Cisco Unified Communications SRND Based on Cisco Unified Communications Manager 5.x
Call Processing
Downloads: This chapterpdf (PDF - 737.0KB) The complete bookPDF (PDF - 15.41MB) | Feedback

Call Processing

Table Of Contents

Call Processing

Unified CM Cluster Guidelines

Hardware Platforms

Unified CM Cluster Services

Intracluster Communications

Intracluster Security

Publisher

Call Processing Subscriber

TFTP Server

CTI Manager

IP Voice Media Streaming Application

Voice Activity Detection

Unified CM Applications

Unified CM Platform Capacity Planning

Capacity Calculations

Cisco Unified CM Capacity Tool

Gatekeeper Design Considerations

Hardware Platform Selection

Gatekeeper Redundancy

Hot Standby Router Protocol (HSRP)

Gatekeeper Clustering (Alternate Gatekeeper)

Directory Gatekeeper Redundancy

Interoperability of Unified CM and Unified CM Express

Multisite IP Telephony Deployments Using Unified CM and Unified CME with SIP Trunks

Multisite IP Telephony Deployments Using Unified CM and Unified CME, with H.323 Trunks and an IP-to-IP Gateway


Call Processing


Last revised on: September 27, 2007

 

This chapter provides design guidance for scalable and resilient call processing systems with Cisco Unified Communications Manager 5.0. This chapter also discusses how to choose the appropriate hardware and deployment scenario for Cisco Unified Communications Manager (Unified CM) based on specific requirements that include the following:

Scale — The number of users, gateways, applications, and so forth

Performance — The call rate

Resilience — The amount of redundancy

This chapter focuses on the following topics:

Unified CM Cluster Guidelines

This section discusses the minimum hardware requirements for Unified CM. This section also explains the various feature services that can be enabled on the Unified CM server and the purpose of each.

Unified CM Platform Capacity Planning

This section provides guidelines for using the Cisco Unified CM Capacity Tool, which must be used when planning an IP Telephony deployment. The Cisco Unified CM Capacity Tool provides guidance on resources used on a Unified CM server, based on certain deployment requirements.

Gatekeeper Design Considerations

This section explains how gatekeepers can be used in an IP Telephony deployment. Cisco Gatekeeper may also be paired with other standby gatekeepers or may be clustered for higher performance and resilience. Gatekeepers may also be used for call routing and call admission control.

Interoperability of Unified CM and Unified CM Express

This section explains the H.323 and SIP integration between Cisco Unified CM and Cisco Unified Communications Manager Express (Unified CME) in a distributed call processing deployment.

Unified CM Cluster Guidelines

The Unified CM architecture enables a group of physical servers to work together as a single IP PBX system. This grouping of servers is known as a cluster. A cluster of Unified CM servers may be distributed across an IP network, within design limitations, allowing for spatial redundancy and, hence, resilience to be designed into the IP Communications system.

This section describes the various functions performed by the servers that form a Unified CM cluster, and it provides guidelines for deploying the servers in ways that achieve the desired scale, performance, and resilience.

Hardware Platforms

Unified CM clusters utilize various types of servers, depending on the scale, performance, and redundancy required. They range from non-redundant, single-processor servers to highly redundant, multi-processor units.

Table 8-1 lists the general types of servers you can use in a cluster, along with their main characteristics.

Table 8-1 Types of Cisco Unified CM Servers 

Server Type
Cisco Server Model
Characteristics

Standard server (not high availability)

MCS 7815 or equivalent

Single processor

Single power supply

Non-RAID hard disk

High-availability standard server

MCS 7825 or equivalent

Single processor

Multiple power supplies

Single SCSI RAID hard disk array

High-performance server

MCS 7835 and MCS 7845 or equivalent

Multiple processors

Multiple power supplies

Multiple SCSI RAID hard disk arrays


Cisco Unified CM 5.0 is supported on specific Cisco MCS 7815, MCS 7825, MCS 7835, and MCS 7845 servers or on customer-provided HP and IBM servers that have been verified by Cisco to meet the following minimum requirements:

Processor speed must be 2.0 GHz or greater

Physical memory size must be 2 GB or greater

Physical hard disk size must be 72 GB or larger

For a complete list of currently supported hardware configurations, refer to the documentation available at

http://www.cisco.com/go/swonly

Servers should be deployed in an environment that provides high availability, not only for the IP network but also for power and cooling. Servers should be powered from an uninterruptible power supply (UPS) if building power does not have the required availability. Servers with dual power supplies could also be plugged into two different power sources to avoid the failure of one power circuit causing the server to fail.

Connectivity to the IP network should also ensure maximum performance and availability. The Unified CM servers should be connected to the Ethernet at 100 Mbps full-duplex. If 100 Mbps is not available on smaller deployments, then use 10 Mbps full-duplex. Many servers also include the capability of using Gigabit Ethernet, which is also an option. Ensure that servers are connected to the network using full-duplex, which can be achieved with 10 Mbps and 100 Mbps by hard-coding the switch port and the server NIC. For 1000 Mbps, Cisco recommends using Auto/Auto for speed and duplex configuration on both the NIC and the switch port. The default for Cisco Unified CM 5.0 is Auto/Auto, and this setting is also the default following an upgrade from a previous Unified CM release.


Note A mismatch will occur if either the server port or the Ethernet switch port is left in Auto mode and the other port is configured manually. The best practice is to configure both the server port and the Ethernet switch port manually, with the exception of Gigabit Ethernet ports which should be set to Auto/Auto.


NIC Teaming for Network Fault Tolerance

Hewlett-Packard (HP) server platforms with dual Ethernet network interface cards (NICs) can support NIC teaming for Network Fault Tolerance with Cisco Unified CM 5.0. This feature allows a server to be connected to the Ethernet via two NICs and, hence, two cables. NIC teaming prevents network downtime by transferring the workload from the failed port to the working port. NIC teaming cannot be used for load balancing or increasing the interface speed.

General Clustering Guidelines

The following guidelines apply to all Unified CM clusters:


Note A cluster may contain a mix of server platforms, but all servers in the cluster must run the same Unified CM software release.


Under normal circumstances, place all members of the cluster within the same LAN or MAN. Cisco does not recommend placing all members of a cluster on the same VLAN or switch.

For redundancy, the members of the cluster should be deployed in the following manner to minimize the impact of any failures in the infrastructure or building:

Different access switches connected to the same distribution or core switch

Different access switches attached to different distribution or core switches

Different buildings within the same LAN or MAN

If the cluster spans an IP WAN, follow the guidelines for clustering over an IP WAN as specified in the section on Clustering Over the IP WAN, page 2-17.

Unified CM Cluster Services

Within a Unified CM cluster, there are servers that provide unique services. Each of these services can coexist with others on the same physical server. For example, in a small system it is possible to have a single server be a database publisher, backup subscriber, music on hold (MoH) server, TFTP server, CTI Manager, and Conference Bridge. As the scale and performance requirements of the cluster increase, many of these services should be moved to a single, dedicated physical server.

With the release of Cisco Unified CM 5.0, a cluster may contain as many as 20 servers, of which a maximum of eight may run the Cisco CallManager Service that provides call processing. The other servers may be configured as a dedicated database publisher, dedicated Trivial File Transfer Protocol (TFTP) server, or music on hold (MoH) servers. Media streaming applications (conference bridge or media termination point) may also be installed on a separate server that registers with the cluster.

