Cisco Unified Contact Center Express Solution Reference Network Design Release 9.0(2)
Estimating Bandwidth Consumption
Downloads: This chapterpdf (PDF - 1.45MB) The complete bookPDF (PDF - 5.12MB) | The complete bookePub (ePub - 1.95MB) | Feedback

Estimating Bandwidth Consumption

Bandwidth plays a large role in deployments involving:
  • The centralized call processing model (Unified CCX at the central site)
  • Any call deployment model that uses call admission control or a gatekeeper

Remote Agent Traffic Profile

Unified CCX signaling represents only a very small portion of control traffic (Agent/Supervisor Desktop to and from the Unified CCX Server) in the network. For information on TCP ports and Differentiated Services Code Point (DSCP) marking for Unified CCX and CTI traffic, see the sections on Security, and QoS and Call Admission Control.

Bandwidth estimation becomes an issue when voice is included in the calculation. Because WAN links are usually the lowest-speed circuits in an IP Telephony network, particular attention must be given to reducing packet loss, delay, and jitter where voice traffic is sent across these links. G.729 is the preferred codec for use over the WAN because the G.729 method for sampling audio introduces the least latency (only 30 msecs) in addition to any other delays caused by the network.

Where voice is included in bandwidth, system architects should consider the following factors:

  • Total delay budget for latency (taking into account WAN latency, serialization delays for any local area network traversed, and any forwarding latency present in the network devices). The generally agreed-upon limit for total (one-way) latency for applications in a network is 150 milliseconds.
  • Impact of delays inherent in the applications themselves. 8 seconds is the average Unified CCX agent login time with no WAN delay. This includes the exchange of approximately 1,000 messages between the agent application and various servers. The overall time to log in agents increases by approximately 30 seconds for each 30 milliseconds of WAN delay.
  • Impact of routing protocols. For example, Enhanced Interior Gateway Routing Protocol (EIGRP) uses quick convergence times and conservative use of bandwidth. EIGRP convergence also has a negligible impact on call processing and Unified CCX agent logins.
  • Method used for silently monitoring and recording agent calls. The method used dictates the bandwidth load on a given network link.

Silent Monitoring Bandwidth Usage

The silent monitoring feature of the CAD desktop software, which includes both listening to and recording agent calls, has the largest bandwidth requirements for the CAD product. Properly configuring this feature is especially important for remote agents who are connected to the main site by a WAN connection.

An agent's call can be listened to or recorded by the CAD software. To do this, a request is sent to a VoIP provider. The VoIP provider captures the voice streams representing the call (two voice streams per call) and sends them back to the requestor. The bandwidth requirements detailed in this section are for the network links between the requestor and provider.

Silent Monitoring Requestors

There are two possible requestors in the CAD software:
  • CSD
  • Recording service

CSDs will send requests when the supervisor wishes to listen to an agent's call in real-time. The VoIP provider will capture the voice streams and send them back to the supervisor's desktop where they can be listened to over the desktop's speakers.

A Recording service will send requests when either a supervisor or agent wishes to record the call. The VoIP provider will send the voice streams and the Recording service will save the streams to disk so they can be listened to at a later time.

In Unified CCX, the Recording service is installed on the Unified CCX server.

Silent Monitoring Providers

There are also two possible VoIP providers in the CAD software:

  • Cisco Agent Desktop
  • VoIP Monitor service

The Cisco Agent Desktop application contains a service referred to as the Desktop Monitor service that runs on the agent's desktop. It is responsible for processing silent monitoring requests only for the agent logged into the CAD application on the desktop. It captures voice packets sent to the IP or soft phone associated with the logged-in agent. The IP phone must be connected in series with the agent desktop on the network for this to work.

By default, this service is active on all agent desktops when the application is started. After initial installation of the CAD servers, all agents are already configured to use the Desktop Monitor service for the silent monitoring feature.

A VoIP Monitor service is able to handle multiple requests for silent monitoring simultaneously. It captures packets directly from the switch via the switch's Switched Port Analyzer (SPAN) configuration. In a deployment without high availability, Unified CCX supports one VoIP Monitor service, which is installed on the Unified CCX server. When high availability is deployed, two VoIP Monitor services are installed, one on each Unified CCX server.


