Cisco Unified Contact Center Enterprise Solution Reference Network Design, Release 9.x
Deployment Models
Downloads: This chapterpdf (PDF - 3.83MB) The complete bookPDF (PDF - 12.19MB) | The complete bookePub (ePub - 7.08MB) | Feedback

Deployment Models

Contents

Deployment Models

There are numerous ways that Unified Contact Center Enterprise (Unified CCE) can be deployed, but the deployments can generally be categorized into the following major types or models:
  • Single Site
  • Multisite Centralized Call Processing
  • Multisite Distributed Call Processing
  • Clustering over the WAN
Many variations or combinations of these deployment models are possible. The primary factors that cause variations within these models are as follows:
  • Location of Unified CCE servers
  • Location of Voice Gateways
  • Choice of inter-exchange carrier (IXC) or local exchange carrier (LEC) trunks
  • Pre-routing availability
  • IVR queuing platform and location
  • Transfers
  • Traditional ACD, PBX, and IVR integration
  • Sizing
  • Redundancy

This topic discusses the impact of these factors (except for sizing) on the selection of a design. With each deployment model, this topic also lists considerations and risks that must be evaluated using a cost benefit analysis. Scenarios that best fit a particular deployment model are also noted.

In this topic, section titles are prefaced by the type of factor discussed in the section. The factors are classified into the following categories:
  • IPT—Cisco Unified Communications deployment factors (how Cisco Unified Communications Manager and the Voice Gateways are deployed)
  • Unified CCE—Unified CCE deployment factors (such as what PG is used)
  • IVR—IVR and queuing deployment factors (if Unified CVP or Unified IP IVR is used)

A combination of these deployment models is also possible. For example, a multisite deployment may have some sites that use centralized call processing (probably small sites) and some sites that use distributed call processing (probably larger sites). Examples of scenarios where combinations are likely are identified within each section.

Also in this chapter is a section on integration of traditional ACD and IVR systems into a Unified CCE deployment with considerations on hybrid PBX/ACD deployments. Sizing and redundancy are discussed in later chapters of this Unified CCE design guide. For more information about the network infrastructure required to support a Unified CCE solution, see the latest version of the Cisco Network Infrastructure Quality of Service Design Guide at http:/​/​www.cisco.com/​en/​US/​netsol/​ns742/​networking_​solutions_​program_​category_​home.html .

For more information about deployment models for Unified CCE and Cisco Unified Communications, see the latest version of the Cisco Unified Communications Solution Reference Network Design (SRND) Guide at http:/​/​www.cisco.com/​en/​US/​products/​sw/​voicesw/​ps556/​tsd_​products_​support_​series_​home.html.

What’s New in This Chapter

The following topics are new in this chapter or have changed significantly from previous releases of this document.

  • Precision Routing feature support
  • Congestion Control feature support

The new and changed information in this chapter is extensive; read the entire chapter.

General Deployment Options

This section describes options that can apply to many of the specific deployment models listed in the rest of this document. It describes at a high level the tradeoffs that can be made when installing the Unified CCE software.

Agent Peripheral Types

There are two types of Agent Peripherals that can be installed to handle Unified CCE agents. This section discusses the strengths and weaknesses of each.

Enterprise Unified CCE Peripheral

In Enterprise Unified CCE Peripheral deployments, the Unified CCE software treats the VRU and Unified CM as separate peripherals. This means that routing must be done once for each peripheral, and Termination Call Detail records are created for each peripheral each time a call touches the peripheral. Translation routes must be used to send calls between the VRU and Unified CM.

The Unified CM PG and VRU PG can be deployed independently, or the Unified CM and VRU can be deployed in a Generic PG with separate PIMs for Unified CM and VRU.

These deployments provide a large degree of flexibility in configuration. For example, this deployment is capable of using either a Unified CVP or a Unified IP IVR attached to the VRU peripheral and load balancing can be done between multiple IVRs.

These Unified CCE deployments may not be used where Unified CCE is a child to a Unified ICM; a Unified CCE System Peripheral deployment must be used for that solution. For more information, see the section Parent/​Child.

Unified CCE System Peripheral

The Unified CCE System Peripheral combines the functionality of both the VRU peripherals (up to five Unified IP IVR peripherals) and a single Unified CM peripheral together into a single logical Unified ICM peripheral. The Unified CCE treats these Unified IP IVR and Unified CM peripherals as a single peripheral eliminating the need to translation-route calls to the Unified IP IVR for treatment and queuing. If multiple Unified IP IVRs are configured, the Unified CCE System peripheral automatically load-balances calls between the Unified IP IVRs that have available capacity.

Additionally, because the Unified CCE System PG is a single peripheral, Termination Call Detail (TCD) records and other reporting data include the information for the call during the entire time it is on the peripheral. Instead of getting up to three TCDs for each call (one for the original route, one for the IVR, and one for the agent handle time), only a single record is created with the Unified CCE System PG.

The Unified CCE System PG does not support Unified CVP, therefore all queuing and treatment in the Unified CCE System PG is done using Unified IP IVR. Note that a separate Unified CVP on its own PG can be used in conjunction with the Unified CCE System Peripheral.


Note


Precision Routing does not support the Unified CCE System Peripheral.


Unified CCE Administration and Data Server

Several Administration & Data Server deployments with different roles are available. The deployments have varying functionality and can handle varying amounts of reporting data.

Roles

The Administration & Data Servers are classified into the following roles based on the system configuration and the call load that it can handle:

Administration Server (Configuration and Real-Time Reporting)

This role is similar to the former Distributor AW model which provides the capability for configuration changes as well as real-time reporting. The real-time reporting is supported using Cisco Unified Intelligent Center (Reporting client). No historical reporting is supported.

Figure 1. Configuration and Real-Time Reporting

Administration Server (Configuration-Only AW)

In this Administration & Data Server deployment role, the HDS is not enabled and real-time reporting is turned OFF. This distributor deployment provides the capability for configuration changes only. No real-time and historical reporting is supported.

This deployment role allows Contact Center Hosted using CCMP to configure a specific Unified CCE Customer Instance through the ConAPI interface. The load is low enough on such a lightweight Administration & Data Server that a single server is sufficient if deployed using VMware.

Figure 2. Configuration-Only AW

Administration Server, Historical Data Server, and Detail Data Server (AW-HDS-DDS)
This Administration & Data Server deployment role is similar to the existing Distributor AW with HDS model which provides the capability for configuration changes as well as both real-time and historical reporting. The real-time and historical reporting is supported using Cisco Unified Intelligence Center (Unified Intelligence Center Reporting client). Call detail and call variable data are supported for custom reporting data extraction to feed historical data.

Note


Unified Intelligence Center is not part of the out-of-the-box solution.


Figure 3. Administration Server, Historical Data Server, and Detail Data Server (AW-HDS-DDS)

Administration Server And Historical Data Server (AW-HDS)
This Administration & Data Server deployment role provides the capability for configuration changes as well as both real-time and historical reporting. Real-time and historical reporting are supported using Cisco Unified Intelligence Center Reporting user.

Note


Cisco Unified Intelligence Center is not part of the out-of-the-box solution.


In addition, the following features are disabled and not supported:
  • Call Detail, Call Variable, and Agent State Trace data
  • Custom reporting data extraction
Figure 4. Administration Server And Historical Data Server (AW-HDS)

Historical Data Server and Detail Data Server (HDS-DDS)

The HDS-DDS deployment model is used specifically for data extraction and for custom reports for call detail (TCD and RCD) only.

In addition, the following features are disabled and not supported:
  • Real-time data reporting
  • Ability to make configuration changes

This deployment role is limited to one per Logger side.

Figure 5. Historical Data Server And Detail Data Server (HDS-DDS)

Deployment Options

The following deployment options are based on the varying capacity in terms of number of agents, call load, number of reporting users, and other operating conditions mentioned in the Sizing Unified CCE Components and Servers chapter.

Option 1: Less than 4000 Agents

This deployment option is applicable to the following limitations:
  • Sizing within the operating conditions listed in the Sizing Unified CCE Components and Servers chapter in the SRND and the Hardware and System Software Specification (Bill of Material) for Cisco Unified ICM/Contact Center Enterprise and Hosted specification
  • A maximum of 4,000 active agents
  • A maximum of 400 reporting users

The number of HDSs is limited to two AW-HDS-DDSs per Logger side.

Option 2: Greater than 4000 Agents

This deployment option is applicable to the following limitations:
  • Sizing within the operating conditions listed in the Sizing Unified CCE Components and Servers chapter in the SRND and the Hardware and System Software Specification (Bill of Material) for Cisco Unified ICM/Contact Center Enterprise and Hosted specification
  • More than 4,000 active agents
  • More than 400 reporting users

The number of Administration & Data Server roles is limited to three AW-HDSs and 1 HDS-DDS or AW-HDS-DDS per Logger side.

Cisco Unified Intelligence Center (Reporting client) is supported but not part of the out-of-the-box solution.

Parent/Child

The Unified CCE Gateway PG allows Unified CCE or Unified CCX to appear as a traditional ACD connected to the Unified ICM system. The Unified CCE Gateway PG does this by providing a PG to the Unified ICM system that communicates to the CTI interface of Unified CCE System PG or Unified CCX.

When the Unified CCE Gateway PG is used in a deployment, its relationship to Unified ICM is termed a parent and Unified CCE is called the child:
Parent
The Unified ICM system that serves as the network or enterprise routing point. The child looks like an ACD to the parent, which uses the appropriate Unified CCE Gateway PG (Enterprise or Express) to communicate to the CTI interface on the child Unified CCE. The parent can perform all functions that a Unified ICM can usually perform, including pre- and post-routing and end-to-end call tracking using translation routes.
Child
The Unified CCE System PG or Unified CCX system that is set up to function as an ACD. The child can receive calls that are translation-routed from the parent, but it is not aware of any other peripherals attached to the parent. The child can also post-route calls from the Unified CCE to the parent where the call can be handled like any other Unified ICM call. For example, the call can be translation-routed to any TDM or IP ACD controlled by the Unified ICM or queued in the Unified ICM network queue point with Unified CVP.

Note


Precision Routing is not supported in a parent/child deployment.


In the parent/child model, the child Unified CCE is configured to function completely on its own and does not need the connection to the parent to route calls to agents. This independence provides complete local survivability for mission-critical contact centers if the network between the child and parent goes down or if there is a problem with the parent or the Unified CCE Gateway PG connection.

Configuration objects entered into the child system can automatically be sent to the parent Unified ICM and inserted into the Unified ICM configuration, thus eliminating the need to configure objects twice (once in the local ACD and again to match the configuration in the Unified ICM itself for routing and reporting). This functionality can also be turned off for situations where the customer does not want automatic configuration updates, such as with an outsourcer using the Unified CCE child system where not all of the agents, skill groups, and call types on that child system apply to the customer’s Unified ICM system.

The Unified CCE Gateway PG can connect to a Unified CCE child that is using the Unified CCE System PG or to Unified CCX. If the Unified CCE child has multiple Unified CCE System PGs and peripherals, a separate Unified CCE Gateway PG peripheral must be installed and configured for each one in the Unified ICM parent system. When deployed on a separate server, a Unified CCE Gateway PG can manage multiple child Unified CCE peripherals or multiple child Unified CCX systems (up to five child systems).


Note


The Unified CCE Gateway PG does not support Unified CCE System PG and Unified CCX integration on the same CCE Gateway PG instance.


In the Unified CCE child, you can deploy IP IVR or Unified CVP for call treatment and queuing. If you deploy Unified CVP, you must configure an additional VRU PG. This model does not follow the single peripheral model used when IP IVR is deployed. For this reason, information about calls queued at the child (and queue time of a call) is not available on the parent, so any computation involving queue time (such as minimum expected delay (MED) and average answer wait time) are inaccurate.

Special Note on Whisper Announcement and Agent Greeting for Parent/Child Systems

Beginning with Unified CCE Release 8.5(2), the Whisper Announcement and Agent Greeting feature are supported on a child Unified CCE.

Agent Greeting for Parent/Child is only supported for a very specific Parent/Child configuration, where calls are queued at a CCX (IP-IVR) on the child system PG, and requires a dedicated CVP at the child on a dedicated VRU PG to provide the agent greetings. Agent Greeting in Parent/Child configurations must be approved by the Cisco Assessment to Quality (A2Q) process and requires Cisco's Design Mentoring Service to assure that the deployment model is designed and sized correctly to support the Agent Greeting feature.

To deploy the child Unified CCE, you must complete the following:
  • Use IP-IVR on the child for call treatment, queuing, and Whisper Announcement.
  • Use CVP on the child only for Agent Greeting. Do not use CVP for call queuing. You must use a separate VRU PG for the CVP.

Special Note on Network Consultative Transfer for Parent/Child Systems

One restriction of the parent/child deployment is that calls terminating on child systems cannot be transferred by network consultative transfer (NCT) through the NIC or any other routing client on the parent. Although NCT works for TDM ACDs, and at first glance the parent/child deployment seems virtually identical in architecture, the parent/child deployment is not the same.

For a TDM PG, the CTI Server is connected to the ACD PG, which is part of the parent system. This is the equivalent of having a CTI-Server connected to the Gateway PG. To think of it another way, it is like using CTI directly to an ACD instead of the CTI Server, in which case network consultative transfer is not possible either. In parent/child deployments, CTI is connected to the child PG. Having CTI connected to the child PG does not provide the necessary network call ID and other information necessary to allow network consultative transfer.


