On which runtime planes is clustering not available?

Clustering in MuleSoft typically refers to the capability of distributing workloads across multiple instances of a runtime to ensure high availability and scalability. In the context of CloudHub, it is worth noting that it is a fully managed integration platform-as-a-service (iPaaS) provided by MuleSoft for deploying applications in the cloud.

The architecture of CloudHub does not support traditional clustering as seen in on-premises deployments. Instead of clustering, CloudHub uses a concept called "worker scaling," where you can scale instances vertically or horizontally, but these instances do not operate as a clustered environment in the traditional sense. The lack of clustering is a design decision to optimize for a cloud environment where automatic scaling and load balancing are managed by the platform itself.

In contrast, customer-hosted runtime planes, hybrid cloud platforms, and dedicated servers can implement clustering because they provide more control over the infrastructure, allowing for setup configurations that support service clustering. This allows these environments to share workloads and enhance resilience in case of hardware or software failures. Thus, when referencing where clustering is not available, CloudHub serves as the correct answer, as it defaults to a model that abstracts away the need for end-users to manage clustering directly while providing its own load balancing and scaling features.