Note


Agents who do not have a CAD destkop (such as IPPA and CADBE agents) must be configured to use a VoIP Monitor service for the silent monitoring feature.


The figure below shows a representative Unified CCX installation supporting a remote office over a WAN. Both the main office and the remote office have a VoIP and Recording service on site.
Figure 1. Contact Center Representation

It is easy to see where the bandwidth will be required for the silent monitoring feature when you can locate the requestors and providers.

Regardless of who the requestor and VoIP provider are, the bandwidth requirement between these two points is the bandwidth of the IP call being monitored and/or recorded. You can think of each monitoring and/or recording session as being a new phone call (2 voice streams) for calculating bandwidth. Therefore, to calculate bandwidth to support the Silent Monitoring feature, you can use the same calculations used to provision the network to handle call traffic.

IP Call Bandwidth Usage

An IP phone call consists of two streams of data. One stream is sent from phone A to phone B. The other stream is sent from phone B to phone A. The voice data is encapsulated into packets that are sent over the network. The amount of data required to store a voice stream is dependent upon the CODEC used to encode the data. The CAD software can support both the G.711 and G.729 CODEC.

The voice data itself is transmitted over the network using the Real-Time Transport Protocol (RTP). The RTP protocol supports the idea of silence suppression. When silence suppression is used, no voice packets are sent over the network if there is no sound.

Otherwise, even packets that contain silence are sent. This lowers the average required bandwidth for a call. Although CAD supports silence suppression, the lower bandwidth requirements for silence suppression should not be used when provisioning the network because the worst case scenario would be where there is not silence in the call, requiring the full call bandwidth as if silence suppression was not enabled.

When calculating bandwidth for an IP call, you must use the size of the RTP packet plus the additional overhead of the networking protocols used to transport the RTP data through the network.

For example, G.711 packets carrying 20 ms of speech data require 64 kbps (kilobytes per second) of network bandwidth per stream. These packets are encapsulated by four layers of networking protocols (RTP, UDP, IP, and Ethernet). Each of these protocols adds its own header information to the G.711 data. As a result, the G.711 data, once packed into an Ethernet frame, requires 87.2 kbps of bandwidth per data stream as it travels over the network. Since an IP phone call consists of two voice streams, in this example, a call would require 174.4 kbps.

The amount of voice data in a single packet also influences the size of the packet and bandwidth. The example above used packets containing 20 milliseconds of speech for its calculations, but this value can be changed in the Unified CM configuration for each supported CODEC. Configuring packets to contain more speech information reduces the number of packets sent over the network and reduces the bandwidth since there are fewer packets containing the additional networking headers, but the packet sizes increase.

The following table shows the bandwidth required for a phone call for the different combinations of CODEC and amount of speech per packet.
Table 1 Per-call Packet Size Bandwidth Requirements

CODEC

Milliseconds of speech per packet

Bandwidth required (Kbps) for a call

G.711

10

220.8

G.711

20

174.4

G.711

30

159.0

G.729

10

108.8

G.729

20

62.4

G.729

30

47.0

G.729

40

39.2

G.729

50

34.6

G.729

60

31.4


Note


  • The calculations are based on G.711 using a sampling rate of 64 kbps speech encoding and the G.729 using 8kbps. This means one second of speech encoded into the G.711 CODEC requires 65,536 bits (or 8,192 bytes) to represent one second of sound.
  • For full-duplex connections, the bandwidth speed applies to both incoming and outgoing traffic. (For instance, for a 100-Mbps connection, there is 100 Mbps of upload bandwidth and 100 Mbps of download bandwidth.) Therefore, an IP phone call consumes the bandwidth equivalent of a single stream of data. In this scenario, a G.711 IP phone call with no silence suppression and containing 20 milliseconds of speech per packet requires 87.2 kbps (174.4 / 2) of the available bandwidth.
  • Unified CCX supports a-law and μ-law for G.711.
  • If a prompt is recorded with G711 a-law phones and uploaded, you may encounter an error while playing the recorded prompt.