Note


Network blind transfer is still possible using any client (for example, Unified CVP or a NIC) on the parent system when a post route is initiated to the parent system from the child.


SIP Support

Unified IP IVR is notified of caller entered digits (DTMF input) by way of JTAPI messages from Unified CM. Unified IP IVR and Unified CM do not support mechanisms to detect in-band DTMF digits. In deployments with SIP Voice Gateways or SIP phones that support only in-band DTMF (or are configured to use in-band DTMF In accordance with RFC 2833), Unified CM must invoke an MTP resource to convert the in-band DTMF signaling to out-of band signaling so that the Unified IP IVR can be notified of caller entered digits. Therefore, in environments that include SIP phones or gateways, it is necessary to provision sufficient MTP resources. Keep this in mind if the phones need to interact with Unified IP IVR. Likewise, CTI ports do not support in-band DTMF (RFC 2833). The Mobile Agent feature relies on CTI ports, so MTP resources are required when in-band DTMF (RFC 2833) is negotiated.

SIP trunking is supported using CVP deployments with Cisco IOS gateways (IOS GW) and Unified Border Elements (CUBE). For more information about deployment models for Unified CCE and Cisco Unified Communications, see the latest version of the Cisco Unified Communications Solution Reference Network Design (SRND) Guide at http:/​/​www.cisco.com/​en/​US/​products/​sw/​voicesw/​ps556/​tsd_​products_​support_​series_​home.html.

Q.SIG Support

Cisco Unified CCE does not support using Q.SIG trunks with the Unified CM deployment.

IPv6 Support

Unified CCE 8.0 supports interoperability with an IPv6-enabled Unified CM cluster. All the Unified CCE components run with IPv4 enabled including IP-IVR, Unified CVP, the CTI OS Agent Desktops, and Agent Phones. The Unified CCE Agent PG integration with the Unified CM CTI Manager uses IPv4.

Caller phones or Voice Gateways can run IPv4 or IPv6. If the caller's environment is IPv6 only, then Media Termination Point (MTP) resources are required for call treatment using IP-IVR and Unified CVP VXML Gateways.

Agent phones must run IPv4; IPv6 is not supported for agent phones. If the caller’s phone is IPv6, then an MTP must also be inserted between the resources.

Mobile Agent and Outbound Option endpoints (CTI ports and Dialer ports) must be configured as IPv4 devices.

Service Advertisement Framework Call Control Discovery

The Service Advertisement Framework Call Control Discovery (SAF CCD) feature is a way to replace the need for gatekeepers and SIP proxies. The JTAPI impact has a new reason code for when calls are redirected from another cluster back to the local PSTN. For more details, see the Cisco Unified Communications SRND.

Cisco Unified Mobile Agent

For deployments using Cisco Unified Mobile Agent, it is important to consider the location of the Voice Gateways that is used to call agents because their location has design considerations for Silent Monitoring, call admission control, and other areas. For design guidance and considerations for implementing Cisco Unified Mobile Agent, see Cisco Unified Mobile Agent.

CAD Multi-server Support

Multiple instances of CAD services are supported against a single PG. This deployment model requires that the CAD services reside on separate servers (such as when supporting different languages), but the model does not require a separate PG for each one. Up to 10 instances of CAD services are allowed per PG.

The benefit of this type of deployment is that the additional instances of CAD services required to achieve a specific agent number do not require a separate PG. There is, however, a limit of 1000 CAD agents per PG, regardless of the number of multiple instances of CAD. While the separate instances of CAD services do have to be deployed on separate servers, the hardware requirements for CAD services are minimal. For information about those requirements, see the Cisco CAD Installation Guide, Release 9.0 .

Cisco Finesse Multi-Server Support

Two instances of Finesse are supported on a single PG, with each Finesse server capable of supporting the maximum of 2,000 users supported by the CTI Server. Each Finesse server requires a single CTI connection. See the next figure.

Figure 6. Multiple Finesse Servers

Cisco Finesse must be deployed as a virtual machine in a VMware virtual environment. Finesse can be deployed according to coresidency policies outlined in the Unified Communications Virtualization DocWiki (http:/​/​docwiki.cisco.com/​wiki/​Unified_​Communications_​Virtualization).

Finesse can be deployed with CTI-OS with the following limitations:
  • The maximum number of CTI all events connections supported by the CTI Server (7) is not exceeded.
  • The total number of combined Finesse and CTI OS agents does not exceed the capacity of the common PG.

Virtualization Support

For detailed information about Unified Communications applications running in virtual machines on UCS hardware, please see the Unified Communications Virtualization DocWiki page at: http:/​/​docwiki.cisco.com/​wiki/​Unified_​Communications_​Virtualization.

Precision Routing Support

Beginning in Unified CCE 9.0(1), Precision Routing is available as an alternative and enhanced routing strategy. The Precision Routing feature targets agents based on a set of attributes or competencies that agents possess rather than a set of fixed skill group associations. The set of agent attributes represent the abilities that an agent has, for example, the ability to speak a language at a specific proficiency level, their skill in selling a product, or the fact that they reside in a particular location. These are all attributes which can then be targeted in a precision queue.

Limitations

Precision Routing is currently available for voice agents only, and does not support multimedia. Precision Routing is available only for the unified CCE PG, and does not support the Unified System PG.

Precision Routing is not supported in a parent/child deployment.

Precision Routing is not supported for Cisco Outbound Option. However, agents who participate in an outbound campaign (through an outbound skill group) can also be logged into a Precision Queue to handling inbound calls.

Whisper Announcement Support

Whisper Announcement is defined as the ability to play a pre-recorded announcement to an Agent right before the caller is connected. The announcement is played to the Agent only. The announcement is not heard by the Customer. Only one Whisper Announcement file plays for each call.

Whisper Announcement is supported with Enterprise deployments of Unified CCE only. The feature operates with a type 10 Network VRU based on CVP with SIP as the CVP call control protocol. H323 is not supported.

Whisper time is not specifically categorized in Unified CCE reports. In agent and skill group reports, the period during which the announcement plays is reported as reserved agent state time. In the Termination Call Detail records, it is treated as ring time.

Whisper Announcement for Parent/Child is only supported for a very specific Parent/Child configuration, where calls are queued and whisper announcement are sourced by a CCX (IP-IVR) on the child system PG, and requires a dedicated CVP at the child on a dedicated VRU PG to provide agent greetings if so desired. W hisper Announcement in Parent/Child configurations must be approved by the Cisco Assessment to Quality (A2Q) process and requires Cisco Design Mentoring Service to assure that the deployment model is designed and sized correctly to support the Whisper Announcement feature.

Agent Greeting Support

Agent Greeting is the capability for a contact center agent to record a greeting and then automatically play it to the caller and agent at the same time when a new call is received by the agent. The agent greeting playback is immediately followed by a caller and agent connection in a call.

Only one Agent Greeting file plays per call. Agent Greeting is supported with Enterprise deployments of Unified CCE only. The feature operates with a type 10 Network VRU based on CVP. SIP is the only CVP call control protocol supported (not H.323).

Agent Greetings are recorded using g.711 encoding. CVP supports a mixed codec environment in which the IVR uses g.711 encoding while customer/agent calls can use g.729. g.722 is not supported.

Greeting time is not specifically categorized in Unified CCE reports. In agent and skill group reports, the period during which the greeting plays is reported as talk time. Record time is counted as an internal call by the default skill group.

Agent Greeting is supported for in-house (that is, non-mobile) agents in Release 8.5(1). Release 8.5(2) adds support for mobile agents. It works with Unified CM supported IP phones with BIB. See the Unified CCE Compatibility Matrix for the list of supported phones.

The CVP VXML server is required for recording.

Agent Greeting Deployment Models

For the agent greeting call leg, the Agent PIM sends a separate route request to the router when the agent answers the call. The PIM instructs the UCM to connect to the VRU to add media to the call and then the VRU plays the agent greeting based on the instruction from the router. This extra call leg adds extra load on the system and more capacity needs to be estimated when sizing the system. Special attention needs to be paid for this call leg when deploying systems with the agent greeting feature.

One Network VRU is used for call treatment, call queuing, and playing agent greeting

In this deployment model, one network VRU is configured in the Unified CCE and this VRU is used for playing prompts, call queuing, and playing or recording agent greeting. The SendToVRU node can be used in the routing script for the agent greeting call leg.

For small deployments where one CVP server is used for call queuing and treatment, a SIP trunk can be configured in UCM to directly send the call to the CVP to play the agent greeting.

For those deployments where more than one CVP servers are needed for call queuing and treatment,  each CVP server is associated with a VRU PIM and all of those VRU PIMs are associated with the same network VRU.

A proxy server can be configured between the UCM and CVP servers to balance the load among those CVP servers for the agent greeting call leg. See the following figure.

Figure 7. Single-Site Deployment Model for Agent Greeting

Dedicated Network VRU for agent greeting

This is supported in Unified CCE Release 8.5(2) and later. For details, see the at http:/​/​www.cisco.com/​en/​US/​products/​sw/​custcosw/​ps1844/​products_​feature_​guides_​list.html.

Special requirements for multiple sites deployment
As the agent greeting is played to both the caller and agent, excessive network latency may cause the caller to hear silence before the greeting is played. The following requirements must be met when the system is deployed at multiple sites:
  • The latency between the Unified CCE Central Controllers and remote PGs (both Agent PG and VRU PG) cannot exceed 50 ms one way (100 ms round trip).
  • When multiple CVP servers are deployed for the call load, it is possible that the call is queued at one CVP and Agent Greeting is played by a different CVP at another site. For better performance, it is ideal to have the Agent Greeting played by the VXML Voice Gateway which is closest to the agent phone.

IPT: Single Site

A single-site deployment refers to any scenario where all Voice Gateways, agents, desktops, phones, and call processing servers (Unified CM, Unified ICM, Unified CCE, and Unified IP IVR or Cisco Unified Customer Voice Portal (Unified CVP)) are located at the same site and have no WAN connectivity between any Unified CCE software modules.

Figure 8. Single-Site Deployment

This figure shows a Unified IP IVR, a Unified CM cluster, redundant Unified CCE servers, two Administration & Data Servers, and a direct connection to the PSTN from the Voice Gateways.

The Unified CCE server in this scenario is running the following major software processes:
  • Call Router
  • Logger and Database Server
  • Unified CCE System PG with Unified CM Peripheral Interface Manager (PIM) and Unified IP IVR PIM
  • CTI Server
  • CTI Object Server (CTI OS)
  • Optionally, Cisco Agent Desktop (CAD) servers can be co-located on the Unified CCE servers as well.

Optionally, the Central Controller and Unified CCE System PG can be split onto separate servers. For information about when to install the Central Controller and PG on separate servers, see Sizing Unified CCE Components and Servers.

Traditional Unified CCE must be deployed in a redundant fashion. Simplex deployments are supported only for lab or non-production deployments. For information about Unified CCE redundancy, see Design Considerations for High Availability.

The number of Unified CM nodes and the hardware model used is not specified along with the number of Unified IP IVRs. For information about determining the number and type of servers required, see Sizing Unified CCE Components and Servers.

Also not specified in this model is the specific data switching infrastructure required for the LAN, the type of Voice Gateways, or the number of Voice Gateways and trunks. Cisco campus design guides and Cisco Unified Communications design guides are available to assist in the design of these components. Sizing Contact Center Resources discusses how to determine the number of gateway ports.

Another variation of this model is to have the Voice Gateways connected to the line side of a PBX instead of the PSTN. Connection to multiple PSTNs and a PBX all from the same single-site deployment is also possible. For example, a deployment can have trunks from a local PSTN, a toll-free PSTN, and a traditional PBX/ACD. For more information, see Traditional ACD Integration and Traditional IVR Integration.

This deployment model also does not specify the type of signaling (ISDN, MF, R1, and so on) to be used between the PSTN and Voice Gateway, or the specific signaling (H.323, SIP or MGCP) to be used between the Voice Gateway and Unified CM.

The amount of digital signal processor (DSP) resources required for placing calls on hold, consultative transfers, and conferencing is also not specified in this model. For information about sizing of these resources, see the latest version of the Unified Communications System Solution Reference Network Design (SRND) at http:/​/​www.cisco.com/​en/​US/​docs/​voice_ip_comm/​uc_system/​design/​guides/​UCgoList.html.

The main advantage of the single-site deployment model is that there is no WAN connectivity required. Given that there is no WAN in this deployment model, there is generally no need to use g.729 or any other compressed Real-Time Transport Protocol (RTP) stream (so transcoding is not required).

Unified CCE: Unified CCE System PG

In this deployment model, the agent PG that is deployed is a Unified CCE System PG. Only a single peripheral is needed to handle both the Unified CM and any Unified IP IVRs that may exist. This peripheral unifies the appearances of the multiple PIMs and also handles the load balancing of calls between multiple Unified IP IVRs. Alternatively, this model may be configured to use Unified CVP. When Unified CVP is used, its connectivity to Cisco Unified Presence handles load balancing by distributing the incoming calls among Unified CVP Call Servers. In this deployment, the VRU PIMs (up to 10) communicating with the Unified CVP Call Servers reside on their own PG and not under the Unified CCE System PG. The following figure shows a single-site deployment using Unified CVP instead of IP-IVR in a Unified CCE system. In this model, no longer do all calls reside under a single peripheral; Unified CVP is its own peripheral.