When deploying a cluster with Cisco MCS 7815 or equivalent servers, there is a minimum limit of two servers in a cluster: one as the publisher, TFTP server, and backup call processing server, and the other as the primary call processing server. A maximum of 500 phones is supported in this configuration with Cisco Unified CM 5.1 on a Cisco MCS 7815 (or a maximum of 300 phones with Cisco Unified CM 5.0 on the MCS 7815). When deploying a two-server cluster with higher-capacity servers, Cisco recommends that you do not exceed 1250 users in the cluster. Above 1250 users, a dedicated publisher and separate servers for primary and secondary call processing services is recommended, thus increasing the number of servers in a cluster.

It is also possible to deploy a single-server cluster with an MCS 7825 or greater servers. With an MCS 7825 or equivalent server, the limit is 500 users; with a higher-availability server, the single-server cluster should not exceed 1000 users. In a single-server configuration, there is no redundancy unless Survivable Remote Site Telephony (SRST) is also deployed to provide service during periods when the Unified CM is not available. Cisco does not recommend a single-server deployment for production environments. The load balancing option is not available when the publisher is a backup call processing subscriber.

Figure 8-1 illustrates a typical Unified CM cluster.

Figure 8-1 Typical Unified CM Cluster

Intracluster Communications

There are two primary kinds of communication within a Unified CM cluster (intracluster communications). (See Figure 8-2.) The first is a mechanism for distributing the database that contains all the device configuration information (see Database Replication in Figure 8-2). The configuration database is stored on a publisher server, and a read-only copy is replicated to the subscriber members of the cluster. Changes made on the publisher are communicated to the subscriber databases, ensuring that the configuration is consistent across the members of the cluster, as well as facilitating spatial redundancy of the database.

The second type of Intracluster communication is the propagation and replication of run-time data such as registration of devices, locations bandwidth, and shared media resources (see ICCS in Figure 8-2). This information is shared across all members of a cluster running the Cisco CallManager Service, and it assures the optimum routing of calls between members of the cluster and associated gateways.

Figure 8-2 Intracluster Communications

Intracluster Security

With the release of Cisco Unified CM 5.0, a different architecture and mechanism are implemented for securing the Unified CM applications and communication links. Each server in a cluster runs an internal dynamic firewall. The application ports on Unified CM are protected using source IP filtering. The dynamic firewall opens these applications ports only to authenticated or trusted servers. (See Figure 8-3.)

Figure 8-3 Intracluster Security

This security mechanism is applicable only between servers in a single Unified CM cluster. Unified CM subscribers are authenticated in a cluster before they can access the publisher's database. The Intra-cluster communication and database replication take place only between authenticated servers. During the installation process, a subscriber is authenticated to the publisher using a pre-shared key authentication mechanism. The authentication process involves the following steps:

1. Install the publisher server using a security password.

2. Configure the subscriber server on the publisher by using Unified CM Administration.

3. Install the subscriber server using the same security password used during publisher server installation.

4. After the subscriber is installed, the server attempts to establish connection to the publisher on a management channel using UDP 8500. The subscriber sends all the credentials to the publisher, such as hostname, IP address, and so forth. The credentials are authenticated using the security password used during the install process.

5. The publisher verifies the subscriber's credentials using its own security password.

6. The publisher adds the subscriber as a trusted source to its dynamic firewall table if the information is valid. The subscriber is allowed access to the database.

7. The subscriber gets a list of other subscriber servers from the publisher. All the subscribers establish a management channel with each other, thus creating a mesh topology.

Publisher

The publisher is a required server in all clusters, and there can currently be only one per cluster. This server is the first to be installed and provides the database services to all other members in the cluster. The publisher server is the only server that has read and write access to the configuration database. When configuration changes are made, other members of the cluster have a read-only copy of the database. On larger systems with more than 1250 users, Cisco recommends a dedicated publisher to prevent administrative operations from affecting the telephony users. A dedicated publisher does not have any call processing services or TFTP services running on the server. Other servers run the TFTP and Unified CM services.

Servers in the cluster will attempt to use the publisher's database when initializing. If the publisher is not available, they will use the local read-only copy from their hard drives.

