Clustering: Concepts and Principles

Clustering is the process of grouping a set of similar physical systems (often servers).

In order to work in cluster mode, you must have previously set up high availability on your system. Below is a list of the main concepts associated with clustering:

• High availability is a design approach that ensures a level of operational continuity and minimises the risk of unplanned downtime. It is achieved through load balancing and failover.

  Information noteNote: High availability with Talend refers only to the scheduling of task executions.

• Load balancing allows you to distribute work across the clustered Web servers. It is often combined with failover.

• Failover allows you to automatically switch to a standby server if the primary server is down or temporarily unreachable. For example, if a request to run a Job is sent to a server that is unreachable, it is automatically redirected to the backup server which will perform the desired operation.