Figure 9. Single-Site Deployment Using Unified CVP

When using this configuration, the VRU PGs are deployed in a duplex PG configuration with one to ten Unified CVP Call Servers (depending on the number of agents and call volume required).

IVR: Treatment and Queuing with Unified IP IVR

In this deployment model, all initial and subsequent queuing is done on the Unified IP IVR. Up to five Unified IP IVRs can be deployed in this model (with the Unified CCE System PG). The Unified IP IVRs are placed behind Unified CM, using the Unified CM dial plan and call switching under control of Unified CCE. All calls come into a CTI Route Point on Unified CM, controlled by Unified CCE, and are then automatically translation-routed to the Unified IP IVR by the Unified CCE System PG. The Unified CCE handles load balancing between available Unified IP IVR ports. Configuring translation routes between the Unified IP IVR and Unified CM is not needed.

IVR: Treatment and Queuing with Unified CVP

Unified CVP can be used to provide the call treatment and queuing in this model as well. Because Unified CVP is not part of the Unified CCE System PG peripheral, translation routes must be configured to transfer calls with call data between the peripherals.

In this deployment model, all initial and subsequent queuing is done using Unified CVP. A single server may be used with all Unified CVP processes co-located on that server. Multiple servers, on the other hand, allow scaling and redundancy. For more information about redundancy, see Design Considerations for High Availability.

For more information about Unified CVP, see the Cisco Unified Customer Voice Portal Solution Reference Network Design (SRND).

Unified CCE: Enterprise Unified CCE PG

In these deployment models the Enterprise Unified CCE peripheral is used to handle interactions with Unified CM; a separately configured VRU peripheral is used to handle interactions with the Unified IP IVR or Unified CVP.

IVR: Treatment and Queuing with Unified IP IVR

In this deployment model, all initial and subsequent queuing is done on the Unified IP IVR. If multiple Unified IP IVRs are deployed, use Unified CCE to load-balance calls across those Unified IP IVRs. Translation routes must be configured manually between the Unified CM peripheral and the Unified IP IVR peripherals. The translation routes are used to move calls and data between Unified CM and the Unified IP IVRs. Load balancing is done manually in the Translation Route To VRU node in the Unified CCE call routing script.

IVR: Treatment and Queuing with Unified CVP

Unified CVP can be used to provide the call treatment and queuing in this model as well. Unified CVP has its own VRU PG, either loaded on the same Server As the Unified CM PG or as part of a Generic PG combination.

In this deployment model, all initial and subsequent queuing is done using Unified CVP. A single server may be used with all Unified CVP processes co-located on that server. Multiple servers, on the other hand, allow scaling and redundancy. For more information about redundancy, see Design Considerations for High Availability.

For more information about Unified CVP, see the Cisco Unified Customer Voice Portal Solution Reference Network Design (SRND).

Unified CCE: Transfers

In this deployment model (as well as in the multisite centralized call processing model), both the transferring agent and target agent are on the same peripheral. This also implies that both the routing client and the peripheral target are the same peripheral. The transferring agent generates a transfer to a particular dialed number configured as a CTI Route Point in Unified CM (for example, looking for any specialist in the specialist skill group).

The Agent peripheral (either the Unified CCE System peripheral or the Enterprise Unified CCE peripheral) generates a route request to the Call Router. The Call Router matches the dialed number to a call type and activates the appropriate routing script. The routing script looks for an available specialist.

If a target agent (specialist) is available to receive the transferred call, the Call Router returns the appropriate label based on the Agent Target Rules (Dynamic) or a DeviceTarget Label (Static) to the requesting routing client (the Agent peripheral). In this scenario, the label is typically just the extension of the phone where the target agent is currently logged in. On receiving the route response (label), the Unified CM PIM initiates the transfer by sending a JTAPI transfer request to Unified CM.

At the same time that the label is returned to the routing client, pre-call data (which includes any call data that has been collected for this call) is delivered to the peripheral target. In this scenario, the routing client and peripheral target are the same Agent peripheral. This is because the transferring agent and the target agent are both associated with the same peripheral. In some of the more complex scenarios to be discussed in later sections, the routing client and peripheral target are not the same.

If a target agent is not available to receive the transferred call, the Call Routing script is typically configured to transfer the call to an IVR so that queue treatment can be provided. In this scenario, the logic in the Unified CCE System PG differs from the logic in the Unified CCE PG if the IP-IVR variant is used.

In both cases, the label is a dialed number that instructs Unified CM to transfer the call to an IVR. The translation-route or correlationID is not needed when using the Unified CCE System peripheral but is needed when deploying Unified CVP.

IPT: Multisite with Centralized Call Processing

A multisite deployment with centralized call processing refers to any scenario where call processing servers (Unified CM, Unified CCE, and Unified IP IVR or Unified CVP) are located at the same site, while any combination of Voice Gateways, agents, desktops, and phones are either located centrally or remotely across a WAN link.

There are two variations of this IPT model:
  • IPT: Centralized Voice Gateways
  • IPT: Distributed Voice Gateways

IPT: Centralized Voice Gateways

If an enterprise has small remote sites or offices in a metropolitan area where it is not efficient to place call processing servers or Voice Gateways, then this model is most appropriate. As sites become larger or more geographically dispersed, use of distributed Voice Gateways might be a better option.

The following figure illustrates this model using a Unified CCE deployment. The figure shows the deployment using IP-IVR, but it can also use Unified CVP instead of IP-IVR.

Figure 10. Multisite Deployment with Centralized Call Processing and Centralized Voice Gateways

Advantages

  • Only a small data switch and router, IP phones, and agent desktops are needed at remote sites where only a few agents exist. Only limited system and network management skills are required at remote sites.
  • No PSTN trunks are required directly into these small remote sites and offices, except for local POTS lines for emergency services (911) in the event of a loss of the WAN link.
  • PSTN trunks are used more efficiently because the trunks for small remote sites are aggregated.
  • Unified CCE Queue Points (Unified IP IVR or Unified CVP) are used more efficiently because all Queue Points are aggregated.
  • No VoIP WAN bandwidth is used while calls are queuing (initially or subsequently). Calls are extended over the WAN only when there is an agent available for the caller.

As with the single-site deployment model, all the same options exist when using Unified CCE configurations. For example, multisite deployments can run the Unified CCE software all on the same server or on multiple servers. The Unified CCE software can be deployed either with the Unified CCE System PG or the Unified CCE PG. The number of Unified CM and Unified IP IVR or Unified CVP servers is not specified by the deployment model, nor are the LAN/WAN infrastructure, Voice Gateways, or PSTN connectivity. For other variations, see IPT: Single Site.

Best Practices

  • VoIP WAN connectivity is required for RTP traffic to agent phones at remote sites.
  • RTP traffic to agent phones at remote sites might require compression to reduce VoIP WAN bandwidth usage. It may be desirable for calls within a site to be uncompressed, so transcoding might also be required depending on how the Cisco Unified Communications deployment is designed.
  • Skinny Client Control Protocol (SCCP) or SIP call control traffic from IP phones to the Unified CM cluster flows over the WAN.
  • CTI data to and from the Unified CCE Agent Desktop flows over the WAN. Adequate bandwidth and QoS provisioning are critical for these links.
  • Because there are no Voice Gateways at the remote sites, customers might be required to dial a long-distance number to reach what is normally a local PSTN phone call if Voice Gateways with trunks were present at the remote site. This situation is mitigated if the business requirements are to dial toll- free numbers at the central site. An alternative is to offer customers a toll-free number to dial and have those calls all routed to the centralized Voice Gateway location. However, this requires the call center to incur toll-free charges that can be avoided if customers had a local PSTN number to dial.
  • The lack of local Voice Gateways with local PSTN trunks can also impact access to 911 emergency services; this must be managed through the Unified CM dial plan. In most cases, local trunks are configured to dial out locally and for 911 emergency calls.
  • Unified CM location-based call admission control failure results in a routed call being disconnected. Therefore, it is important to provision adequate bandwidth to the remote sites. Also, an appropriately designed QoS WAN is critical.

    Note


    For calls controlled by Unified CVP, call admission control failures can be recovered with the proper Unified CCE Scripting configuration with the Unified CCE Router Re-query feature enabled.


  • Automated Alternate Routing (AAR) provides a mechanism to reroute calls through the PSTN or other network by using an alternate number when Unified CM blocks a call due to insufficient location bandwidth. For deployments with Unified CVP ingress call control, do not enable AAR. This allows Unified CVP Router Re-query to take control of the call in the event of a failure or timeout. For deployments using IP-IVR, enable AAR to route the call through the PSTN or other network component using an alternative number.

IVR: Treatment and Queuing with Unified IP IVR

As in the single-site deployment, all call queuing is done on the Unified IP IVR at a single central site. While calls are queuing, no RTP traffic flows over the WAN. If re-queuing is required during a transfer or reroute on ring-no-answer, the RTP traffic flow during the queue treatment also does not flow over the WAN. This reduces the amount of WAN bandwidth required to the remote sites.

IVR: Treatment and Queuing with Unified CVP

In this model, Unified CVP is used in the same way as Unified IP IVR.

Unified CCE: Transfers

Transfers in this scenario are, from the point of view of the contact center, the same as in the single-site scenario. Therefore, the same call and message flows occur as in the single-site model, whether the transferring agent is on the same LAN as the target or on a different LAN. The only differences are that QoS must be enabled and that appropriate LAN and WAN routing must be established. For details on provisioning your WAN with QoS, see the latest version of the Cisco Network Infrastructure Quality of Service Design Guide.

During consultative transfers where the agent (not the caller) is routed to a Unified IP IVR port for queuing treatment, transcoding is required because the Unified IP IVR can generate only g.711 media streams.

IPT: Distributed Voice Gateways

A variation of the centralized call processing model can include multiple ingress Voice Gateway locations. This distributed Voice Gateway model might be appropriate for a company with many small sites, each requiring local PSTN trunks for incoming calls. This model provides local PSTN connectivity for local calling and access to local emergency services.

Figure 11. Multisite Deployment with Centralized Call Processing and Distributed Voice Gateways

In this deployment model, shown with Unified IP IVR for queuing and treatment, it might be desirable to restrict calls arriving at a site to be handled by an agent within that site, but this is not required. By restricting calls to the site where it arrived, the following conditions apply:
  • VoIP WAN bandwidth is reduced for calls going to agents from the ingress Voice Gateway.
  • Calls still cross the VoIP WAN during the time they are in a queue or are receiving treatment from the centralized Unified IP IVRs.
  • Customer service levels for calls arriving into that site might suffer due to longer queue times and handling times.
  • Longer queue times can occur because even though an agent at another site is available, the Unified CCE configuration may continue to queue for an agent at the local site.
  • Longer handling times can occur because even though a more qualified agent exists at another site, the call may be routed to a local agent to reduce WAN bandwidth usage.

To restrict a call to the site at which it arrived in this deployment model, you must create separate skill groups for agents at each location. To route a call to any agent in a given skill group regardless of location, you can use an enterprise skill group to combine the location-specific skill groups.

When using Precision Routing, to target agents which are local to a particular site you can use a location based attribute within a Precision Queue (as opposed to using location specific skill groups).

It is important for deployment teams to carefully assess the trade-offs between operational costs and customer satisfaction levels to establish the right balance on a customer-by-customer basis. For example, it may be desirable to route a specific high-profile customer to an agent at another site to reduce their queue time and allow the call to be handled by a more experienced representative, while another customer may be restricted to an agent within the site where the call arrived.

A Unified CCE deployment may actually use a combination of centralized and distributed Voice Gateways. The centralized Voice Gateways can be connected to one PSTN carrier providing toll-free services, while the distributed Voice Gateways can be connected to another PSTN carrier providing local phone services.

Inbound calls from the local PSTN can be both direct inward dial (DID) and contact center calls. It is important to understand the requirements for all inbound and outbound calling to determine the most efficient location for Voice Gateways. Important details include identifying who is calling, why they are calling, where they are calling from, and how they are calling.

CVP Call Treatment and Call Routing

In multisite environments with distributed Voice Gateways, Unified CVP can be used to leverage the ingress Voice Gateways at the remote sites as part of the Unified CCE system. Unified CVP treats and queues calls locally in the ingress Voice Gateway rather than requiring the call to cross the VoIP WAN to a centralized queue platform. Unified CVP provides call treatment (VRU) using the VoiceXML Browser built into the Cisco IOS Voice Gateway. Only call signaling passes over the WAN to instruct the remote site Voice Gateway how to treat, queue, and transfer the call to an agent. In these models, pre-routing to the site might not be necessary because Unified ICM/CCE takes control of the call as soon as it arrives at the site. Basic carrier percent allocation can be used to allocate calls to the sites and fail-over (rollover) trunks can be used to address local failures as needed.

Traditional Pre-routing

In a traditional deployment with Unified ICM (parent and child or hybrid) with multisite deployments and distributed Voice Gateways, the Unified ICM pre-routing capability can also be used to load-balance calls dynamically across the multiple sites. For a list of PSTN carriers that offer Unified ICM pre-routing services, see the Pre-installation Planning Guide for Cisco ICM Enterprise & Hosted Editions.

In multisite environments in which Voice Gateways have both local PSTN trunks and separate toll-free trunks that deliver contact center calls, the Unified ICM pre-routing software can load-balance the toll-free contact center calls around the local contact center calls. For example, suppose you have a two-site deployment where Site 1 currently has all agents busy and there are many locally-originating calls in the queue. Site 2 has only a few calls in the queue or maybe even a few agents are currently available. In this scenario, configure the Unified ICM to instruct the toll-free provider to route most or all of the toll-free calls to Site 2. This type of multisite load balancing provided by Unified ICM is dynamic and automatically adjusts as call volumes change at all sites.

