Cisco Cisco IP Contact Center Release 4.6.2 Design Guide

Page of 388
 
12-2
Cisco Unified Contact Center Enterprise 7.5 SRND
Chapter 12      Bandwidth Provisioning and QoS Considerations
What's New in This Chapter
What's New in This Chapter
 lists the topics that are new in this chapter or that have changed significantly from previous 
releases of this document.
 
Unified CCE Network Architecture Overview
Unified CCE is a distributed, resilient, and fault-tolerant network application that relies heavily on a 
network infrastructure with sufficient performance to meet the real-time data transfer requirements of 
the product. A properly designed Unified CCE network is characterized by proper bandwidth, low 
latency, and a prioritization scheme favoring specific UDP and TCP application traffic. These design 
requirements are necessary to ensure both the fault-tolerant message synchronization of specific 
duplexed Unified CCE nodes (Central Controller and Peripheral Gateway) as well as the delivery of 
time-sensitive system status data (agent states, call statistics, trunk information, and so forth) across the 
system. Expeditious delivery of PG data to the Central Controller is necessary for accurate call center 
state updates and fully accurate real-time reporting data.
In a Cisco Unified Communications deployment, WAN and LAN traffic can be grouped into the 
following categories:
  •
Voice and video traffic
Voice calls (voice carrier stream) consist of Real-Time Transport Protocol (RTP) packets that 
contain the actual voice samples between various endpoints such as PSTN gateway ports, 
Unified IP IVR Q-points (ports), and IP phones. This traffic includes voice streams of silently 
monitored and recorded agent calls.
  •
Call control traffic
Call control consists of packets belonging to one of several protocols (H.323, MGCP, SCCP, or 
TAPI/JTAPI), according to the endpoints involved in the call. Call control functions include those 
used to set up, maintain, tear down, or redirect calls. For Unified CCE, control traffic includes 
Table 12-1
New or Changed Information Since the Previous Release of This Document 
New or Revised Topic
Described in:
Best practices and recommendations for Cisco 
Agent Desktop Service Placement
Bandwidth requirements for Intra-Cluster 
Communication Signaling (ICCS)
Multilink Point-to-Point Protocol (MLPPP) for 
private networks
Latency requirements for high-priority traffic
Microsoft Packet Scheduler
Port utilization for Public Network
Private bandwidth example calculation
Queuing policy