Bandwidth Available for Monitoring and Recording

The following tables display the percentage of total bandwidth available, based on the network connection, which is required for simultaneous monitoring sessions handled by a VoIP provider.

Table 2 Available Upload Bandwidth Percentage for Simultaneous Monitoring Sessions with G.711 CODEC

Number of Simultaneous Monitoring Sessions

Percentage of Available Bandwidth Required (No Silence Suppression)

100 Mbps

10 Mbps

1.544 Mbps

640 kbps

256 kbps

128 kbps

64 kbps

56 kbps

Call only

0.1

0.9

5.6

13.6

34.1

68.1

Not supported (NS)1

1

0.3

2.6

16.8

40.9

NS

NS

NS

NS

2

0.4

4.4

28.1

68.1

NS

NS

NS

NS

3

0.6

6.1

39.3

95.4

NS

NS

NS

NS

4

0.8

7.8

50.5

NS

NS

NS

NS

NS

5

1.0

9.6

61.7

NS

NS

NS

NS

NS

6

1.1

11.3

72.9

NS

NS

NS

NS

NS

7

1.3

13.1

84.2

NS

NS

NS

NS

NS

8

1.5

14.8

95.4

NS

NS

NS

NS

NS

9

1.7

16.6

NS

NS

NS

NS

NS

NS

10

1.8

18.3

NS

NS

NS

NS

NS

NS

1 The bandwidth of the connection is not large enough to support the number of simultaneous monitoring sessions.
Table 3 Available Upload Bandwidth Percentage for Simultaneous Monitoring Sessions with G.711 CODEC

Number of Simultaneous Monitoring Sessions

Percentage of Available Bandwidth Required (No Silence Suppression)

100 Mbps

10 Mbps

1.544 Mbps

640 kbps

256 kbps

128 kbps

64 kbps

Call only

0.0

0.3

2.0

4.9

12.2

24.4

48.8

1

0.1

0.9

6.0

14.6

36.6

73.1

Not supported (NS)2

2

0.2

1.6

10.0

24.4

60.9

NS

NS

3

0.2

2.2

14.1

34.1

85.3

NS

NS

4

0.3

2.8

18.1

43.9

NS

NS

NS

5

0.3

3.4

22.1

53.6

NS

NS

NS

6

0.4

4.1

26.1

63.4

NS

NS

NS

7

0.5

4.7

30.1

73.1

NS

NS

NS

8

0.5

5.3

34.1

82.9

NS

NS

NS

9

0.6

5.9

38.1

92.6

NS

NS

NS

10

0.7

6.6

42.2

NS

NS

NS

NS

2 The bandwidth of the connection is not large enough to support the number of simultaneous monitoring sessions.
The following notes apply to the bandwidth requirements shown in the tables above:
  • The bandwidth values are calculated based on the best speed of the indicated connections. A connection's true speed can differ from the maximum stated due to various factors.
  • The bandwidth requirements are based on upload speed. Download speed affects only the incoming stream for the IP phone call.
  • The values are based upon each voice packet containing 20 milliseconds of speech.
  • The number of bytes in each packet include the entire Ethernet encapsulation.
  • The data represents the CODECs without silence suppression. With silence suppression, the amount of bandwidth used may be lower.
  • The data shown does not address the quality of the speech of the monitored call. If the bandwidth requirements approach the total bandwidth available and other applications must share access to the network, latency (packet delay) of the voice packets can affect the quality of the monitored speech. However, latency does not affect the quality of recorded speech.
  • The data represents only the bandwidth required for monitoring and recording. It does not include the bandwidth requirements for other Cisco Agent Desktop modules as outlined in other sections of this document.
Related References

Bandwidth Requirements for VoIP Monitor Service

Although the bandwidth requirements are the same between the VoIP Monitor service and the Desktop Monitor service, the VoIP Monitor service can handle more simultaneous sessions (since it runs on the server). The tables below expand upon the tables in IP call bandwidth usage by increasing the number of simultaneous sessions.