Just as in the two previous deployment models, much variation exists in the number and types of Unified ICM/CCE, Unified CM, and Unified IP IVR or Unified CVP servers, LAN/WAN infrastructure, Voice Gateways, PSTN connectivity, and so forth.

Advantages of Using Distributed Voice Gateways

  • Only limited systems management skills are needed for the remote sites because most servers, equipment, and system configurations are managed from a centralized location.
  • Unified CVP or Unified ICM/CCE pre-routing can be used to load-balance calls across sites, including sites with local PSTN trunks in and toll-free PSTN trunks.
  • No WAN RTP traffic is required for calls arriving at each remote site that are handled by agents at that remote site.
  • Unified CVP provides call treatment and queuing at the remote site using the VoiceXML Browser in Cisco IOS on the Voice Gateway itself, thus eliminating the need to move the call over the VoIP WAN to a central queue and treatment point.

Best Practices

  • The Unified IP IVR or Unified CVP, Unified CM and PGs (for both Unified CM and IVR or Unified CVP) are co-located. In this model, the only Unified CCE communications that can be separated across a WAN are the following:
    • Unified Central Controller to PG
    • PG to Unified CCE Agent Desktops
    • Unified CM to Voice Gateways
    • Unified CM to phones
    • Unified CVP Call Control Server to remote Voice Gateway (call control)
  • If calls are not going to be restricted to the site where calls arrive, or if calls are made between sites, more RTP traffic flows across the WAN. It is important to determine the maximum number of calls that flow between sites or locations. Unified CM locations-based call admission control failure results in a routed call being disconnected (rerouting within Unified CM is not currently supported). Therefore, it is important to provision adequate bandwidth to the remote sites, and appropriately designed QoS for the WAN is critical. Calls that are treated by IP IVR at the central site must also be considered.

    Note


    For calls controlled by Unified CVP, call admission control failures can be recovered with the proper Unified CCE Scripting configuration with the Unified CCE Router Re-query feature enabled.


  • H.323, SIP, or MGCP signaling traffic between the Voice Gateways and the centralized Unified CM servers flows over the WAN. Proper QoS implementation on the WAN is critical and signaling delays must be within tolerances listed in the latest version of the Cisco Unified Communications System Solution Reference Network Design (SRND) Guide at http:/​/​www.cisco.com/​en/​US/​docs/​voice_ip_comm/​uc_system/​design/​guides/​UCgoList.html.

Unified CCE: Unified CCE System PG

Because the deployment of contact center components is essentially the same as in other multisite centralized call processing deployments, the same benefits and restrictions apply to Unified CCE deployed using the Unified CCE System PG.

Additionally, if Unified ICM/Unified CCE pre-routing is used to interact with carriers and distribute calls to the Voice Gateways, translation routes for the NIC routing client to the Unified CCE System PG must be configured manually using the ConfigManager application on the Unified ICM/Unified CCE Admin Workstation.

Unified CCE & IVR: Treatment and Queuing with Unified IP IVR

WAN bandwidth must be provisioned to support all calls that are treated and queued at the central site.

IVR: Treatment and Queuing with Unified CVP

Unified CVP is not supported with a Unified CCE System PG. A separate VRU peripheral must be configured and deployed. This means that translation routes must be configured to transfer calls with call data between the peripherals. However, Unified CVP does provide the benefits of queuing and treatment for callers at the remote distributed ingress Voice Gateways in this model because the calls do not have to cross the VoIP WAN for treatment in the centralized Unified IP IVR.

Using Unified CVP for treatment and queuing allows you to reduce the amount of voice bearer traffic traveling across the WAN. Unified CVP queues and treats calls on the remote gateways, thus eliminating the need to terminate the voice bearer traffic at the central site. WAN bandwidth must still be provisioned for transfers and conferences that involve agents at other locations.

Unified CCE: Unified CCE PG

Because the deployment of contact center components is essentially the same as in other multisite centralized call processing deployments, the same benefits and restrictions apply to Unified CCE deployed using the Unified CCE PG.

Additionally, if Unified ICM/CCE pre-routing is used to interact with carriers and to distribute calls to the Voice Gateways, translation routes must be configured for the NIC routing client using traditional Unified CCE with separate Unified CVP and Unified CM peripherals in the Unified ICM/CCE.

IVR: Treatment and Queuing with Unified IP IVR

WAN bandwidth must be provisioned to support all calls that are treated and queued at the central site.

IVR: Treatment and Queuing with Unified CVP

Using Unified CVP for treatment and queuing allows you to reduce the amount of voice bearer traffic traveling across the WAN. Unified CVP queues and treats calls on the remote gateways, thus eliminating the need to terminate the voice bearer traffic at the central site. WAN bandwidth must still be provisioned for transfers and conferences that involve agents at other locations.

Unified CCE: Transfers

Intra-site or inter-site transfers using the VoIP WAN to send the RTP stream from one site to another occur basically the same way as a single-site transfer or a transfer in a deployment with centralized Voice Gateways.

An alternative to using the VoIP WAN for routing calls between sites is to use a carrier-based PSTN transfer service. These services allow the Unified CCE Voice Gateways to out-pulse DTMF tones to instruct the PSTN to reroute (transfer) the call to another Voice Gateway location. Each site can be configured within the Unified ICM/CCE as a separate Agent Peripheral. The label then indicates whether a transfer is intra-site or inter-site using Take Back and Transfer (*8) or Transfer Connect. These transfer tones are played in-band over the voice path and must be played from a recorded file in Unified IP IVR or out-pulsed as digits from Unified CVP.

IPT: Multisite with Distributed Call Processing

Enterprises with multiple medium to large sites separated by large distances tend to prefer a distributed call processing model. In this model, each site has its own Unified CM cluster, treatment and queue points, PGs, and CTI Server. However, as with the centralized call processing model, you can deploy sites with or without local Voice Gateways. Some deployments may also contain a combination of distributed Voice Gateways (possibly for locally dialed calls) and centralized Voice Gateways (possibly for toll-free calls) as well as centralized or distributed treatment and queue points.

Regardless of how many sites are being deployed in this model, there is still only one logical Unified CCE Central Controller. If the Unified CCE Central Controller is deployed with redundancy, Sides A and B can be deployed side-by-side or geographically separated (remote redundancy). For details on remote redundancy, see the Unified ICM/​CCE product documentation.

Unified CCE: Distributed Voice Gateways with Treatment and Queuing Using Unified IP IVR

This deployment model is a good choice if the company has multiple medium to large sites. In this model, Voice Gateways with PSTN trunks terminate into each site. Just as in the centralized call processing model with distributed Voice Gateways, it might be desirable to limit the routing of calls to agents within the site where the call arrived (to reduce WAN bandwidth). An analysis of benefits from customer service levels versus WAN costs is required to determine whether limiting calls within a site is appropriate. The following figure illustrates this model using a traditional Unified CCE deployment with the Unified CCE System PG.

Figure 12. Multisite Deployment with Distributed Call Processing and Distributed Voice Gateways with Unified IP IVR

As with the previous models, many options are possible. The number and type of Unified CCE Servers, Unified CM servers, and Unified IP IVR servers can vary. The LAN and WAN infrastructure, Voice Gateways, PSTN trunks, redundancy, and so forth are also variable within this deployment model. Central processing and gateways may be added for self-service, toll-free calls and support for smaller sites. In addition, the use of a pre-routing PSTN Network Interface Controller (NIC) is also an option.

Advantages

  • Scalability — Each independent site can scale up to the maximum number of supported agents per Unified CM cluster and there is no software limit to the number of sites that can be combined by the Unified CCE Central Controller to produce a single enterprise-wide contact center (provided that the total concurrent agent count is less than the maximum supported agent count in a Unified CCE System). For scalability and sizing information, see the chapter on Sizing Contact Center Resources.
  • All or most VoIP traffic can be contained within the LAN of each site, if desired. The VoIP WAN shown in Figure 18 is required for voice calls to be transferred across sites. Use of a PSTN transfer service (for example, Take Back and Transfer or Transfer Connect) can eliminate that need. If desired, a small portion of calls arriving at a particular site can be queued for agent resources at other sites to improve customer service levels.
  • Unified ICM/CCE pre-routing can be used to load-balance calls (based on agent or Unified IP IVR port availability) to the best site to reduce WAN usage for VoIP traffic.
  • Failure at any one site has no impact on operations at another site.
  • Each site can be sized according to the requirements for that site.
  • The Unified CCE Central Controller provides centralized management for configuration of routing for all calls within the enterprise.
  • The Unified CCE Central Controller provides the capability to create a single enterprise-wide queue.
  • The Unified CCE Central Controller provides consolidated reporting for all sites.

Best Practices

  • The PG, Unified CM cluster, and Unified IP IVR must be co-located at the contact center site.
  • The communication link from the Unified CCE Central Controller to the PG must be sized properly and provisioned for bandwidth and QoS. (For details, see the chapter Bandwidth Provisioning and QoS Considerations.)
  • Gatekeeper-based or RSVP Agent-based call admission control can 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 CCE 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 CCE Central Controller and PG. For example, in the event of a lost Unified CCE Central Controller connection, the Unified CM CTI route points 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 CCE Central Controller. For more information about these options, see the Design Considerations for High Availability chapter.
  • While two inter-cluster call legs for the same call do not cause unnecessary RTP streams, two separate call signaling control paths remain intact between the two clusters (producing logical hair-pinning and reducing the number of inter-cluster trunks by two). Consider the percentage of inter-site transfers when sizing inter-cluster trunks capacities.
  • Latency between Unified CCE Central Controllers and remote PGs cannot exceed 200 ms one way (400 ms round-trip).

Treatment and Queuing

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, perform the queuing 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, queue the call to a Unified IP IVR at Site 2 to avoid generating another inter-cluster call. A second inter-cluster call is made only if an agent at Site 1 was selected for the transfer. The RTP flow at this point is directly from the Voice Gateway at Site 1 to the agent’s phone at Site 1. However, the two Unified CM clusters still logically see two calls in progress between the two clusters.

Transfers

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 inter-cluster 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 out-pulse 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 then routes the call to Site 2, but the call uses two Voice Gateway ports at Site 1 for the remainder of the call.

Unified CCE: Unified CCE System PG

The Unified CCE System PG acts as a single peripheral that joins the Unified CM and Unified IP IVR peripherals of former versions to simplify installation, configuration, and routing. In this model, the PGs at the remote sites can be installed as Unified CCE System PGs to combine the Unified IP IVR and Unified CM peripherals under a single logical PG instance and peripheral.

This model is perhaps more typical of outsourcers that set up a call center specifically for a single client and deploy it as a Unified CCE System PG to allow their client company to connect their Unified CCE Enterprise system to the outsourcer.

Unified CCE: Unified CCE PG

This model, as designed with multiple remote locations, is more suited for the traditional Unified CCE design with multiple distributed peripheral gateways. The system can deployed with the Generic PG or both Unified CM and Unified IP IVR PGs at the sites; however, the new Unified CCE System PG that combines both of these peripherals into a single peripheral for routing and reporting under the traditional model might be easier for new deployments of this solution. Existing customers upgrading to Unified CCE 8.0 may stay on their existing Generic PG or multi-PG model.

Alternative: Parent/Child

The Unified ICM Enterprise (parent) and Unified CCE (child) model is an appropriate alternative deployment to provide local, distributed call processing with a local Unified CM and Unified CCE at each site (child), controlled under a centralized Unified ICM Enterprise parent for enterprise-wide routing, reporting, and call control. This model has the advantage of being more tolerant of WAN outages and each site is completely survivable. The following figure shows this same model deployed using the parent/child model.
Figure 13. Multisite Deployment with Distributed Call Processing and Parent/Child

In this design, there is a parent Unified ICM Enterprise system deployed with Unified CVP and its own Administration & Data server. At each distributed site, there is a complete Unified CCE deployment consisting of Central Controller on one or more servers. There is also a local Administration & Data Server for Unified CCE to perform configuration, scripting, and reporting tasks for that specific site. There is a Unified CCE Gateway PG that connects Unified CCE to the Unified ICM parent and it is part of the Peripheral Gateways deployed on the parent Unified ICM.

An optional deployment for the Unified CCE Gateway PG is to co-locate it with the Unified CCE System PG, while adhering to the following guidelines:
  • If the Unified CCE Gateway PG and Unified CCE System PG Instance Numbers are the same, then the PG number for the Unified CCE Gateway PG and Unified CCE System PG must be different.
  • If the Unified CCE Gateway PG and Unified CCE System PG Instance Numbers are different, then the PG number for the Unified CCE Gateway PG and the Unified CCE System PG may be the same.
  • No additional PGs (such as a VRU PG or MR PG) can be added to this Server.
  • For scalability limits of the co-resident Unified CCE Gateway PG and Unified CCE System PG, see Sizing Unified CCE Components and Servers for additional details.

In this design, the local Unified CCE deployments act as their own local IP ACDs with no visibility to any of the other sites in the system. Users at Site 1 cannot see any of the calls or reports from Site 2 in this model. Only the Unified ICM Enterprise parent system has visibility to all activity at all sites connected to the Unified ICM Enterprise system.