If the system is running but the publisher is not available, the following operations will not be available:

Call forwarding changes

Operations that requires licensing service

Configuration changes

Extension Mobility login or logout operations

Extension Mobility will not function without the publisher because it requires access to the read/write database. Therefore, Cisco recommends running this service on the publisher only.

The choice of hardware platform for the publisher is based on the scale and performance of the cluster. Cisco recommends that the publisher have the same performance capability as the call processing subscribers. Ideally the publisher should also be a high-availability server to minimize the impact of a hardware failure.

Call Processing Subscriber

When installing the Unified CM software, you can define two types of servers, publisher and subscriber. These terms are used to define the database relationship during installation. Once the software is installed, only the database and network services are enabled. All subscribers will subscribe to the publisher to obtain a read-only copy of the database information.

A call processing subscriber is a server that has the Cisco CallManager Service enabled. A single server license is required to enable this service on a subscriber. The Cisco CallManager Service cannot be enabled on a server if the publisher is not available because the publisher acts as a licensing server and distributes the licenses needed to activate the Cisco CallManager Service. Once this service is enabled, the server is able to perform call processing functions. Devices such as phones, gateways, and media resources can register and make calls only to servers with this service enabled. Cisco Unified CM 5.0 supports up to eight servers in a cluster with the Cisco CallManager Service enabled.

Depending on the redundancy scheme chosen (see Call Processing Redundancy), the call processing subscriber will be either a primary (active) subscriber or a backup (standby) subscriber. In the load-balancing option, the subscriber can be both a primary and backup subscriber. When planning the design of a cluster, you should generally dedicate the call processing subscribers to this function. In larger-scale or higher-performance clusters, the call processing service should not be enabled on the publisher and TFTP server. Call processing subscribers normally operate in either dedicated pairs or shared pairs, depending on the redundancy scheme adopted. One-to-one redundancy uses dedicated pairs, while two-to-one redundancy uses two pairs of servers that share one server from each pair (the backup server).

The choice of hardware platform depends on the scale, performance, redundancy, and cost of the servers. Scale and performance are covered in the section on Unified CM Platform Capacity Planning, and redundancy is covered in the section on Call Processing Redundancy.

Call Processing Redundancy

With Cisco Unified CM 5.0, you can choose from the following redundancy configurations:

Two to one (2:1) — For every two primary subscribers, there is one shared backup subscriber.

One to one (1:1) — For every primary subscriber, there is a backup subscriber.

The 1:1 redundancy scheme allows upgrades with only the failover periods impacting the cluster. The failover mechanism has been enhanced so that you can achieve failover rates for Skinny Client Control Protocol (SCCP) IP phones of approximately 125 registrations per second. The failover mechanism for Session Initiation Protocol (SIP) phones is approximately 40 registrations per second.

With the release of Cisco Unified CM 5.0, a cluster can be upgraded without impacting the services. Unified CM 5.0 allows two versions of Unified CM to be on the same server, one in the active partition and other in the inactive partition. All services and devices use the Unified CM version in the active partition for all Unified CM functionality. During the upgrade process, the cluster operations continue using its current release of Unified CM in the active partition, while the upgrade version gets installed in the inactive partition. Once the upgrade process is completed, the servers can be rebooted to switch the inactive partition to the active partition, thus running the new version of Unified CM.

Upgrading from Unified CM 5.0.x to Unified CM 5.0.x

The 1:1 redundancy scheme enables you to upgrade the cluster using the following method:


Step 1 Install the new version of Unified CM on the publisher. Do not reboot.

Step 2 Install the new version of Unified CM on all subscribers simultaneously. Do not reboot.

Step 3 Reboot only the publisher. Switch to the new version of Unified CM and allow some time for the database to initialize.

Step 4 Reboot the TFTP server(s) one at a time. Switch to the new version of Unified CM and wait for the configuration files to be rebuilt before upgrading any further servers in the cluster.

Step 5 Reboot the dedicated music on hold (MoH) server(s) one at a time. Switch to the new version of Unified CM.

