# Oban Pro

Oban Pro is a licensed add-on for the Oban job orchestration framework. It provides extensions
that expand what Oban is capable of while making complex workflows possible.

Get started with [installation](installation.md), learn how to transition from OSS to Pro
with [adoption](adoption.md), then explore the features below.

## Operations

Configure how Pro runs, without changes to your workers.

### Drop-In Optimizations

Install Pro and your application automatically benefits from bulk inserts, bulk acking, and
accurate orphan rescue—no code changes required. [Learn more →](adoption.md)

### Multi-Process Execution

Bypass the GIL and utilize multiple CPU cores for CPU-intensive workloads. Just switch from
`oban start` to `obanpro start`. [Learn more →](multi_process.md)

### Erlang Notifier

Replace Postgres pub/sub with direct communication between processes. Works with transaction
poolers, without a payload size limit, and keeps notifications out of the database. It uses the
Distributed Erlang protocol to interop with Oban in Elixir, but doesn't require Erlang or Elixir.
[Learn more →](erlang_notifier.md)

### Smart Concurrency

Control job execution with global concurrency limits across all nodes, rate limiting
(e.g., 60 jobs/minute), and partitioned queues that apply limits per worker, tenant, or
any argument. [Learn more →](smart_concurrency.md)

### Pro Pruner

Replace the standard pruner with fine-grained retention rules per queue, worker, or state. Prune on
a cron schedule, manage rules at runtime, and archive jobs rather than deleting them.
[Learn more →](pruner.md)

## Jobs

Extend what an individual job can do.

### Unique Jobs

Prevent enqueueing duplicate jobs based on worker, args, queue, or specific keys. Configure
time windows and detect conflicts. [Learn more →](unique_jobs.md)

### Relay

Dispatch jobs and wait for their results with synchronous-style semantics. Perfect for
request/response workflows and service coordination. [Learn more →](relay.md)

### Awaiting Signals

Pause a job mid-execution to wait for an external decision and resume when a signal arrives.
Perfect for human approval flows, third-party callbacks, and durable state machines.
[Learn more →](signals.md)

## Composition

Coordinate many jobs together.

### Workflows

Compose jobs together for durable workflows with fault-tolerant execution patterns. Supports
sub-workflows, automatic result passing, and runtime grafting for dynamic pipelines. [Learn more
→](workflow.md)

### Chunks

Process jobs together in groups based on a size or wait time, e.g. 1,000 jobs or every 10
seconds. Partition chunks by worker, args, or meta, and apply outcomes to individual jobs.
[Learn more →](chunk.md)

```{toctree}
:maxdepth: 2
:caption: Getting Started

installation
adoption
```

```{toctree}
:maxdepth: 2
:caption: Operations

multi_process
erlang_notifier
smart_concurrency
pruner
```

```{toctree}
:maxdepth: 2
:caption: Jobs

unique_jobs
relay
signals
```

```{toctree}
:maxdepth: 2
:caption: Composition

workflow
chunk
```

```{toctree}
:maxdepth: 2
:caption: Reference

changelog
```

```{toctree}
:maxdepth: 1
:caption: Upgrading

upgrading/v0.7
```