The Unified CVP at the Unified ICM parent site is used to control the calls coming into the distributed sites providing local call queuing and treatment in the VoiceXML Browser in the Voice Gateway. When configuring the Unified ICM parent CVP to use Unified CVP Router Re-query to take control of the call in the event of a failure or answer timeout, the child Unified CCE cannot terminate the ingress call to a child Unified CVP or a child Unified IP IVR. The local Unified IP IVR servers are used only for a local backup if the connection from these Voice Gateways is lost to the parent Unified CVP Call Control server. The local Unified IP IVR also provides local queue treatment for calls that are not answered by the local agents (RONA) rather than sending the call back to the Unified CVP to be re-queued.

The child Unified CCE deployments can also transfer calls across the system between the sites using Unified ICM post-routing by the Unified CCE Gateway PG. The Unified CCE Gateway PG allows the child Unified CCE to ask the Unified ICM to transfer a call to the best agent at another site or to queue it centrally for the next available agent.

Unlike traditional Unified CCE models with distributed Unified CM Peripheral Gateways, the parent/child model provides for complete local redundancy at the contact center site. The local Unified CCE takes over call processing for inbound calls from the Unified CVP gateways and provide local call queuing and treatment in the local Unified IP IVR. This is an excellent design for call center sites that require complete redundancy or 100% up-time and that cannot be down because of a WAN failure.

This design is a good approach for customers who have Unified ICM already installed with their TDM ACD platforms and who want either to add new sites with Unified CCE or to convert an existing site to Unified CCE. It allows the Unified ICM to continue performing enterprise-wide routing and reporting across all of the sites while inserting new Unified CCE technology on a site-by-site basis.


Note


Unified CVP can be at both the parent and child. This is virtually identical to Unified CVP at the parent and IP-IVR at the child from a call flow perspective. One key difference is that information about queued calls at the child Unified CVP are not available at the parent (through the Unified CCE Gateway PG), as is the case if IP-IVR is used. This means that minimum expected delay (MED) over services cannot be used.


Advantages

  • Unified CVP provides a virtual network queue across all the distributed sites controlled by the parent Unified ICM. The parent Unified ICM has visibility into all the distributed sites and sends the call to the next available agent from the virtual queue.
  • Each distributed site can scale up to the maximum number of supported agents on a single Unified CCE deployment. Multiple Unified CCE Central Controllers can be connected to a single Unified CM cluster to scale up to the maximum number of supported agents per cluster. The Unified CCE systems are connected to the parent Unified ICM using the Unified CCE Gateway PG on the parent Unified ICM, which can scale up to the maximum number of supported agents per parent Unified ICM Enterprise system.
  • All or most VoIP traffic can be contained within the LAN of each site, if desired. The QoS WAN shown in Figure 19 is required for voice calls to be transferred across sites. Use of a PSTN transfer service (for example, Take Back and Transfer or Transfer Connect) eliminates that need. If desired, a small portion of calls arriving at a particular site can be queued for agent resources at other sites to improve customer service levels.
  • Unified ICM pre-routing can be used to load-balance calls based on agent or Unified CVP session availability and to route calls to the best site to reduce WAN usage for VoIP traffic.
  • Failure at any one site has no impact on operations at another site.
  • Each site can be sized according to the requirements for that site
  • The parent Unified ICM Central Controller provides centralized management for configuration of routing for all calls within the enterprise.
  • The parent Unified ICM Central Controller provides the capability to create a single enterprise-wide queue.
  • The parent Unified ICM Central Controller provides consolidated reporting for all sites.

Disadvantages

Server count — The number of servers that are required to manage the parent/child model is usually higher due to the increased number of software components (additional Unified CCE Gateway PGs required if co-locating with Unified CCE System PG is not an option, additional Central Controller for each child, and so forth).

Best Practices

  • Co-locate the Unified CCE Gateway PG, Unified Communications Manager cluster, Unified IP IVR, and Unified CCE (if possible) at the contact center site.
  • The communication link from the parent Unified ICM Central Controller to the Unified CCE Gateway PG must be sized properly and provisioned for bandwidth and QoS. (For details, see the Bandwidth Provisioning and QoS Considerations chapter.)
  • Gatekeeper-based or RSVP agent-based call admission control is 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 Unified CCE Gateway PG and the parent Unified ICM Central Controller is lost, then all contact center routing for calls at that site is put under control of the local Unified CCE. Unified CVP-controlled ingress Voice Gateways have survivability TCL scripts to redirect inbound calls to local Unified Communications Manager CTI route points and the local Unified IP IVR are used to handle local queuing and treatment during the WAN outage. This is a major feature of the parent/child model to provide complete local survivability for the call center. For more information, see the Design Considerations for High Availability chapter.
  • While two inter-cluster call legs for the same call do not cause unnecessary RTP streams, two separate call signaling control paths remain intact between the two clusters (producing logical hair-pinning and reducing the number of inter-cluster trunks by two). Consider the percentage of inter-site transfers when sizing inter-cluster trunks capacities.
  • Latency between parent Unified ICM Central Controllers and remote Unified CCE Gateway PGs must not exceed 200 ms one way (400 ms round-trip).

IVR: Distributed Voice Gateways with Treatment and Queuing Using Unified CVP

This deployment model is the same as the previous model, except that Unified CVP is used instead of Unified IP IVR for call treatment and queuing. In this model, Voice Gateways with PSTN trunks terminate into each site. Just as in the centralized call processing model with distributed Voice Gateways, it might be desirable to limit the routing of calls to agents within the site where the call arrived (to reduce WAN bandwidth). Call treatment and queuing can also be achieved at the site where the call arrived, further reducing the WAN bandwidth needs. Figure 20 illustrates this model using a traditional Unified CCE deployment.

Figure 14. Multisite Deployment with Distributed Call Processing and Distributed Voice Gateways with Unified CVP

As with the previous models, many options are possible. The number and type of Unified CCE Servers, Unified CM servers, and Unified CVP servers can vary. The LAN and WAN infrastructure, Voice Gateways, PSTN trunks, redundancy, and so forth are also variable within this deployment model. Central processing and gateways may be added for self-service, toll-free calls and support for smaller sites. The use of a pre-routing PSTN Network Interface Controller (NIC) is also an option.

Advantages

  • Unified CVP Servers can be located either centrally or remotely. Call treatment and queuing is still distributed, executing on the local gateway, regardless of Unified CVP server location. Unified CVP is shown centrally located in Figure 20.
  • For information about the number of agents supported per PG and across the entire system, see Sizing Unified CCE Components and Servers.
  • All or most VoIP traffic can be contained within the LAN of each site, if desired. The QoS WAN are required for voice calls to be transferred across sites. Use of a PSTN transfer service (for example, Takeback N Transfer) eliminates that need. If desired, a small portion of calls arriving at a particular site can be queued for agent resources at other sites to improve customer service levels.
  • Unified CCE pre-routing can be used to load-balance calls and route them to the best site to reduce WAN usage for VoIP traffic.
  • Failure at any one site has no impact on operations at another site.
  • Each site can be sized according to the requirements for that site.
  • The Unified CCE Central Controller provides centralized management for configuration of routing for all calls within the enterprise.
  • The Unified CCE Central Controller provides the capability to create a single enterprise-wide queue.
  • The Unified CCE Central Controller provides consolidated reporting for all sites.

Best Practices

  • The Unified CM PG and Unified CM cluster must be co-located. The Unified CVP PG and Unified CVP servers must be co-located.
  • The communication link from the Unified CCE Central Controller to PG must be properly sized and provisioned for bandwidth and QoS. A tool is accessible to Cisco partners and Cisco employees for computing the bandwidth needed between the Unified CCE Central Controller and the VRU Peripheral Gateway. This tool is called the VRU Peripheral Gateway to Unified CCE Central Controller Bandwidth Calculator, and it is also available through the Steps to Success Portal.
  • If the communication link between the PG and the Unified CCE Central Controller is lost, then all contact center routing for calls at that site is 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 CCE Central Controller and PG.
  • Latency between Unified CCE Central Controllers and remote PGs must not exceed 200 ms one way (400 ms round-trip)

IVR: Treatment and Queuing

Unified CVP queues and treats calls on the remote gateways eliminating the need to terminate the voice bearer traffic at the central site. Unified CVP servers may be located at the central site or distributed to remote sites. WAN bandwidth must still be provisioned for transfers and conferences that involve agents at other locations.

Unlike Unified IP IVR, with Unified CVP the call legs are torn down and reconnected, avoiding signaling hairpins. With Unified IP IVR, two separate call signaling control paths remains intact between the two clusters (producing logical hair-pinning and reducing the number of inter-cluster trunks by two).

Transfers

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 inter-cluster 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 out-pulse 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 then routes the call to Site 2, but the call uses two Voice Gateway ports at Site 1 for the remainder of the call.

Unified CCE: Unified CCE System PG

The Unified CCE System PG is not a good fit for this model because it does not support Unified CVP for queuing and the IVR PIMs on the Unified CCE System PG are unused.

Unified CCE: Unified CCE PG

The Unified CCE PG is the required PG for this deployment model.

Unified CCE: Distributed Unified CCE Option with Distributed Call Processing Model

Figure 21 illustrates this deployment model.

Figure 15. Distributed Unified CCE Option Shown with Unified IP IVR

Advantages

The primary advantage of the distributed Unified CCE option is the redundancy gained from splitting the Unified CCE Central Controller between two redundant sites.

Best Practices

  • Unified CCE Central Controllers (Routers and Loggers) require a separate network path or link to carry the private communications between the two redundant sites. In a non-distributed Unified CCE model, the private traffic usually traverses an Ethernet crossover cable or LAN connected directly between the Side A and Side B Unified CCE Central Controller components. In the distributed Unified CCE model, the private communications between the A and B Unified CCE components travel across a dedicated link with at least as much bandwidth as a T1 line.
  • Latency across the private separate link must not exceed 100 ms one way (200 ms round-trip), but 50 ms (100 ms round-trip) is preferred.
  • Latency between Unified CCE Central Controllers and remote PGs must not exceed 200 ms one way (400 ms round-trip).
  • The private link cannot traverse the same path as public traffic. The private link must have path diversity and must reside on a link that is completely path-independent from Unified CCE public traffic. This link is used as part of the system fault tolerant design. For more information, see Design Considerations for High Availability.
  • The redundant centralized model is explored in the next section on IPT Clustering over the WAN.

IPT: Clustering Over the WAN

As part of the centralization of call processing, many customers prefer to combine the redundancy of the distributed Unified CM call processing model with the simplicity of having a single Unified CM cluster for a single dial plan and voice system to administer. This combination of models provides for a single Unified CM cluster with its subscriber servers split across data center locations to provide a single cluster with multiple distributed call processing servers for a highly available and redundant design (known as clustering over the WAN).

Unified CM clustering over the WAN may also be used with Unified CCE for contact centers to allow full agent redundancy in the case of a data center (central site) outage. Implementation of clustering over the WAN for Unified CCE does have several strict requirements that differ from other models. Bandwidth between central sites for Unified CCE public and private traffic, Unified CM intra-cluster communication signaling (ICCS), and all other voice-related media and signaling must be properly provisioned with QoS enabled. The WAN between central sites must be highly available (HA) with redundant links and redundant routers.

Advantages

  • No single point of failure, including loss of an entire central site.
  • Cisco Unified Mobile Agents (Remote Agent) require no reconfiguration to remain fully operational in case of site or link outage. When outages occur, agents and agent devices dynamically switch to the redundant site.
  • Central administration for both Unified CCE and Unified CM.
  • Reduction of servers for distributed deployment.

Best Practices

  • Deploy a minimum of three WAN links for systems that employ the clustered over the WAN model. Deploy at least two links for the high-availability network that carries the Unified CCE public traffic (see Figure 22). Use a separate WAN link for the Unified CCE private traffic (see Figure 23). If QoS and bandwidth are configured correctly (see the guidelines in the Bandwidth Provisioning and QoS Considerations chapter), these WAN links can be converged with other corporate traffic as long as the private and public traffic are not carried over the same link. Carry the Unified CM ICCS traffic over the high-availability network used by the Unified CCE public communications.
    Figure 16. High-Availability WAN Network for the Unified CCE Public Traffic

    Figure 17. Separate WAN Link for Unified CCE Private Traffic

  • It is possible to deploy Unified CCE clustering over the WAN with two links, if the following rules are applied:
    • During normal operations, the Unified CCE public and private traffic must be carried over separate links; they must not be carried over the same link.
    • Carry the Unified CM traffic over the Unified CCE public link in normal conditions (see Figure 24).
    • Two routers are required on each side of the WAN for redundancy; connect these to different WAN links.
    Figure 18. Network Architecture Under Normal Operations

  • In case of network failure, configure the WAN link that carries the Unified CCE public traffic to fail-over to the other link that carries the Unified CCE private traffic (see Figure 25). Temporarily allow the Unified CM ICCS traffic to fail-over to the private link. This prevents situations where a CTI Manager that connects to the active Agent PG loses its WAN connection to the Unified CM node to which the agent phones are registered. Restore the link as fast as possible so that the public and private Unified CCE traffic are carried over separate links. If the redundant link that carries the Unified CCE private traffic also fails, Unified CCE instability and data loss can occur, including the corruption of one Logger database. Manual intervention may be required. This is why it is very important to actively monitor any network failure at all times.

    The links must also be sized correctly to accommodate this fail-over situation where the private link carries the entire WAN traffic, including the public and ICCS traffic. QoS and bandwidth must be configured according to the guidelines in the Bandwidth Provisioning and QoS Considerations chapter.