Step 6 Reboot the backup subscriber(s) one at a time. Switch to the new version of Unified CM. This step might impact some users if 50/50 load balancing is implemented.

Step 7 Fail-over the devices from the primary subscribers to their backups.

Step 8 Reboot the primary subscriber(s) one at a time. Switch to the new version of Unified CM.


With this upgrade method, there is no period (except for the failover period) when devices are registered to subscriber servers that are running different versions of the Unified CM software.


Note Once the upgrade process starts on the inactive partition, any changes made in the active partition of the publisher's database will not be migrated to the new version of the database.


The 2:1 redundancy scheme allows for fewer servers in a cluster, but it can potentially result in an outage during upgrades.


Note You must use 1:1 redundancy when 10,000 or more IP phones are registered on the two primary subscribers because there cannot be more than 10,000 backup registrations on a single backup subscriber.



Note Before you do an upgrade, Cisco recommends that you back up the Unified CM and Call Detail Record (CDR) database to an external network directory using the Disaster Recovery Framework. This practice will prevent any loss of data if the upgrade fails.


Call Processing Subscriber Redundancy

The following figures illustrate typical cluster configurations to provide call processing redundancy with Unified CM.

Figure 8-4 Basic Redundancy Schemes

Figure 8-4 illustrates the two basic redundancy schemes available. In each case the backup server must be capable of handling the capacity of at least a single primary call processing server failure. In the 2:1 redundancy scheme, the backup might have to be capable of handling the failure of a single call processing server or potentially both primary call processing servers, depending on the requirements of a particular deployment. Sizing the capacity of the servers and choosing the hardware platforms is covered in the section on Unified CM Platform Capacity Planning.

Figure 8-5 1:1 Redundancy Configuration Options

In Figure 8-5 the five options shown all indicate 1:1 redundancy. Option 1 is used for clusters supporting less than 1250 users. Options 2 through 5 illustrate increasingly scalable clusters. The exact scale will depend on the hardware platforms chosen or required.

Note that the illustrations show only publisher and call processing subscribers.

Figure 8-6 2:1 Redundancy Configuration Options

Load Balancing

Normally a backup server has no devices registered to it unless its primary is unavailable. This model allows for:

Easier troubleshooting — Because all call processing takes place on primary servers, obtaining traces and alert notifications becomes easier.

Less configuration — Because all the devices are registered to the primary server, the need to define additional Unified CM redundancy groups or device pools for the various devices can be reduced by 50%.

The 1:1 redundancy scheme enables you to balance distribution of the devices over the primary and backup server pairs. With load balancing, you can move up to half of the device load from the primary to the secondary subscriber by using the Unified CM redundancy groups and device pool settings. This model allows for:

Load sharing — The call processing load is distributed on multiple servers, which can provide faster response time.

Faster failover and failback — Because all devices (such as IP phones, CTI ports, gateways, trunks, voicemail ports, and so forth) are distributed across all active subscribers, only some of the devices fail over to the secondary subscriber if the primary subscriber fails. In this way, you can reduce by 50% the impact of any server becoming unavailable.

To plan for 50/50 load balancing, calculate the capacity of a cluster without load balancing, and then distribute the load across the primary and backup subscribers based on devices and call volume. To allow for failure of the primary or the backup server, the total load on the primary and secondary subscribers should not exceed that of a single subscriber server.

TFTP Server

The TFTP server performs two main functions:

The serving of files for services such as MoH, configuration files for devices such as phones and gateways, binary files for the upgrade of phones as well as some gateways, and various security files.

Generation of configuration and security files. Most files generated by the Cisco TFTP service are signed and in some cases encrypted before being available for download.

The TFTP service can be enabled on any server in the cluster. However, in a cluster with more than1250 users, other services might be impacted by configuration changes that can cause the TFTP service to regenerate configuration files. Therefore, Cisco recommends that you dedicate a specific server to the TFTP service in a cluster with more than 1250 users, with Extension Mobility, or with other features that cause configuration changes.