Table 4 Available Upload Bandwidth Percentage for Simultaneous Monitoring Sessions with G.711 CODEC

Number of Simultaneous Monitoring Sessions

Percentage of Available Bandwidth Required (No Silence Suppression)

100 Mbps

10 Mbps

1.544 Mbps

1

0.3

2.6

16.8

5

1.0

9.6

61.7

10

1.8

18.3

Not supported (NS) 3

15

2.6

26.2

NS

20

3.5

34.9

NS

25

4.4

43.6

NS

30

5.2

52.3

NS

35

6.1

61.0

NS

40

7.0

69.8

NS

45

7.8

78.5

NS

50

8.7

87.2

NS

3 The bandwidth of the connection is not large enough to support the number of simultaneous monitoring sessions.
Table 5 Available Upload Bandwidth Percentage for Simultaneous Monitoring Sessions with G.711 CODEC

Number of Simultaneous Monitoring Sessions

Percentage of Available Bandwidth Required (No Silence Suppression)

100 Mbps

10 Mbps

1.544 Mbps

1

0.1

0.9

6.0

5

0.3

3.4

22.1

10

0.7

6.6

42.2

15

0.9

9.4

60.2

20

1.2

12.5

80.3

25

1.6

15.6

Not supported (NS)

30

1.9

18.7

NS

35

2.2

21.8

NS

40

2.5

25.0

NS

45

2.8

28.1

NS

50

3.1

31.2

NS

CAD Desktop Applications Bandwidth Usage

The CAD desktop applications include the following:
  • Cisco Agent Desktop
  • Cisco Supervisor Desktop
  • Cisco Desktop Administrator

These applications also require a certain amount of bandwidth, although far less than the Desktop Monitor service. In addition, the type of communication across the network is bursty. In general, bandwidth usage is low when the agents are not performing any actions. When features or actions are requested, the bandwidth increases for the time it takes to perform the action, which is usually less than one second, then drop down to the steady state level. From a provisioning standpoint, one must determine the probability of all the CAD agents performing a particular action at the same time. It might be more helpful to characterize the call center and determine the maximum number of simultaneous actions (in the worst case) to determine instantaneous bandwidth requirements, then determine what amount of delay is tolerable for a percentage of the requested actions.

For example, the raw bandwidth requirement for 300 CAD agents logging in simultaneously is about 4.5 Kilobytes/second and the login time is about 9 seconds (with no network delay) for each agent. If the WAN link did not have this much bandwidth, logins would take longer as packets were queued before being sent and received. If this caused the login attempts to take twice as long (18 seconds), would this delay be acceptable? If not, more bandwidth should be provisioned.

Each of these applications communicates with the base CAD services running on server machines. In addition, the agent desktop application communicates with the CTI server for call control actions and state changes. The following table displays the types of messaging for each application.

Table 6 Messaging Type by CAD Desktop Application

Application Name

Message types

Cisco Agent Desktop

Login/Logoff, Agent state changes, Call control, Call status Information,
Desktop Monitoring/Recording, Chat messages,
Team Performance messages,  
Report generation, and Real-time data refresh

Cisco Supervisor Desktop

Login/Logoff, Agent state updates, Call status updates,
Report generation, Silent Monitoring, Call Recording, Call Playback,
Chat messages, Team Performance messages, and Real-time data refresh

Cisco Desktop Administrator

Configuration information retrieval and storage
Configuration data refresh

Cisco Agent Desktop Bandwidth Usage

CAD agents are able to login and logoff their agents, change their agent state, handle calls, and send reporting information to the base servers. The bandwidth requirements for these activities are fairly small but can add up when many agents are considered.

The following table displays the average bandwidth requirements for different numbers of agents. This information is derived from bandwidth testing and extrapolation of bandwidth data. Since there are many variables that can affect bandwidth, a configuration that resulted in higher bandwidth usage was chosen to provide near worst-case scenarios. If the agent's WAN link meets or exceeds the bandwidth requirements shown in this table, Cisco Agent Desktop will be able to run without delays in message passing.
Table 7 Average Bandwidth Requirements for Cisco Agent Desktop

Number of agents

Average Download Bandwidth (Kilobytes/second)

Average Upload Bandwidth (Kilobytes/second)

1

0.03

0.05

10

0.2

0.3

50

1.1

1.5

100

2.2

3.0

150

3.3

4.5

200

4.4

6.0

250

5.5

7.0

300

6.6

9.0


Note


The bandwidth requirements shown do not include the bandwidth of the RTP streams for the call, recording, or monitoring session.