    Figure 19. Network Architecture After Failure of the Unified CCE Public Network

  • It is also possible to allow the private link to fail-over to the public link. However, if the total fail-over latency takes more than 500 ms (five times the TCP keep alive interval of 100 ms), the Unified CCE system considers the private link to be down. If the public link is also down, Unified CCE instability and data loss can occur, including the corruption of one Logger database. Manual intervention is required. The total fail-over latency typically includes the round-trip transmission latency, the routing protocol convergence delay, the HSRP convergence delay (if applicable), queuing and packetization delays, and any other delay that is applicable. If the total fail-over latency is higher than 500 ms, or if you suspect possible recurrent network flapping, deploy three WAN links and keeping the private traffic separate from the public traffic at all times. Also, the links must be sized correctly to accommodate this fail-over situation where the public link carries the entire WAN traffic, including the private and ICCS traffic. Restore the link as fast as possible so that the public and private Unified CCE traffic are carried over separate links.
  • If QoS and bandwidth are configured correctly (see the guidelines in the Bandwidth Provisioning and QoS Considerations chapter for more details), these WAN links can be converged with other corporate traffic.
  • With a SONET fiber ring, which is highly resilient and has built-in redundancy, the public and private traffic can be carried over the same SONET ring under normal operations or following a network fail-over. A separate link for the private traffic is not required in this case. Also, two routers are required on each side of the WAN for redundancy. Under normal operations, use one router for the Unified CCE public traffic and use the other router for the Unified CCE private traffic. The other rules described in this section also apply.
    Figure 20. Network Architecture Based on a SONET Ring Under Normal Operations

  • The high-availability (HA) WAN between the central sites must be fully redundant with no single point of failure. (For information regarding site-to-site redundancy options, see the WAN infrastructure and QoS design guides.) In case of partial failure of the high-availability WAN, the redundant link must be capable of handling the full central-site load with all QoS parameters. For more information, see Bandwidth Requirements for Unified CCE Clustering Over the WAN.
  • A high-availability (HA) WAN using point-to-point technology is best implemented across two separate carriers, but this is not necessary when using a ring technology.
  • Latency requirements across the high-availability (HA) WAN must meet the current Cisco Unified Communications requirements for clustering over the WAN. Currently, a maximum latency of 40 ms one way (80 ms round-trip) is allowed with Unified CM Release 6.1 or later releases. With prior versions of Unified CM, the maximum latency is 20 ms one way. For full specifications, see the Unified Communications System Solution Reference Network Design (SRND) Guide at http:/​/​www.cisco.com/​en/​US/​docs/​voice_ip_comm/​uc_system/​design/​guides/​UCgoList.html.
  • Unified CCE latency requirements can be met by conforming to Cisco Unified Communications requirements. However, the bandwidth requirements for Unified CM intra-cluster communications differ between Unified CCE and Cisco Unified Communications. For more information, see the section on Bandwidth Requirements for Unified CCE Clustering Over the WAN.
  • Bandwidth requirements across the high-availability (HA) WAN include bandwidth and QoS provisioning for:
    • Unified CM intra-cluster communication signaling (ICCS)
    • Communications between Unified CCE Central Controllers
    • Communications between Unified CCE Central Controller and PG
    • Communications between CTI Object Server (CTI OS) and CTI Server, if using CTI OS
    • See Bandwidth Requirements for Unified CCE Clustering Over the WAN.
  • Deploy separate dedicated links for Unified CCE private communications between Unified CCE Central Controllers Side A and Side B and between PGs Side A and Side B to ensure path diversity. Path diversity is required due to the architecture of Unified CCE. Without path diversity, the possibility of a dual (public communication and private communication) failure exists. If a dual failure occurs even for a moment, Unified CCE instability and data loss can occur, including the corruption of one Logger database. The separate links for Unified CCE private communications can be converged with other corporate traffic if QoS and bandwidth are configured correctly, but they cannot be converged with the Unified CCE public traffic.
  • The separate private links may be either two links (one for Central Controller private traffic and one for Unified CM PG private traffic) or one converged link containing both Central Controller and PG private traffic. See Site-to-Site Unified CCE Private Communications Options for more information.
  • Separate paths must exist from agent sites to each central site. Both paths must be capable of handling the full load of signaling, media and other traffic if one path fails. These paths may reside on the same physical link from the agent site with a WAN technology such as Frame Relay using multiple permanent virtual circuits (PVCs).
  • The minimum cluster size using Unified IP IVR as the treatment and queuing platform is five nodes (publisher plus four subscribers). This minimum is required to allow Unified IP IVR at each site to have redundant connections locally to the cluster without traversing the WAN. JTAPI connectivity between Unified CM and Unified IP IVR is not supported across the WAN in this model. Local gateways also need local redundant connections to Unified CM.
  • The minimum cluster size using Unified CVP as the treatment and queuing platform is three nodes (publisher plus two subscribers). However, a deployment with five nodes is preferable, especially if there are phones (either contact center or non-contact center) local to the central sites, central gateways, or central media resources, which requires local fail-over capabilities.
  • In a deployment with clustering over the WAN, the VRU PG can connect to a local IP IVR or Unified CVP or to a redundant IP IVR or Unified CVP across the WAN. For information about bandwidth requirements, see the Bandwidth Provisioning and QoS Considerations chapter.
  • In a deployment with clustering over the WAN, the Unified CM PG must be on the same LAN segment with the CTI Manager to which it is connected.
  • Cisco Finesse supports a WAN connection between the Cisco Finesse Server and the CTI Server. CTI reconnect time is determined by a number of factors, including the number of configured agents, the number of concurrent agents who are taking calls, the number of skill groups, and bandwidth. To determine the bandwidth requirements for your specific deployment, review the Finesse Bandwidth Calculator.

    Any of the network connections shown as arrows in the following diagram can be over a LAN or WAN.

    Figure 21. Cisco Finesse Network Connections

Centralized Voice Gateways with Centralized Call Treatment and Queuing Using Unified IP IVR

In the following model, the Voice Gateways are located in the central sites. Unified IP IVR is centrally located and used for treatment and queuing on each side. Figure 27 illustrates this model.

Figure 22. Centralized Voice Gateways with Centralized Call Treatment and Queuing Using Unified IP IVR

Advantages

  • Component location and administration are centralized.
  • Calls are treated and queued locally, eliminating the need for queuing across a WAN connection.

Best Practices

  • WAN connections to agent sites must be provisioned with bandwidth for voice as well as control and CTI. See Bandwidth Requirements for Unified CCE Clustering Over the WAN for more information.
  • A local Voice Gateway might be needed at remote sites for local out-calling and 911. For more information, see the Cisco Unified Communications Solution Reference Network Design (SRND) Guide at http:/​/​www.cisco.com/​en/​US/​products/​sw/​voicesw/​ps556/​tsd_​products_​support_​series_​home.html.
  • Central site outages include loss of half of the ingress gateways, assuming a balanced deployment. Gateways and IVRs must be scaled to handle the full load in both sites if one site fails.
  • Carrier call routing must be able to route calls to the alternate site in the case of a site or gateway loss. Pre-routing may be used to balance the load, but it cannot prevent calls from being routed to a failed central site. Pre-routing is best deployed only as a last resort.

Clustering Over the WAN with Unified CCE System PG

Clustering over the WAN with Unified CCE System PG is now supported. However, due to the fact that a single Unified CCE System peripheral is controlling all of the Unified IP IVRs and the Unified Communications Manager, the load-balancing of calls between Unified IP IVRs does not take into account which site the call came into; it simply distributes the calls to whichever Unified IP IVR is least loaded. This means that calls coming into Site A might be treated by a Unified IP IVR in Site B. Additionally, both the A-Side And B-side Unified CCE System PG know about all of the Unified IP IVRs. PIM activation logic determines if the A-side or the B-side PIM connects to each of the Unified IP IVRs. This means that the PG at site A might connect to the Unified IP IVR at site B and traffic might not be sent optimally over the WAN. In this model, make sure the WAN is sized for proper operation given this fact. To avoid this bandwidth overhead, you can consider Clustering over the WAN deployments with Unified CVP in place of Unified IP-IVR.

Centralized Voice Gateways with Centralized Call Treatment and Queuing Using Unified CVP

In this model, the Voice Gateways are VoiceXML gateways located in the central sites. Unified CVP is centrally located and used for treatment and queuing.

Figure 23. Centralized Voice Gateways with Centralized Call Treatment and Queuing Using Unified CVP

Advantages

  • Component location and administration are centralized.
  • Calls are treated and queued locally, eliminating the need for queuing across a WAN connection.
  • There is less load on Unified CM because Unified CVP is the primary routing point. This allows higher scalability per cluster compared to Unified IP IVR implementations. See the Sizing Unified CCE Components and Servers chapter for more information.

Best Practices

Considerations for Clustering Over the WAN

The following figure illustrates a deployment with clustering over the WAN.

Figure 24. Clustering Over the WAN

The following guidelines and considerations apply to deployments with clustering over the WAN:
  • The network deployment supports high availability, converged Visible, and Private Networks. The Unified ICM and Unified CCE Central Controller's Private traffic and Visible (Public) traffic are isolated and converge on different edge devices.
  • WAN considerations for communications between the two Data Centers may include a Multiprotocol Label Switching (MPLS) backbone with VPN routing and forwarding table VRFs.
  • Design the network to prevent any single points of failure. The visible network and private networks need to converge on separate switches and routers before connecting to the WAN.
  • Isolation of the private network is not required. Central Controllers and Unified CCE System PGs can share a common private network path.
  • Multiple private network paths can be provisioned. (Central Controllers and Unified CCE System PGs have separate private networks.)
  • Bandwidth must be guaranteed across the WAN for the private network path traffic and visible (public) network traffic with appropriate traffic prioritization. For more information, see the Bandwidth Provisioning and QoS Considerations chapter.
  • Currently there is no bandwidth calculator for the private network bandwidth between the gateway and system PG pairs because this has not been certified. For guidance, see the section on Bandwidth Provisioning.
  • A side-to-side private network of a duplexed Central Controller and PGs has a maximum one-way latency of 100 ms, but 50 ms is preferred.

The underlying network infrastructure for LAN and WAN provisioning must meet all the above requirements. Key factors are isolation of visible and private paths as well as critical low-latency and bandwidth, especially on the private path. The isolated private networks for PGs and Central Controllers provide some degree of independence from each other's private link failures. The more path and route diversity provisioned, the greater the fault tolerance. For example, if the private network between the parent central controllers goes down, the child central controllers can still continue to function in duplex mode.

MPLS Considerations

If an MPLS network can guarantee the route diversity, latency, and bandwidth; and if it is configured to support label switch paths that route through independent topologies and hardware elements to meet the above requirements, then the application works as designed. It is important to ensure that the route autonomy is not compromised over time through adaptive change.

For additional information regarding best practices and high-availability deployments, see the section on IPT: Clustering over the WAN.

Distributed Voice Gateways with Distributed Call Treatment and Queuing Using Unified CVP

In this model, the Voice Gateways are VoiceXML gateways distributed to agent locations. Unified CVP is centrally located and used for treatment and queuing on the remote gateways.

Figure 25. Distributed Voice Gateways with Distributed Call Treatment and Queuing Using Unified CVP

Advantages

  • No or minimal voice RTP traffic across WAN links if ingress calls and gateways are provisioned to support primarily their local agents. Transfers and conferences to other sites traverse the WAN.
  • Calls are treated and queued at the agent site, eliminating the need for queuing across a WAN connection.
  • Local calls incoming and outgoing, including 911, can share the local VoiceXML gateway.
  • There is less load on Unified CM because Unified CVP is the primary routing point. This allows higher scalability per cluster compared to Unified IP IVR implementations. See Sizing Unified CCE Components and Servers for more information.

Best Practices

  • Distributed gateways require minimal additional remote maintenance and administration over centralized gateways.
  • The media server for Unified CVP may be centrally located or located at the agent site. Media may also be run from gateway flash. Locating the media Server At the agent site reduces bandwidth requirements but adds to the server count and maintenance costs.

Site-to-Site Unified CCE Private Communications Options

Unified CCE private communications must travel on a separate path from the public communications between Unified CCE components. There are two options for achieving this path separation: dual and single links.

Unified CCE Central Controller Private and Unified CM PG Private Across Dual Links

Dual links separate Unified CCE Central Controller Private traffic from VRU/CM PG Private traffic.

Figure 26. Unified CCE Central Controller Private and Unified CM PG Private Across Dual Links

Advantages

  • Failure of one link does not cause both the Unified CCE Central Controller and PG to enter simplex mode, thus reducing the possibility of an outage due to a double failure.
  • The QoS configuration is limited to two classifications across each link, therefore links are simpler to configure and maintain.
  • Resizing or alterations of the deployment model and call flow might affect only one link, thus reducing the QoS and sizing changes needed to ensure proper functionality.
  • Unanticipated changes to the call flow or configuration (including misconfiguration) are less likely to cause issues across separate private links.

Best Practices

  • Deploy separate links across separate dedicated circuits. The links, however, do not have to be redundant and must not fail-over to each other.
  • Link sizing and configuration must be examined before any major change to call load, call flow, or deployment configuration.
  • Make the link a dedicated circuit; not tunneled across the high-availability (HA) WAN. See Best Practices, at the beginning of the section about IPT: Clustering Over the WAN for more information about path diversity.

