How Do Cloudera Hybrid Environments Work?
Cloudera Hybrid Environments use a novel compute-only cloud replica model based on in-place data access.
Workloads submitted to compute services in Cloudera Hybrid Data Hubs access data and metadata directly from an associated Cloudera on premises cluster.
This architecture relies on the following building blocks.
- Unified Authentication
- Implementing a two-way Kerberos cross-realm trust between the hybrid cloud and on-premises clusters. This enables centralized authorization and governance.
- Metadata Synchronization
- Associating Cloudera on premises cluster with the Cloudera Hybrid Data Hub cluster as its metadata, authorization, and governance context.
- Network Connectivity
- Ensuring stable, bi-directional network connectivity exists between the on-premises cluster and hybrid cloud environments to support in-place data read/write operations for active jobs.
- Workload Portability
- Maintaining a unified runtime version across both the Hybrid Datahub and the on-premises cluster for default workload portability.