The parameters that affect bandwidth and apply to table 28 are shown below.

  • Number of skills per agent: 10
  • Number of agents per team: 20
  • Number of teams: 50
  • Number of agent state changes per agent per hour: 10 (does not count state changes due to handling calls)
  • Calls per agent per hour: 60
  • Team Performance Messages per team per hour: 8
  • Chat messages sent/received per hour: 20
  • Average chat message size (in bytes): 40
  • Number of calls recorded per hour: 0

Cisco Agent Desktop provides a bandwidth calculator that can be used with both Unified CCX and Unified CCE. Additional information about the Cisco Agent Desktop Bandwidth Calculator is available at http:/​/​www.cisco.com/​univercd/​cc/​td/​doc/​product/​icm/​bandcalc/​index.htm.

Cisco Supervisor Desktop Bandwidth Usage

A CSD will receive events for all the agent's of the team that the supervisor is logged into. This information includes state changes, call handling, login/logoff, and so on. The more agents, skills, and calls there are, the more data will be sent to supervisors. In addition, particular reports are automatically refreshed periodically to provide real-time data while the supervisor is viewing the report. Refreshing reports requires additional bandwidth.

The following table uses the same basic configuration parameters used to determine the bandwidth numbers in Cisco Agent Desktop bandwidth usage, with the following differences:
  • The calculations are based on 1 supervisor per 10 agents
  • The Team Agent Statistics Report is viewed
  • The Team Skill Statistics Report is viewed

Number of agents

Average Download Bandwidth (Kilobytes/second)

Average Upload Bandwidth (Kilobytes/second)

1

0.05

0.05

10

0.05

0.05

50

0.2

0.2

100

0.5

0.5

150

0.7

0.7

200

1.0

1.0

250

1.2

1.2

300

1.4

1.5

Cisco Desktop Administrator Bandwidth Usage

The bandwidth requirements for CDA are very small and are only seen when an administrator is actively changing configurations. In general, the bandwidth used by CDA is negligible from a provisioning standpoint.

Web Chat Feature

When deploying the Unified CCX along with Cisco SocialMiner observe the following network requirements.

Delay: The maximum allowed round-trip time (RTT) between the Unified CCX server and SocialMiner is 150 ms.

Bandwidth: In addition to the requirements for the Unified CCX and Unified CM clusters, you must provision sufficient bandwidth for SocialMiner, the customer web server, and remote agent/supervisor desktops to deploy Web Chat successfully. You must take into account the bandwidth required for the following components:

  • Unified CCX and SocialMiner: If SocialMiner and the Unified CCX are not co-located, there is an additional bandwidth requirement for the communication and contact signaling.
  • SocialMiner and Cisco Agent Desktop: When a chat session starts, depending on the chat transcript size and communication frequency, there is an additional bandwidth requirement between SocialMiner and the Cisco Agent Desktop.
  • SocialMiner and customer website: The customer website transfers all new chat contact requests to SocialMiner. When the chat contact reaches SocialMiner, an active session is maintained and on commencement of chat, the chat agent and the customer website have to and from data traffic to carry chat messages. If the customer website is not on the same network as SocialMiner, you must ensure that the bandwidth requirement is based on mean per session chat traffic.

The following table shows the minimum bandwidth requirement for the Unified CCX and SocialMiner when they are not located in the same network.

Between Unified CCX and SocialMiner (Kilobytes/second)

Between SocialMiner and Agent Desktop (Kilobytes/second)

Between Customer Webserver and SocialMiner (Kilobytes/second)

100

100*

100*

*Based on a maximum of 50 supported concurrent chat sessions and on the assumption that the average chat message length is 500 characters, and of all the active sessions at any point in time, only 10% contribute to message traffic on the wire. For more information, see http:/​/​docwiki.cisco.com/​wiki/​SocialMiner_​Release_​9.0(1)#Provisioning

Quality of Service: The criteria and considerations are the same as the Unified CCX High Availability deployment.