Unified CCE Central Controller Private and Unified CM PG Private Across Single Link

A single link, as shown in Figure 32, carries both Unified CCE Central Controller Private traffic and VRU/CM PG Private traffic. Single-link implementations are more common and less costly than dual-link implementations.

Figure 27. Unified CCE Central Controller Private and Unified CM PG Private Across a Single Link

Advantages

  • Less costly than separate-link model.
  • Fewer links to maintain, but is more complex.

Best Practices

  • The link does not have to be redundant. If a redundant link is used, however, latency on fail-over must not exceed 500 ms.
  • Separate QoS classifications and reserved bandwidth are required for Central Controller high-priority and PG high-priority communications. For details, see the Bandwidth Provisioning and QoS Considerations chapter.
  • Link sizing and configuration must be examined before any major change to call load, call flow, or deployment configuration. This is especially important in the single-link model.
  • Make the link a dedicated circuit fully isolated from, and not tunneled across, the high-availability (HA) WAN. See Best Practices, at the beginning of the section on IPT: Clustering Over the WAN for more information about path diversity.

Failure Analysis of Unified CCE Clustering Over the WAN

This section describes the behavior of clustering over the WAN for Unified CCE during certain failure situations. The stability of the high-availability (HA) WAN is extremely critical in this deployment model. Failure of the high-availability WAN is considered outside the bounds of what normally happens.

For illustrations of the deployment models described in this section, see the figures shown in IPT: Clustering Over the WAN.

Entire Central Site Loss

Loss of the entire central site is defined as the loss of all communications with a central site, as if the site were switched off. This can result from natural disasters, power issues, major connectivity issues, and human error. If a central site retains some but not all connectivity, it is not considered a site loss but rather a partial connectivity loss. This scenario is covered in subsequent sections.

When an entire central site has completely lost Unified CCE clustering over the WAN, mobile agents failover properly to the redundant site. Fail-over times can range from 1 to 60 seconds for agents. Variations are due to agent count, phone registration location, and agent desktop server used.

When using distributed VoiceXML gateways and Unified CVP, the gateways must fail-over from one site to another if their primary site is lost. This fail-over takes approximately 30 seconds, and calls coming into the remote gateways during those 30 seconds are lost.

Private Connection Between Site 1 and Site 2

If the private connection between Unified CCE Central Controller Sides A and B fails, one Unified CCE Call Router goes out-of-service and the other Unified CCE Call Router then runs in simplex mode until the link is reinstated. PGs with active connections to the Unified CCE Router that goes out-of-service realign message streams to the remaining active Unified CCE Router side. This situation does not cause any call loss or failure.

If the private connection between PG Side A and PG Side B fails, the disabled synchronizer (Side B) initiates a test of its peer synchronizer by using the TOS procedure on the Public or Visible Network connection. If PG Side B receives a TOS response stating that the A side synchronizer is enabled or active, then the B side immediately goes out of service, leaving the A side to run in simplex mode until the Private Network connection is restored. The PIM, OPC, and CTI SVR processes become active on PG Side A, if not already in that state, and the CTI OS Server process still remains active on both sides as long as the PG Side B server is healthy.

If the B side does not receive a message stating that the A side is enabled, then Side B continues to run in simplex mode and the PIM, OPC, and CTI SVR processes become active on PG Side B, if not already in that state. This condition (PG Side B going Active) occurs only if the PG Side A server is truly down or unreachable due to a double failure of visible and private network paths. The Side A or Side B PG runs in simplex mode until the private link is reinstated.

There is no impact to the agents, calls in progress, or calls in queue because the agents stay connected to their already established CTI OS Server process connection. The system can continue to function normally; however, the PGs are in simplex mode until the private network link is restored.

When using a combined private link, Unified CCE Central Controller and PG private connections are lost if the link is lost. This causes both components to switch to simplex mode as described above. This situation does not cause any call loss or failure.

Connectivity to Central Site from Mobile Agent Site

If connectivity to one of the central sites is lost from a Mobile agent site, all phones and agent desktops immediately switch to the second central site and begin processing calls. Failover typically takes between 1 and 60 seconds.

High-Availability WAN Failure

By definition, a high-availability (HA) WAN does not fail under normal circumstances. If the HA WAN is dual-path and fully redundant, a failure of this type is highly unusual. This section discusses what happens in this unlikely scenario.

If the HA WAN is lost for any reason, the Unified CM cluster becomes split. The primary result from this occurrence is that Unified CCE loses contact with half of the agent phones. Unified CCE is in communication with only half of the cluster and cannot communicate with or see any phones registered on the other half. This causes Unified CCE to immediately log out all agents with phones that are no longer visible. These agents cannot log back in until the HA WAN is restored or their phones are forced to switch cluster sides.

Mobile Agent Over Broadband

An organization might want to deploy Unified CCE to support mobile agents (for example, at-home agents) using a Cisco Unified IP Phone over a broadband internet connection. This section outlines the mobile agent solution that can be deployed using a desktop broadband asymmetric digital subscriber line (ADSL) or Cable connection as the remote network. Another option is to use the Cisco Unified Mobile Agent solution (for details, see the Cisco Unified Mobile Agent chapter). Both Cisco Unified Mobile Agent and Mobile agent over Broadband can be supported concurrently using the same back-end infrastructure with the Cisco Virtual Office solution, which is an underlying end-to-end secure infrastructure for remote tele-workers utilizing a converged VPN architecture.

The Cisco Voice and Video Enabled IPSec VPN (V3PN), ADSL, or Cable connection can use a Cisco 800 Series router as an edge router to the broadband network. The Cisco 800 Series router can provide the mobile agent with V3PN, Encryption, Network Address Translation (NAT), Firewall, Cisco IOS Intrusion Detection System (IDS), and QoS on the broadband network link to the Unified CCE campus. Mobile agent V3PN aggregation on the campus is provided through LAN to LAN VPN routers.

Use the Cisco 800 Series router with the following features for mobile agent over broadband:
  • Quality of Service (QoS) with Low-Latency Queuing (LLQ) and Class-Based Weighted Fair Queuing (CBWFQ) support
  • Managed Switch
  • Power over Ethernet (optional)

The Cisco 830, 870, and 880 Series routers are examples of acceptable routers. Avoid using the Cisco 850 and 860 Series routers for this application because they have limited QoS feature support.

Advantages

  • A mobile agent deployment results in money saved for a contact center enterprise, thereby increasing return on investment (ROI).
  • Mobile agents can be deployed with standard Unified CCE agent desktop applications such as Cisco CTI OS, Cisco Agent Desktop, or customer relationship management (CRM) desktops.
  • The Broadband Agent Desktop Always-on connection is a secure extension of the corporate LAN in the home office.
  • Mobile agents have access to the same Unified CCE applications and most Unified CCE features in their home office as when they are working at the Unified CCE contact center and they can access those features in exactly the same way.
  • The mobile-agent router provides high-quality voice using IP phones with simultaneous data to the agent desktop over existing broadband service.
  • Unified CCE home agents and home family users can securely share broadband Cable and DSL connections with authentication of Unified CCE corporate users providing access to the VPN tunnel.
  • The mobile-agent routers can be managed centrally by the enterprise using a highly scalable and flexible management product such as Cisco Unified Operations Manager.
  • The mobile-agent-over-broadband solution is based on Cisco IOS VPN Routers for resiliency, high availability, and a building-block approach to high scalability that can support thousands of home agents.
  • All traffic, including data and voice, is encrypted with the Triple Data Encryption Standard (3DES).
  • The remote-agent router can be deployed as part of an existing Unified CM installation.
  • Mobile agents can have the same extension type as campus agents.

Best Practices

  • Follow all applicable V3PN and Cisco Virtual Office design guidelines outlined in the documentation available at http:/​/​cisco.com/​go/​cvo.
  • Configure mobile agent IP phones to use g.729 with minimum bandwidth limits. Higher-quality voice can be achieved with the g.711 codec. The minimum bandwidth to support g.711 is 512 kbps upload speed.
  • Implement fault and performance management tools such as NetFlow, Service Assurance Agent (SAA), and Internetwork Performance Monitor (IPM).
  • Wireless access points are supported; however, their use is determined by the enterprise security polices for Mobile agents.
  • Only one mobile agent per household is supported.
  • Configure the conference bridge on a DSP hardware device. There is no loss of conference voice quality using a DSP conference bridge. This is the preferred solution even for pure Cisco Unified Communications deployments.
  • The remote-agent-over-broadband solution is supported only with centralized Unified CCE and Unified CM clusters.
  • There might be times when the ADSL or Cable link goes down. When the link is back up, you might have to reset your ADSL or Cable modem mobile agent router and IP phone. This task requires Mobile agent training.
  • Only unicast Music on Hold (MoH) streams are supported.
  • There must be a Domain Name System (DNS) entry for the mobile agent desktop, otherwise the agent cannot connect to a CTI server. DNS entries can be updated dynamically or entered as static updates.
  • The mobile agent workstation and IP phone must be set up to use Dynamic Host Configuration Protocol (DHCP).
  • The mobile agent workstation requires Windows 7 or Windows XP Pro for the operating system. For Windows XP, XP Remote Desktop Control must also be installed.
  • The Cisco Unified IP Phone requires a power supply if the remote-agent router does not have the Power over Ethernet option.
  • Mobile agent broadband bandwidth requires a minimum of 256 kbps upload speed and 1.4 Mbps download speed for ADSL and 1 Mbps download for Cable. Before actual deployment, make sure that the bandwidth is correct. If you are deploying Cable, take into account peak usage times. If link speeds fall below the specified bandwidth, the home agent can encounter voice quality problems such as clipping.
  • Mobile agent round-trip delay to the Unified CCE campus is not to exceed 180 ms for ADSL or 60 ms for Cable. Longer delay times can result in voice jitter, conference bridge problems, and delayed agent desktop screen pops.
  • If the Music on Hold (MoH) server is not set up to stream using a g.729 codec, then a transcoder must be set up to enable outside callers to receive MoH.
  • CTI OS Supervisor home and campus supervisors can silently monitor, barge in, and intercept (but not record) home agents. CTI OS home and campus supervisors can send and receive text messages, make an agent ready, and log out home agents.
  • Connect the agent desktop to the RJ45 port on the back of the IP phone. Otherwise, CTI OS Supervisor cannot voice-monitor the agent phone.
  • Only IP phones that are compatible with Cisco Unified CCE are supported. For compatibility information, see the following documentation:
  • You can find a test for the broadband line speed at http:/​/​broadbandreports.comBroadbandreports.com. From this website, you can execute a test that benchmarks the home agent’s line speed (both upload and download) from a test server.

Mobile Agent with Unified IP Phones Deployed by using the Cisco Virtual Office Solution

In this model, the mobile agent IP phone and workstation are connected by using the VPN tunnel to the main Unified CCE campus. Customer calls routed to the mobile agent are handled in the same manner as campus agents, as shown in Figure 33.

Figure 28. Mobile Agent with IP Phones Deployed via the Cisco Virtual Office Solution

Advantages

  • High-speed broadband enables cost-effective office applications.
  • Site-to-site always-on VPN connection.
  • Advanced security functions allow extension of the corporate LAN to the home office.
  • Supports full range of converged desktop applications including CTI data and high-quality voice.

Best Practices

  • Minimum broadband speed supported is 256 kbps upload and 1.0 Mbps download for cable.
  • Minimum broadband speed supported is 256 kbps upload and 1.4 Mbps download for ADSL.
  • Agent workstation must have 500 MHz and 512 MB RAM or greater.
  • IP phone must be configured to use g.729 on minimum broadband speeds.
  • QoS is enabled only at the remote-agent router edge. Currently, service providers are not providing QoS.
  • Enable security features on the remote-agent router.
  • The Cisco 7200 VXR and Catalyst 6500 IPSec VPN Services Module (VPNSM) offer the best LAN-to-LAN performance for agents.
  • The mobile agent home phone must be used for 911 calls.
  • Use Redirect-on-no-answer (RONA) when a mobile agent is logged in and ready but is unavailable to pick up a call.

Traditional ACD Integration

Enterprises that want to integrate traditional ACDs with their Unified CCE use a parent/child deployment where the Unified ICM and Unified CCE each have a Central Controller or a hybrid deployment where Unified ICM and Unified CCE use a shared Central Controller. Several options exist within those categories depending on how the calls are routed within the deployment.

Hybrid Deployment with PSTN Pre-routing

Enterprises that want to load-balance calls between a traditional ACD site and a Unified CCE site add a pre-routing Network Interface Controller (NIC) as shown in Figure 34. This configuration requires that the Unified ICM have a NIC that supports the PSTN service provider. In this scenario, the PSTN queries the Unified ICM/CCE Central Controller (through the NIC) to determine which site is best and instructs the PSTN on where (which site) to deliver the call. Any call data provided by the PSTN to the Unified ICM/CCE is passed to the agent desktop (traditional ACD or Unified CCE).

To transfer calls between two sites (ACD site and Unified CCE site), use a PSTN transfer service. Use of a PSTN transfer service avoids any double trunking of calls at either site. An alternative to using a PSTN transfer service is to deploy TDM voice circuits between the traditional ACD and Unified CCE Voice Gateways. In that environment, any transfer of a call back to the original site results in double trunking between the two sites. Each additional transfer between sites results in an additional TDM voice circuit being used.

