Cisco Cisco IPCC Web Option Leaflet

The Unified ICM Central Controller provides the capability to create a single enterprise-wide queue.

The Unified ICM Central Controller provides consolidated reporting for all sites. 

The PG, Unified CM cluster, and Unified IP IVR must be co-located at the contact center site.

The communication link from the Unified ICM Central Controller to the PG must be sized properly 
and provisioned for bandwidth and QoS. (For details, refer to the chapter on 

Gatekeeper-based or RSVP Agent-based call admission control could be used to reroute calls 
between sites over the PSTN when WAN bandwidth is not available. It is best to ensure that 
adequate WAN bandwidth exists between sites for the maximum amount of calling that can occur.

If the communication link between the PG and the Unified ICM Central Controller is lost, then all 
contact center routing for calls at that site is also lost. Therefore, it is important to implement a 
fault-tolerant WAN. Even when a fault-tolerant WAN is implemented, it is important to identify 
contingency plans for call treatment and routing when communication is lost between the Unified 
ICM Central Controller and PG. For example, in the event of a lost Unified ICM Central Controller 
connection, the Unified CM CTI route points could send the calls to Unified IP IVR ports to provide 
basic announcement treatment or to invoke a PSTN transfer to another site. Another alternative is 
for the Unified CM cluster to route the call to another Unified CM cluster that has a PG with an 
active connection to the Unified ICM Central Controller. For more information on these options, 
refer to the chapter on 

.

While two intercluster call legs for the same call will not cause unnecessary RTP streams, two 
separate call signaling control paths will remain intact between the two clusters (producing logical 
hairpinning and reducing the number of intercluster trunks by two).

Latency between Unified ICM Central Controllers and remote PGs cannot exceed 200 ms one way 
(400 ms round-trip).

Initial call queuing is done on a Unified IP IVR co-located with the voice gateways, so no transcoding 
is required. When a call is transferred and subsequent queuing is required, the queuing should be done 
on a Unified IP IVR at the site where the call is currently being processed. For example, if a call comes 
into Site 1 and gets routed to an agent at Site 2, but that agent needs to transfer the call to another agent 
whose location is unknown, the call should be queued to a Unified IP IVR at Site 2 to avoid generating 
another intercluster call. A second intercluster call would be made only if an agent at Site 1 was selected 
for the transfer. The RTP flow at this point would be directly from the voice gateway at Site 1 to the 
agent’s phone at Site 1. However, the two Unified CM clusters would still logically see two calls in 
progress between the two clusters.

Transfers within a site function just like a single-site transfer. Transfers between Unified CM clusters 
use either the VoIP WAN or a PSTN service.

If the VoIP WAN is used, sufficient intercluster trunks must be configured. An alternative to using the 
VoIP WAN for routing calls between sites is to use a PSTN transfer service. These services allow the 
Unified CCE voice gateways to outpulse DTMF tones to instruct the PSTN to reroute (transfer) the call 
to another voice gateway location. Another alternative is to have the Unified CM cluster at Site 1 make 
an outbound call back to the PSTN. The PSTN would then route the call to Site 2, but the call would use 
two voice gateway ports at Site 1 for the remainder of the call.