High availability

High availability for the Db2® SaaS performance plan is provided by native Db2 HADR.

Procedure

  1. Node architecture

    Each HA system consists of 2 nodes.

    Figure 1. Figure 1. Schematic view of the 2 nodes
    Schematic view of the 2 nodes
  2. Standby and replication behavior

    The primary node processes read and write transactions. The standby node is replicated synchronously, providing read-only capability. Each transaction is committed on at least 2 nodes to be considered successful.

    The standby node is also ready to take over write processing in case of failure or maintenance on the primary.

    Figure 2. Figure 2. Schematic view of primary node failover
    Schematic view of primary node failover
  3. Failover behavior

    During failover events, expect 10–20 seconds of restricted transactions. Clients can fail over seamlessly by using automatic client reroute (ACR) along with appropriate retry logic.