> ## Documentation Index
> Fetch the complete documentation index at: https://enterprise-docs.crewai.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Worker Configuration

> Configuration for the CrewAI Platform background worker processes.

<ParamField path="worker.name" type="string" default="worker">
  Name identifier for worker component.
</ParamField>

<ParamField path="worker.replicaCount" type="integer" default="1">
  Number of worker replicas.

  **Scaling Guidelines:**

  | Job Volume | Description          | Recommended Replicas |
  | ---------- | -------------------- | -------------------- |
  | Low        | \< 100 jobs/hour     | 1-2                  |
  | Medium     | 100-1000 jobs/hour   | 2-4                  |
  | High       | 1000-10000 jobs/hour | 4-8                  |
  | Very High  | > 10000 jobs/hour    | 8+                   |

  **Considerations:**

  * Each worker replica increases database connection pool usage
  * Monitor queue depth to determine if more workers are needed
  * Scale horizontally rather than increasing resources per worker
</ParamField>

### `worker.resources.*`

Resource limits and requests for worker container.

<ParamField path="worker.resources.limits.cpu" type="string" default="6">
  CPU limit for worker pod.

  **Sizing Guidelines:** Workers have similar resource needs to web pods. Job complexity affects resource usage.
</ParamField>

<ParamField path="worker.resources.limits.memory" type="string" default="12Gi">
  Memory limit for worker pod.

  **Tuning:** Monitor actual usage and adjust based on job requirements.
</ParamField>

<ParamField path="worker.resources.requests.cpu" type="string" default="500m">
  Guaranteed CPU allocation for worker pod.
</ParamField>

<ParamField path="worker.resources.requests.memory" type="string" default="6Gi">
  Guaranteed memory allocation for worker pod.
</ParamField>

<ParamField path="worker.nodeSelector" type="object" default="{}">
  Node selector for worker pod placement.

  **Example:**

  ```yaml theme={null}
  worker:
    nodeSelector:
      node-role.kubernetes.io/worker: "true"
      workload-type: "background-jobs"
  ```

  **Use Cases:**

  * Separate background job processing from web traffic
  * Ensure workers run on nodes with specific capabilities
  * Control pod placement for optimal resource utilization
</ParamField>