The TFTP server is used by phones and MGCP gateways to obtain configuration information. There is no restriction on the number of servers that can have TFTP service enabled, however Cisco recommends deploying 2 TFTP servers for a large cluster, thus providing redundancy for TFTP service. More than 2 TFTP servers can be deployed in a cluster, but this can result in an extended period for rebuilding of all TFTP files on all TFTP servers. When configuring the TFTP options using DHCP or statically, you can normally define an IP address array (more than one IP address) for a TFTP server. Therefore, you can assign half of the devices to use TFTP server A as the primary and TFTP server B as the backup, and the other half to use TFTP server B as the primary and TFTP server A as the backup. To improve performance on dedicated TFTP servers, you can set service parameters to increase the number of simultaneous TFTP sessions allowed on the server.

When upgrading a Unified CM cluster, Cisco highly recommends that you upgrade the TFTP servers after the publisher and before any other server, also allowing additional time following the upgrade for the TFTP server to rebuild all the configuration files. Either use the typical Cisco TFTP - BuildDuration time or use the real-time monitoring tool to monitor the Cisco TFTP - DeviceBuildCount until it stops incrementing. This upgrade order ensures that any new binaries and configuration changes are available before the upgrade of other services in the cluster. If you are manually adding a specific binary or firmware load for a phone or gateway, be sure to copy the file to each TFTP server in the cluster.

Cisco Unified CM Release 5.0, by default, caches the configuration files in memory and does not store them on the hard drive of the TFTP server. This default setting can be changed to place the configuration files on the hard drive of the TFTP server, but doing so will impact TFTP performance. Therefore, Cisco recommends that you do not change this default setting.

Cisco recommends that you use the same hardware platform for the TFTP servers as used for the call processing subscribers.

CTI Manager

CTI Manager is required in a cluster for applications that use TAPI or JTAPI Computer Telephony Integration (CTI). The CTI Manager acts as a broker between the CTI application and the Cisco CallManager Service. It provides authentication of the application and enables control or monitoring of authorized devices. The CTI application communicates with a primary CTI Manager and, in the event of a failure, will switch to a backup CTI Manager. The CTI Manager should be enabled only on call processing subscribers, thus allowing for a maximum of eight CTI Managers in a cluster. Cisco recommends that you load-balance CTI applications across the various CTI Managers in the cluster to provide maximum resilience, performance, and redundancy.

Generally, it is good practice to associate devices that will be controlled or monitored by an application with the same server pair used for the CTI Manager. For example, an interactive voice response (IVR) application requires four CTI ports. They would be provisioned as follows, assuming the use of 1:1 redundancy and 50/50 load balancing:

Two CTI Ports would have a Unified CM redundancy group of server A as the primary and server B as the backup (or secondary). The other two ports would have a Unified CM redundancy group of server B as the primary and server A as the backup.

The IVR application would be configured to use the CTI Manager on server A as the primary and server B as the backup.

The above example allows for redundancy in case of failure of the CTI Manager on server A and also allows for the IVR call load to be spread across two servers. This approach also minimizes the impact of a Unified CM server failure.

IP Voice Media Streaming Application

Media resources such as conferencing and music on hold may be provided by IP Voice Media Streaming Application services running on the same physical server as the Cisco CallManager Service.

Media resources include:

Music on Hold (MoH) — Provides multicast or unicast music to devices that are placed on hold or temporary hold, transferred, or added to a conference. (See Music on Hold, page 7-1.)

Annunciator service — Provides announcements in place of tones to indicate incorrectly dialed numbers or call routing unavailability. (See Annunciator, page 6-18.)

Conference bridges — Provide software-based conferencing for ad-hoc and meet-me conferences. (See Audio Conferencing, page 6-8.)

