https://cadenceworkflow.io/faq/best-practices · fetched 2026-08-29 · 121c25a690ea
- site_page: https://cadenceworkflow.io/faq/cadence-faq · fetched 2026-08-29 · d3f5f13e2ee8
- site_page: https://cadenceworkflow.io/faq/migrate-from-temporal · fetched 2026-08-29 · 05c3ab7732ef
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| cadence-workflow/cadence | main | 96 |
For agents
markdown · JSON · MCP: product_card(name="cadence-workflow/cadence")
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem