# CloudHub Worker Sizing and Configuration

*MuleSoft Platform Architect Study Guide · Deploying API implementations to CloudHub · Last updated November 13, 2026 · https://www.certifyforce.com/study-guide/mulesoft-platform-architect/deploying-api-implementations-to-cloudhub/cloudhub-worker-sizing-and-configuration*

Size classic CloudHub workers by sustained compute, heap and disk, then plan bounded autoscaling against available account capacity.

- Certification: Salesforce MuleSoft Platform Architect
- Exam section: Deploying API implementations to CloudHub
- Study guide: https://www.certifyforce.com/study-guide/mulesoft-platform-architect

**Official exam objective:**

> Select CloudHub worker sizes and configuration as appropriate.

## What the exam expects

- Separate worker size from worker count and account entitlement.
- Read heap, worker memory and disk as different capacities.
- Select sustained compute capacity rather than relying on startup bursts.
- Plan classic CloudHub autoscaling within its limits, evaluation periods and available resources.

A **worker** is an independently managed Mule runtime instance; a **vCore** expresses configured compute capacity; **heap** is the Java object-memory area; **scale-up** changes worker size; **scale-out** changes worker count; an **autoscaling policy** adjusts resources in response to sustained usage ([Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture), [Autoscaling](https://docs.mulesoft.com/cloudhub/autoscaling-in-cloudhub)). These are classic worker-based CloudHub controls, rather than CloudHub 2 replica settings.

## Key facts

| Fact | Detail | Source |
| --- | --- | --- |
| Fractional compute | 0.1- and 0.2-vCore workers offer limited capacity with short bursts. | [Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture) |
| Consistent compute | Workers with 1 or more vCores provide performance consistency. | [Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture) |
| Heap comparison | 1 vCore has 2 GB heap; 2 vCores has 4 GB heap. | [Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture) |
| Disk comparison | 1, 2 and 4 vCores provide 12, 40 and 88 GB disk respectively. | [Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture) |
| Disk scope | Allocation includes operating-system and application storage. | [Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture) |
| Policy ceiling | Autoscaling supports at most four workers and 16-vCore size. | [Autoscaling](https://docs.mulesoft.com/cloudhub/autoscaling-in-cloudhub) |
| Scaling delay | Each step involves evaluation time plus cool-down. | [Autoscaling](https://docs.mulesoft.com/cloudhub/autoscaling-in-cloudhub) |
| CPU scope | Overall worker CPU, not JVM CPU alone. | [Autoscaling](https://docs.mulesoft.com/cloudhub/autoscaling-in-cloudhub) |

## Check yourself

- Why does the initial 128 MB metaspace size not establish its maximum?
- When available vCores are insufficient, why can an application exist without starting?
- When investigating earlier autoscaling events, where can the configured alert's history be inspected?

## Sources

- [CloudHub Architecture](https://docs.mulesoft.com/cloudhub/cloudhub-architecture)
- [Autoscaling in CloudHub](https://docs.mulesoft.com/cloudhub/autoscaling-in-cloudhub)
- [CloudHub High Availability Features](https://docs.mulesoft.com/cloudhub/cloudhub-fabric)
- [Deploy to CloudHub](https://docs.mulesoft.com/cloudhub/deploying-to-cloudhub)

## About the full unit

The "Core concepts", "Exam traps" and "Worked scenario" sections of this unit, its 6 linked practice questions, and unlimited timed mock exams are included with the MuleSoft Platform Architect pack (USD $19): https://www.certifyforce.com/practice-exams