Filter
Top Products
12-16
Cisco Broadband Access Center 3.8 Administrator Guide
OL-27172-01
Chapter 12 Configuring CWMP Service
Provisioning Group Scalability and Failover
Provisioning Group Scalability and Failover
This section describes the scalability and failover features provided in this release of Cisco BAC.
Cisco BAC’s scalability and failover provide a high degree of availability to suit networks of virtually
any size, even those with millions of devices deployed. The product also provides such critical features
as DPE redundancy and failover protection.
Scalability in a Cisco BAC deployment is accomplished through provisioning groups, each of which is
a cluster of redundant DPE servers that communicate with a set of CPE. Provisioning groups enhance
the scalability of the Cisco BAC network by making each provisioning group responsible only for a
subset of devices.
This partitioning of devices can be along regional groupings or any other policy defined by the service
provider.
To scale a deployment, the administrator can:
• Upgrade existing DPE server hardware.
• Add DPE servers to a provisioning group.
• Add provisioning groups and redistribute devices among them.
Cisco BAC supports explicit assignment and automatic membership of devices to provisioning groups.
For more information, see CPE Management Overview, page 4-1.
Redundancy in Cisco BAC
Redundancy helps to ensure:
• High availability for your network applications.
• Users do not experience long network delays or black holes due to a single point of failure.
Cisco BAC supports local as well as regional server redundancy.
Local Redundancy
The Cisco BAC provisioning group is a cluster of redundant DPE servers that communicate with a set
of devices. A single URL identifies each provisioning group.
You can configure local redundancy by using any software or hardware that provides load-balancing
features, such as the Cisco Application Control Engine (ACE) 4710.
Regional Redundancy
Regional redundancy ensures that DPEs in a different location temporarily process CPE requests in case
of regional failure. The simplest way of facilitating this deployment is to set up DPEs within a
provisioning group in different geographic locations. In such a setup, CPE requests are serviced by DPEs
located in different regions, providing regional failover within the confines of the provisioning group.
In some deployments, however, you may require that CPE requests be processed by servers in one
location and fail over to DPEs in another location only in case of failure. In such a scenario, you can also
locate DPEs for a provisioning group in different regions yet configure the network such that, under
normal conditions, CPE requests are directed only to a subset of the DPEs in a particular region.