Remote Agent Traffic Profile

Unified CCX signaling represents only a very small portion of control traffic ( Unified CM CTI and ICD subsystems) in the network. For information on TCP ports and Differentiated Services Code Point (DSCP) marking for Unified CCX ICD and CTI traffic, see Security, and QoS and Call Admission Control.

Bandwidth estimation becomes an issue when voice is included in the calculation. Because WAN links are usually the lowest-speed circuits in an IP Telephony network, particular attention must be given to reducing packet loss, delay, and jitter where voice traffic is sent across these links. G.729 is the preferred codec for use over the WAN because the G.729 method for sampling audio introduces the least latency (only 30 milliseconds) in addition to any other delays caused by the network.

Where voice is included in bandwidth, system architects should consider the following factors:
  • Total delay budget for latency (taking into account WAN latency, serialization delays for any local area network traversed, and any forwarding latency present in the network devices). The generally agreed-upon limit for total (one-way) latency for applications in a network is 150 milliseconds.
  • Impact of delays inherent in the applications themselves. 25 seconds is the initial Unified CCX agent login setup time with no WAN delay. The overall time to log in agents and base delay adds approximately 30 seconds of delay per 30 milliseconds of WAN delay.
  • Impact of routing protocols. For example, Enhanced Interior Gateway Routing Protocol (EIGRP) uses quick convergence times and conservative use of bandwidth. EIGRP convergence also has a negligible impact on call processing and Unified CCX agent logins.
Use the following table to estimate the number of Unified CCX agents that can be maintained across the WAN (with IP Telephony QoS enabled). These numbers are derived from testing where an entire call session to Unified CCX agents, including G.729 RTP streams, is sent across the WAN. Approximately 30% of bandwidth is provisioned for voice. Voice drops are more of an issue when you are running RTP in conjunction with Cisco Agent Desktop and other background traffic across the WAN. These voice drops might occur with a specific number of agents at a certain link speed, and those possible scenarios are denoted by the entry N/A (not applicable) in the following table.
Table 8 Remote Agents Supported by Unified CCX Across a WAN Link

Frame Relay

128 KB

256 KB

512 KB

768 KB

T1

G.729

3

7

15

25

38

G. 711

N/A

N/A

N/A

N/A

14

In remote agent deployments, QoS mechanisms should be used to optimize WAN bandwidth utilization. Advanced queuing and scheduling techniques should be used in distribution and core areas as well. For information on QoS traffic classification, see QoS and Call Admission Control. For provisioning guidelines for centralized call processing deployments, refer the Cisco IP Telephony Solution Reference Network Design documentation, available online at: http:/​/​www.cisco.com/​go/​ucsrnd.

Security

Security can be implemented on many levels. Applications security is clearly dependent upon security implemented at the infrastructure level. For more details on security at the network infrastructure level, refer the security design considerations in the Cisco IP Telephony Solution Reference Network Design documentation, available online at

http:/​/​www.cisco.com/​warp/​public/​779/​largeent/​it/​ese/​srnd.html

Corporate Data Access

Aside from call routing, Unified CCX or Cisco Unified IP IVR scripts often process enterprise data from existing corporate data stores such as a database or a corporate directory server for functions such as account authorization and order status. Often, these data stores already exist and share data with other enterprise applications.

The figure below shows an example of a network where voice and data components reside in separate VLANs and are separated by a firewall.
Figure 2. Unified CCX Accessing Data Stores

Unified CCX can communicate with these external sources through its subsystems, provided Network Address Translation (NAT) is not used.

Port Utilization for Product Revisions

For a list of the TCP and UCP ports used by Unified CCX, including Cisco Unified IP IVR and Unified CCX, see the Port Utilization Guide, which is accessible from the following web page:

http:/​/​www.cisco.com/​en/​US/​products/​sw/​custcosw/​ps1846/​products_​installation_​and_​configuration_​guides_​list.html

Ping NAT PAT and Reverse DNS Lookups

The following configurations and information are required for the CAD software to work properly.

