Cluster of Dgraph nodes

A typical BDD cluster deployment includes a set of machines that are solely dedicated to running the Dgraph. This set of machines is known as the Dgraph cluster.

A Dgraph cluster is a set of Dgraphs that together handle requests for data sets in Big Data Discovery. Requests arriving from Studio are routed and load-balanced between the Dgraph nodes. One of these Dgraph nodes is responsible for handling all write operations (updates, configuration changes), while the remaining Dgraphs serve as read-only. All Dgraph nodes in the cluster utilize an index residing on shared storage.

The leader and follower Dgraph nodes differ in the types of queries they can process, however, this is transparent to the end users of Big Data Discovery. The allocation of leader and follower Dgraph node roles is performed by the BDD cluster automatically.

A Dgraph node is a node in BDD cluster deployment that runs the Dgraph. The Dgraph is the main computational module that provides search, refinement computation, Guided Navigation, and many other features, all of which you can observe and use in Studio.

In a BDD cluster deployment, you can have only one cluster of Dgraph nodes. All nodes in BDD that run Studio and Dgraph Gateway in WebLogic Server talk to the same single cluster of Dgraph nodes. The Dgraph cluster can have any number of nodes, even though a certain number of Dgraph nodes is recommended for production environment. For more information, see the Installation and Deployment Guide.