Media termination point (MTP) services — Provide features for H.323 clients, H.323 trunks, and Session Initiation Protocol (SIP) trunks. (See Media Termination Point (MTP), page 6-13.)

Because of the additional processing and network requirements for media, it is essential to follow all guidelines for running media resources within a cluster. Generally, Cisco recommends non-dedicated servers for multicast MoH and the annunciator, but a dedicated media resource server is recommended for large-scale software-based conferencing and MTP unless those services are within the design guidelines detailed in the chapters on Media Resources, page 6-1, and Music on Hold, page 7-1.

Voice Activity Detection

Cisco also recommends that you leave voice activity detection (VAD) disabled within the cluster. VAD is disabled by default in the Unified CM service parameters, and you should disable it on H.323 and SIP dial peers by using the no vad command.

Unified CM Applications

Various types of applications can be enabled on Unified CM. This section covers only the scalability aspect of Unified CM applications. For detailed design guidance, see the chapter on Cisco Unified CM Applications, page 22-1.

The Unified CM applications include the following:

Cisco Unified Communications Manager Assistant

The Cisco Unified CM Assistant application operates in conjunction with the CTI Manager Service. When Cisco Unified CM Assistant is used in a cluster, the expected capacity and performance can impact the choice of Unified CM and CTI subscriber hardware platform. If other CTI applications are used in a cluster, the number of CTI connections supported in a cluster might limit the maximum Unified CM Assistant configuration.

Currently supported limits for Unified CM Assistant in Cisco Unified CM Release 5.0 are as follows:

Maximum of two Unified CM Assistant servers in a cluster

Maximum of four CTI servers under Unified CM Assistant service parameters

1250 Assistants and 1250 Managers per cluster with Cisco MCS 7845 servers

For additional information on Unified CM Assistant capacity, refer to the Cisco Unified CM and Unified CM Assistant data sheets, documentation, and release notes available at

http://www.cisco.com

Cisco Extension Mobility

When Extension Mobility is used in a cluster, the expected capacity and performance can impact the choice of the publisher hardware platform. As a user logs in or logs out from a phone, the configuration must be updated in the configuration database, the TFTP service has to regenerate the configuration file, and then the device is reset to make the change. Most of this activity takes place on the publisher.

Currently supported limits for Extension Mobility in Cisco Unified CM Release 5.0 are as follows:

A Cisco MCS-7845 publisher server can support 50 sequential logins and/or logouts per minute.

A Cisco MCS-7835 publisher server can support 30 sequential logins and/or logouts per minute.

For additional information on EM capacity, refer to the Cisco Unified CM data sheets, documentation, and release notes available at

http://www.cisco.com

Cisco Unified Communications Manager Attendant Console (AC)

The Cisco AC application interacts with the CTI Manager Service for line monitoring and phone control. When Cisco AC is used in a cluster, the expected capacity and performance can impact the choice of Unified CM and CTI subscriber hardware platform. If other CTI applications are used in a cluster, the number of CTI connections supported in a cluster might limit the maximum AC configuration.

Currently supported limits for AC in Cisco Unified CM Release 5.0 are as follows:

An MCS 7845 can support a maximum of 1250 AC devices. The Attendant Console application can be used in any combination of up to 1250 attendant console devices. For example, you may configure Unified CM to have 125 hunt pilots with 10 members in each hunt pilot or 50 hunt pilots with 25 members in each hunt pilot.

An MCS 7835 supports a maximum of 1000 AC devices, and an MCS 7825 supports a maximum of 750 AC devices.

Unified CM Platform Capacity Planning

Many types of devices can register with Unified CM; for example, IP phones, voicemail ports, CTI (TAPI or JTAPI) devices, gateways, and DSP resources such as transcoding and conferencing. Each of these devices requires resources from the server platform with which it is registered. The required resources can include memory, processor usage, and disk I/O. Each device then consumes additional server resources during transactions, which are normally in the form of calls. For example, a device that makes only 6 calls per hour consumes fewer resources than a device making 12 calls per hour.