The Cisco Agent Desktop application uses the TCP Ping command to verify that it can communicate with the active VoIP servers. This is done even if no agents are configured to use a VoIP Monitor service for the silent monitoring feature. If Ping is disabled on the machine running a CAD VoIP Monitor Server, the silent monitoring feature will not work properly.

There are certain CAD modules that rely upon reverse DNS lookups. If this feature is turned off on the machines running CAD services, there will be a loss of some functionality and errors will be generated and logged.

Network Address Translation (NAT) and Port Address Translation (PAT) are supported between CAD and the Unified CCX servers as long as the CAD client applications are behind a VPN. NAT is supported with IP Phone Agent (IPPA). However, it is required that you use static IP addresses for the IP Phone Agent phones as well as Static NAT. Dynamic NAT and address overloading are not supported. PAT is not supported with IPPA. Refer to the Cisco CAD Installation Guide for more detail.

QoS and Call Admission Control

Quality of Service (QoS) becomes an issue when more voice and application-related traffic is added to an already growing amount of data traffic on your network. Accordingly, Unified CCX and time-sensitive traffic such as voice need higher QoS guarantees than less time-sensitive traffic such as file transfers or emails (particularly if you are using a converged network).

QoS should be used to assign different qualities to data streams to preserve Unified CCX mission-critical and voice traffic. The following are some examples of available QoS mechanisms:
  • Packet classification and usage policies applied at the edge of the network, such as Policy Based Routing (PBR) and Committed Access Rate (CAR).
  • End-to-end queuing mechanisms, such as Low Latency Queuing (LLQ). Because voice is susceptible to increased latency and jitter on low-speed links, Link Fragmentation and Interleaving (LFI) can also be used to reduce delay and jitter by subdividing large datagrams and interleaving low-delay traffic with the resulting smaller packets.
  • Scheduling mechanisms such as Traffic Shaping to optimize bandwidth utilization on output links.

Unified CCX and Application-Related Traffic

The table below lists TCP ports and DSCP markings for use in prioritizing Unified CCX and Unified CM mission-critical CTI traffic. The DSCP Markings for call signaling traffic between Unified CCX and Cisco Unified Communication manager and for voice traffic played from the Unified CCX server are set by default according to the recommended traffic classification guidelines documented in Cisco Unified Communications System Design Guidance, available at: http:/​/​www.cisco.com/​go/​ucsrnd.

Unified CCX does not mark any network traffic other than those mentioned above. As a result, traffic should be marked and prioritized at the edge, according to the recommendations in the below table.

Note


Unified CCX does not mark RTP traffic sources from the Unified CCX engine.
The performance criteria used in classifying such traffic includes:
  • No packet drops on the outbound or inbound interface of the WAN edge router
  • Voice (G.729) loss under 1%
  • One-way voice delay under 150 ms

A detailed description of QoS is not within the scope of this design guide. For QoS design recommendations, refer the Quality of Service design guide available online at: http:/​/​www.cisco.com/​go/​designzone

Table 9 Recommended QoS Classifications for Unified CCX Interfaces

Unified CCX Component

Interface / Protocol

Port

DSCP Marking

Unified CCX Engine - CTI-QBE messaging destined to Unified CM from Unified CCX

CTI-QBE

TCP 2748

CS3

Unified CCX Administration and BIPPA Service - HTTP traffic destined for web administration and BIPPA interface on Unified CCX

HTTP / HTTPS

TCP 8443

AF21

Unified CCX Engine and Unified CCX Administration - SOAP AXL HTTPS messaging destined to Unified CM from Unified CCX

HTTPS / SOAP

TCP 8443

AF21

VoIP Monitor Service - RTP voice bearer traffic destined for CSD from Unified CCX (SPAN-based) or CAD (Desktop-based) monitoring

RTP

UDP 59010

UDP 59012

EF

QoS Considerations for CAD Software

The most important network traffic for quality of service consideration in the CAD software is the voice streams sent between VoIP requestors and providers. The processes that send and receive these voice streams have been set to have higher priorities than other processing threads. This helps assure that there will be no delays in processing these voice streams. However, The voice streams themselves contain no QoS markings. These markings are stripped off when the voice streams are captured by the VoIP provider's software. The networking components used to send these data streams (switches, routers, gateways) should be configured with the appropriate QoS settings to ensure the delivery of these voice streams to meet the intended QoS requirements.

