Common terms

RDBMS Terms

Different RDBMS (relational database management systems) have different terminology when it comes to clusters, instances, databases and schemas.

In DBmarlin we try to take a generic definition of these terms and apply it to all RDBMS platforms that we support.

Cluster

A cluster is a group of instances that are related. Typically the instances will use some kind of replication to propagate data changes between the different instances. This could be Active-Passive with 1 Primary/Source instance handling all writes and 1 or more Secondary/Replica instances receiving replication traffic, or it could be Active-Active with more than 1 instance able to handle writes. Cluster like this is usually done for high availability (HA), disaster recovery (DR), geo-replication or scaling out.

Instance

An instance is the software installed and memory used to access a database. Typically this corresponds to a host/port combination where the port is the listener port for the database software.

Database or Schema

A database or schema, depending on the RDBMS used, defines how the data will be organised (into tables and other objects) and contains the data and metadata stored in physical files on disk.