Cisco Cisco Unified Contact Center Enterprise 9.0(1)
Serviceability Best Practices Guide for Unified ICM/Unified CCE & Unified CCH
©2012 Cisco Systems, Inc.
17
inside of the same PG. Each peripheral on this PG behaves completely independently. Currently the
Generic PG is supported only for the Unified CCE, where it contains a Communications Manager
PIM and a VRU PIM talking to an IP-IVR or Customer Voice Portal (CVP). The second type of
group PG is a Unified CCE System PG. This PG, like the generic PG, has one Call Manager PIM and
one or more VRU PIMs. The System PG ties these multiple PIMs together. In a traditional Unified
CCE, a call that comes into the Communications Manager then gets transferred to the IP-IVR and
then back to an agent looks like three separate calls to the Unified CCE. The new PG coordinates
these calls and makes that call look like a single call. This is what happens on a traditional TDM
ACD, where the ACD also has a queue point.
Generic PG is supported only for the Unified CCE, where it contains a Communications Manager
PIM and a VRU PIM talking to an IP-IVR or Customer Voice Portal (CVP). The second type of
group PG is a Unified CCE System PG. This PG, like the generic PG, has one Call Manager PIM and
one or more VRU PIMs. The System PG ties these multiple PIMs together. In a traditional Unified
CCE, a call that comes into the Communications Manager then gets transferred to the IP-IVR and
then back to an agent looks like three separate calls to the Unified CCE. The new PG coordinates
these calls and makes that call look like a single call. This is what happens on a traditional TDM
ACD, where the ACD also has a queue point.
Figure 3: Peripheral Gateway Architecture
The PG is duplexed using the same technology as the Central Controller, MDS. This means that there
are two PGs operating at any time. All of the messages to the critical process on the PG (OPC) go
through the MDS queue, to keep the two operating in lock-step. However, the PG operates slightly
different from the Router – from a fault tolerance standpoint – in that while both sides share the same
data, for many PG components, only one side is active. Should a fault occur, the opposite side
activates and continues functioning, having the context of the other side without losing calls.
are two PGs operating at any time. All of the messages to the critical process on the PG (OPC) go
through the MDS queue, to keep the two operating in lock-step. However, the PG operates slightly
different from the Router – from a fault tolerance standpoint – in that while both sides share the same
data, for many PG components, only one side is active. Should a fault occur, the opposite side
activates and continues functioning, having the context of the other side without losing calls.
PGs use the Device Management Protocol (DMP) to communicate between themselves and the
central controller. The following figure depicts the components involved in this communication and
the communication links employed.
central controller. The following figure depicts the components involved in this communication and
the communication links employed.
MDS
OPC
Central
Controller
CCAgent
PIM
PIM 1
Peripheral 1
Peripheral 2
CTI Server
CTI-OS
Agent Desktops
CRM
3rd Party ACD
GED 188/
OCCI
GED 188/
ACMI
GED 188
ARM Clients
(BA, CEM, CCS)
ARM
To Parent
ICM