Overview

In active-active mode,

  • The inter-rack deployment is transparent to the adjacent NFs.

  • The inter-rack deployment contains two instances of the CCG function, each instance manifest itself with a set of interface IPs.

  • Each instance support sets of sessions and continue to use the same IP for session consistency.

  • At a specific time period, one CCG instance can be primary only on one rack and standby on the other rack.

  • The set of interface IPs that are associated with the CCG instance, dynamically route to the primary rack of the instance.

cnSGW-CSMF supports primary/standby redundancy in which data is replicated from the primary to standby instance. The primary instance provides services in normal operation. If the primary instance fails, the standby instance becomes the primary and takes over the operation. To achieve inter-rack redundancy, two primary/standby pairs can be set up where each rack is actively processing traffic and standby is acting as backup for the remote rack.

In an Active-Active inter-rack redundancy deployment, consider there are two racks: Rack-1 and Rack-2 located in the same data center. All the NFs are trying to reach instance-1 and instance-2.

Active-Active Inter-Rack Redundancy Deployment

For NFs, both the instances are active. But in real, instance-1 and instance-2 are divided across racks.

Rack-1 has instance-1 and instance-2. In a pre-trigger scenario, instance-1 is local and acts as Primary and instance-2 is in Standby mode.

Rack-2 also has instance-1 and instance-2. In a pre-trigger scenario, instance-2 is local and acts as Primary and instance-1 is in Standby mode.

In case, if Rack-1 goes down, the traffic moves to Rack-2. On Rack-2 both the instances, instance-1 and instance-2 acts as Primary.