Figure 29. Integrating a Traditional ACD with a Unified CCE Site Using a Hybrid Deployment and Pre-routing via PSTN

Hybrid Deployment with Fixed PSTN Delivery

An alternative to pre-routing calls from the PSTN is to have the PSTN deliver calls to just one site or to split the calls across the two sites according to some set of static rules provisioned in the PSTN. When the call arrives at either site, either the traditional ACD or the Unified CM generates a route request to the hybrid Unified ICM/CCE to determine which site is best for this call. If the call needs to be delivered to an agent at the opposite site from where the call was originally routed, then TDM circuits between sites is required. The determination of where calls are routed (and if and when they are transferred between sites) depends on the enterprise business environment, objectives, and cost components.

Hybrid Deployment with Unified CVP

Alternatively, customers may choose to front-end all calls with Unified CVP to provide initial call treatment and queuing across both the TDM ACD and Unified CCE agents, as shown in Figure 35.

Figure 30. Integrating Unified CVP with a Traditional ACD and a Unified CCE Site Using a Hybrid Deployment and Unified CVP

In this design, all calls first come to the Voice Gateway controlled by Unified CVP and they are then directed by the Unified ICM/CCE Call Router. Unified ICM/CCE uses the PG connections to the TDM ACD and Unified CCE PG to monitor for available agents. Calls are queued in Unified CVP until an agent becomes available in either environment. When a call needs to be transferred to the TDM ACD, it hairpins in the Voice Gateway (meaning that it comes into the gateway on a T1 interface from the PSTN carrier network and goes out on a second physical T1 interface) to appear as a trunk on the TDM ACD. Most TDM ACDs are unable to accept inbound calls in IP from the Voice Gateway and require this physical T1 interface connection. Unified CCE agents receive their calls directly over the IP voice network.

Parent/Child Deployment

The parent/child model is illustrated in the following figure.
Figure 31. Parent/Child Model for Integrating a Traditional ACD with a Unified CCE Site

In this model, the Unified ICM Enterprise parent has PGs connected to a Unified CCE (using System PG) at one site (with a complete installation) and a second site with a TDM ACD that is using a Unified ICM TDM ACD PG. In this model, Unified ICM still provides virtual enterprise-wide routing, call treatment, and queuing with the distributed Unified CVP Voice Gateways at the sites. Unified ICM also has full visibility to all the sites for agents and calls in progress. The difference in this model is that Unified CCE provides local survivability. If it loses connection to the Unified ICM parent, the calls are still treated locally just as they are at the TDM ACD site.

Traditional IVR Integration

There are numerous ways that traditional IVRs can be integrated into a Unified CCE deployment. Determination of the best way depends on many factors that are discussed in the following sections. The primary consideration is determining how to eliminate or reduce IVR double trunking when transferring the call from the IVR.

Using PBX Transfer

Many call centers have existing traditional IVR applications that they are not prepared to rewrite. To preserve these IVR applications and integrate them into a Unified CCE environment, the IVR must have an interface to Unified CCE. (See Figure 37)

There are two versions of the IVR interface to Unified CCE. One is simply a post-routing interface (Call Routing Interface or CRI), which just allows the IVR to send a post-route request with call data to Unified CCE. Unified CCE returns a route response instructing the IVR to transfer the call elsewhere. In this scenario, the traditional IVR invokes a PBX transfer to release its port and transfer the call into the Unified CCE environment. Any call data passed from the IVR is passed by Unified CCE to the agent desktop or Unified IP IVR.

The other IVR interface to Unified CCE is the Service Control Interface (SCI). The SCI enables the IVR to receive queuing instructions from Unified CCE. In the PBX model, the SCI is not required.

Even if the IVR has the SCI interface, it is still preferable to deploy Unified CVP or Unified IP IVR for all call queuing because this prevents any additional use of the traditional IVR ports. In addition, use of the Unified IP IVR for queuing provides a way to re-queue calls on subsequent transfers or RONA treatment.

Figure 32. Traditional IVR Integration Using PBX Transfer

In this design, calls come first to the PBX from the PSTN carrier network on a standard T1 trunk interface. The PBX typically uses a hunt group to transfer the call to the IVR, putting all of the IVR ports into the hunt group as agents in auto available mode. The PBX looks like the PSTN to Unified CCE because it does not have a PG connected to the PBX. Unified CCE cannot track the call from the original delivery to the IVR; and it has reporting only from the time the call arrived at the IVR and the IVR informed Unified CCE of the call.

When the caller opts out of the IVR application, the IVR sends a Post-Route to Unified CCE using the Call Routing Interface (CRI). Because this application does not require calls to queue in the IVR, the CRI is the preferred interface option. Unified CCE looks at the agent states across the system and select the agent to send the call to (by using agent phone number or device target) or translation-route the call to the Unified IP IVR for queuing.

When the call is sent to an agent or into queue, it is hair-pinned in the PBX, coming in from the PSTN on a T1 trunk port and then going out to a Voice Gateway on a second T1 trunk port in the PBX. This connection is used for the life of the call.

Alternatively, if you want to track the call from its entry at the PBX or if you need to capture the caller ANI or original dialed number, you can install a PG on the PBX. The PBX can request (through a Post-Route to Unified CCE) which IVR port to send the call to behind the PBX. The PBX cannot use a hunt group to deliver the call from the PBX to the IVR. Unified CCE requires direct DNIS termination to ensure that the translation route maintains the call data collected in the PBX and makes it available to the IVR.

Using PSTN Transfer

This model is very similar to the previous model, except that the IVR invokes a PSTN transfer (instead of a PBX transfer) so that the traditional IVR port can be released. Again, the Unified IP IVR is used for all queuing so that any additional occupancy of the traditional IVR ports is not required and so that any double trunking in the IVR is avoided. Any call data collected by the traditional IVR application is passed by Unified CCE to the agent desktop or Unified IP IVR.

Figure 33. Traditional IVR Integration Using PSTN Transfer

In this model, the TDM IVR is set up as a farm of IVR platforms that have direct PSTN connections for inbound calls. The IVR has a PG connection to Unified CCE which tracks all calls in the system. When a caller opts out of the IVR treatment, the IVR sends a post-route request to Unified CCE which returns a label that directs the call either to an agent or to the Unified IP IVR for queuing.

The label that is returned to the TDM IVR instructs it to send an in-band transfer command using transfer tones (*8 with a destination label in the carrier network). The IVR out-pulses these tones to the service provider with tone generation or plays the tones by using a recorded file.

Using IVR Double Trunking

If your traditional IVR application has a very high success rate where most callers are completely self-served in the traditional IVR and only a very small percentage of callers ever need to be transferred to an agent, then it might be acceptable to double-trunk the calls in the traditional IVR for that small percentage of calls. (See Figure 39)

Unlike the previous model, if the traditional IVR has a Service Control Interface (SCI), then the initial call queuing is done on the traditional IVR. The reason this is beneficial is to queue the call on the Unified IP IVR, a second traditional IVR port is used to transfer the call to the Unified IP IVR. By performing the initial queuing on the traditional IVR, only one traditional IVR port is used during the initial queuing of the call. However, any subsequent queuing as a result of transfers or RONA treatment must be done on the Unified IP IVR to avoid any double trunking.

If the traditional IVR does not have an SCI interface, then the IVR just generates a post-route request to Unified CCE to determine where the call is transferred. All queuing in that scenario has to be done on the Unified IP IVR.

Figure 34. Traditional IVR Integration Using IVR Double Trunking

In this model, the TDM IVR is set up as a farm of IVR platforms that have direct PSTN connections for inbound calls. The IVR has a PG connection to Unified CCE which tracks all calls in the system. When a caller opts out of the IVR treatment, the IVR sends a post-route request to Unified CCE which returns a label that either directs the call to an agent or queues the call locally on the TDM IVR using the Service Control Interface (SCI). The transfer to the agent is done by the TDM IVR selecting a second port to hairpin the call to the Voice Gateway and to the Unified CCE agent. This takes up two ports for the time the call is at the agent.

Using Unified CM Transfer and IVR Double Trunking

Over time, it might become desirable to migrate the traditional IVR applications to Unified CVP or Unified IP IVR. However, if a small percentage of traditional IVR applications still exist for very specific scenarios, then the IVR can be connected to a second Voice Gateway (see Figure 40). Calls arriving at the Voice Gateway from the PSTN are routed by Unified CM. Unified CM can route specific DNs to the traditional IVR or let Unified CCE, Unified CVP, or Unified IP IVR determine when to transfer calls to the traditional IVR. If calls in the traditional IVR need to be transferred to a Unified CCE agent, then a second IVR port, trunk, and Voice Gateway port are used for the duration of the call. Ensure that transfer scenarios do not allow multiple loops to be created because voice quality suffers.

Figure 35. Traditional IVR Integration Using Unified CM Transfer and IVR Double Trunking

In this model, the TDM IVR is front-ended by either Unified CVP using the Voice Gateway or the Unified IP IVR and Unified CM with Unified CCE to determine the location to provide call treatment.

With Unified CVP, calls coming into the Voice Gateway immediately start a routing dialog with Unified CCE using the Service Control Interface (SCI). Based on the initial dialed number or prompting in Unified CVP, Unified CCE decides if the call needs to be sent to the TDM IVR for a specific self-service application or if Unified CVP has the application available for the caller. If the call was sent to the TDM IVR, the TDM IVR sends a route request to Unified CCE when the caller opts out. The reply is not sent back to the TDM IVR but back to Unified CVP as the original routing client. Unified CVP then takes the call leg away from the TDM IVR and transfers it to the Unified CCE agent over the VoIP network or holds it in a queue locally in the Voice Gateway.

With Unified CM, calls coming into the Voice Gateway hit a CTI route point for Unified CM to send a route request to Unified CCE to determine the appropriate call treatment device for the caller. If the CTI route point indicated an application that still is on the TDM IVR, Unified CCE instructs Unified CM to transfer the call to the TDM IVR by hairpinning the call using a second T1 port on the Voice Gateway to connect to the TDM IVR. Unified CCE can also instruct Unified CM to translation-route the call to the Unified IP IVR for call processing or prompting and then make a subsequent transfer to the TDM IVR for further processing. When the caller opts out of the TDM IVR, it sends a post-route request to Unified CCE, and Unified CCE returns a label to the TDM IVR. This label instructs the TDM IVR to transfer the call using a second T1 port on the IVR and to pass the call back to the Voice Gateway and over to the Unified CCE agent under the Unified CM dial plan.

In the model controlled by Unified CM, calls are initially received by the Voice Gateway and are hair-pinned to the TDM IVR on a second T1 port. When the IVR sends the call back to the Unified CCE agent, it uses a second TDM IVR port and a third port on the Voice Gateway. All three ports are used on the Voice Gateway as long as the agent is talking with the caller and both of the TDM IVR ports are used for the duration of this call.

Unified CCE/CCH: Integration with the Genesys Cisco T-Server

Unified CCE integration with the Genesys Cisco T-Server provides the following capabilities:
  • Allows integration of the Genesys Agent Desktop with Unified CCE/H deployments
  • Allows Unified CCE to be the site ACD with Genesys Enterprise Routing
  • Allows Unified CCE to be a backup ACD in the event of a Genesys disconnect, WAN failure, or other failure.

The deployment models listed in this document are supported for the integrated solution. See the Cisco Unified Contact Center Enterprise (Unified CCE) Software Compatibility Guide.

Note the following deployment considerations:
  • In an integrated Unified CCE/Genesys deployment, Cisco Unified CCE or Genesys Universal Routing Server (URS) can perform enterprise level routing—but not both
  • If Genesys URS is the enterprise routing engine, Unified CCE can only support IP-IVR as the local queuing platform
  • CVP as the queuing platform is only supported with Genesys in CTI/Desktop only mode
  • If Unified CCE is the enterprise routing engine, the Genesys T-Server can only be used in a CTI/Desktop only mode
  • Genesys Agent Desktop cannot be used for Unified CCE Mobile Agents (only Cisco CTI OS Desktop can be used)
  • Cisco Outbound Option is not supported by any Unified CCE/Genesys integrated deployment
  • Due to Unified CCE and Genesys having very different reporting architecture and terminology, the reports from Unified CCE and Genesys should not be used for correlation purposes:
    • In case of Unified CCE routing (Genesys is CTI only), there is no impact to Unified CCE reporting data
    • In case of Genesys routing, Genesys enterprise reporting is used
  • For deployments leveraging Genesys routing, Unified CCE must be configured to provide backup routing for the failure scenario where Unified CCE loses connectivity with the Genesys T-Server
  • The Unified CCE configuration data is not synchronized with Genesys (both need to be configured separately)

Note


Current deployments cannot mix Genesys and CTI-OS desktops. CTI OS may be installed for backup ACD functionality but cannot be operated concurrently.


  • All standard Clustering across the WAN (COW) deployments are supported with the added caveat of when the T-Servers are split across the WAN, or in the case of a single T-Server that can link to CTI-Server Across the WAN, the Genesys T-Server deployment guidelines should be followed.
  • For all diagrams, Genesys Desktops can be exchanged with a CTI-OS desktop.
Figure 36. Unified CCE with Genesys CTI/Desktop Only - No Genesys Routing

Figure 37. Unified CCE Parent/Child with Genesys CTI/Desktop Only - No Genesys Routing

Figure 38. Unified CCH with Genesys CTI/Desktop Only

Figure 39. Independent Unified CCE Sites with Genesys Routing

Figure 40. Distributed Unified CCE Sites with Genesys Routing