CAC and RSVP

Unified CM supports Resource-Reservation Protocol (RSVP) between endpoints within a cluster. RSVP is a protocol used for Call Admission Control (CAC) and is used by the routers in the network to reserve bandwidth for calls. The bandwidth being controlled is only for the voice streams, call signalling traffic is not part of CAC.

Before RSVP, each Unified CM cluster maintained its own calculation of how many active calls were traversing between locations in order to calculate bandwidth usage. If more than one Unified CM cluster shared the same link, bandwidth would have to be carved out and dedicated for each cluster, and this led to inefficient use of available bandwidth. RSVP also enables customers to deploy complex network topology while Location-based CAC is limited to a hub-and-spoke type of topology.

RSVP solves this problem by tracing the path between two RSVP Agents that reside on the same LAN as the IP Phones. A software MTP or transcoder resource that runs on Cisco IOS routers can be RSVP Agents. The RSVP Agents are controlled by Unified CM and are inserted into the media stream between the two IP phones when a call is made. The RSVP Agent of the originating IP Phone will traverse the network to the destination IP Phone's RSVP Agent, and reserve bandwidth. Since the network routers (and not Unified CM) are keeping track of bandwidth usage, multiple phone calls can traverse the same RSVP controlled link even if the calls are controlled by multiple Unified CMs.

For more information, see the RSVP chapter in Cisco Unified Communications Solution Reference Network Design (SRND).

Unified CCX selects a call center agent independent of the mechanism, using either RSVP or Location-based CAC; that is, Unified CCX routes a call to an available agent even though the agent phone might not be able to receive the call due to lack of bandwidth. Thus, proper sizing of bandwidth between sites is very important.

For any call transfer, there are moments when two calls are active. If any of the active calls traverses between sites, then CAC is used. Even when the original call is placed on hold during a transfer, that call still takes up the same amount of bandwidth just like an active call.

In the two examples illustrated below, the voice gateway and agents are at a remote site, while the Unified CCX server is at a data center site. A call from PSTN reaches the voice gateway at the remote site and connects to Unified CCX at the data center. This takes one call bandwidth over the WAN link, which is represented by the caller stream. Once an agent is available and selected at the remote site, Unified CCX transfers the call to the agent.

Figure 3. Call from PSTN to Unified CCX Server to Agent

During the transfer, before the agent picks up the call, there is another call setup between Unified CCX and the agent phone. It takes up another call bandwidth over the WAN, and is represented by the agent stream in the example above. Once the agent picks up the call, the voice traffic is between the voice gateway and the agent phone, which are both at the remote site. At that time, no bandwidth is reserved over the WAN, as illustrated in the example below. This example shows how call bandwidth is reserved in a contact center call that is eventually routed to an agent. Depending on where the voice gateway, the agents, and the Unified CCX server are located, proper WAN bandwidth should be provisioned.

Figure 4. After Agent Picks Up Call

Bandwidth Security and QoS Considerations for Unified Intelligence Center

The two bandwidth measurements in a Unified Intelligence Center installation include the following:

  • Bandwidth between the Unified Intelligence Center and data source
  • Bandwidth between the user and Unified Intelligence Center

The Unified CCX database is local to the server. In a normal operating mode, the bandwidth between Unified Intelligence Center and the data source can be ignored.


Note


Each report requires about 2.6 Mbps of bandwidth between the user and Unified Intelligence Center.


The configuration parameters that affect bandwidth include the following:

  • Number of rows in the report: 8000
  • Size of each row: 500 bytes
  • HTML size overhead for each row: 500 bytes
  • Time to transfer the rendered report from Unified Intelligence Center to the browser: 3 seconds