To achieve high availability, you can cluster your API Gateway, API Data Store and Terracotta Server Array instances within a region. Clustering ensures that there is no single point of failure in the system in that region. For information about the cluster deployment, see Cluster Deployment. The typical HA architecture is as follows.

There is another variant to the HA architecture. You can store the Trasactional Events (TE) and logs in an External API Data Store too. For information about the external API Data Store, see Connecting to an External Elasticsearch. The HA architecture with the External API Data Store is as follows.

However, the HA architecture does not protect you from the data center failure in a region due to events like a natural disaster, cyber attack and so on. For protection against data center failures and to ensure business continuity, you must choose one business continuity solution from the following: