Zoomdata microservices and related components can be deployed on-premise, in the cloud, or in hybrid environments, provided that the host server meets operating system and hardware requirements. Additionally, Zoomdata can query supported data sources in the cloud or on-premise, provided network-level connectivity exists between the Zoomdata host and the data store. As a best practice, SSL should be used on connections between Zoomdata and a data store, especially when traversing network enclaves.
As a Java web application, Zoomdata can meet increasing user concurrency requirements through horizontal scaling. Horizontal scale-out is achieved by replicating Zoomdata behind a load balancer with a common metadata store for all nodes. This configuration also provides a higher degree of application availability, as traffic will be routed to healthy nodes in the event of an application or hardware failure on another node.
Another method of scaling available in Zoomdata’s architecture is the independent scale-out of the query engine. In this model, the query engine is deployed to a Hadoop cluster as a Spark on Yarn application.
As the Zoomdata microservices architecture matures, more microservices will be deployed independently for greater availability and optimized load management.
Was this topic helpful?