The recommendations provided in this section are based on calculations made using the Unified CM Capacity Tool, with default trace levels and CDRs enabled. Higher levels of performance can be achieved by disabling, reducing, or reconfiguring other functions that are not directly related to processing calls. Increasing some of these functions can also have an impact on the call processing capabilities of the system. These functions include tracing, call detail recording, highly complex dial plans, and other services that are co-resident on the server. Highly complex dial plans can include multiple line appearances as well as large numbers of partitions, calling search spaces, route patterns, translations, route groups, hunt groups, pickup groups, route lists, extensive use of call forwarding, co-resident services, and other co-resident applications. All of these functions can consume additional resources within the Unified CM server.

To improve system performance, the following techniques can provide useful options:

Install additional certified memory in the server, up to the maximum supported for the particular platform. Doubling the RAM in MCS 7825 and MCS 7835 or equivalent servers is recommended in large configurations for that server class. Verification using a performance monitor will indicate if this memory upgrade is required. As the server approaches maximum utilization of physical memory, the operating system will start to swap to disk. This swapping is a sign that additional physical memory should be installed.

A Unified CM cluster with a very large dial plan containing many gateways, route patterns, translation patterns, and partitions, can take an extended amount of time to initialize when the Cisco CallManager Service is first started. If the system does not initialize within the default time, there are service parameters that can be modified to allow additional time for the configuration to initialize. For details on the service parameters, refer to the online help for Service Parameters in Unified CM Administration.

The following guidelines apply to Cisco Unified CM Release 5.0:

Within a cluster, you may enable a maximum of 8 servers with the Cisco CallManager Service. Other servers may be used for more dedicated functions such as TFTP, publisher, music on hold, and so forth.

You can configure a maximum of 800 CTI connections or associations per standard server, or a maximum of 3200 per cluster if they are equally balanced among the servers.

You can configure a maximum of 2500 CTI connections or associations per high-performance server, or a maximum 10,000 per cluster if they are equally balanced among the servers.

Each cluster can support a maximum of 30,000 SCCP or SIP phones.

Each cluster can support a maximum of 600 H.323 devices (gateways, trunks, and clients), digital MGCP devices, and SIP trunks

Capacity Calculations

The Cisco Unified Communications Manager Capacity Tool (Unified CMCT) for Cisco Unified CM Release 5.0 enables you to calculate the capacity of the system for various configurations. The capacity planning tool is currently available to all www.cisco.com users with a login account. If your system does not meet the following guidelines, or if you consider the system to be more complex and would like to verify the capacity but you do not have access to the Cisco Unified CM Capacity Tool, please contact your Cisco Systems Engineer (SE) or the Cisco Technical Assistance Center (TAC).

The Cisco Unified CM Capacity Tool is available at

http://www.cisco.com/partner/WWChannels/technologies/resources/CallManager/

If your system meets the following requirements, you should not need to verify your configuration with the Cisco Unified CM Capacity Tool:

System contains less than 25% of the maximum number of users for the server platforms

Average number of lines per phone does not exceed 1.1

Average busy hour call attempts (BHCA) per user is below 4 calls per hour

Cluster security is not enabled

No more than 20% trunking (5 users per trunk) either via gateways or trunks

No more than 5% voicemail ports (20 users per voicemail port)

No more than 20 MoH streams per server

No CTI, JTAPI, or TAPI devices

No more than 5% conference bridges (20 users per port on a bridge)

No transcoders

Only MTPs are used to support the maximum calls required for trunking

Less than 20 locations

System contains nothing else that is not defined above and is not an IP Phone, IP Communicator, gateway, media resource, voicemail port, or trunk

The maximum number of users Unified CM can support depends on the server platform, as indicated in Table 8-2.

Table 8-2 Maximum Number of Devices per Server Platform 

Server Platform Characteristics
Maximum Users per Server 1
High-Availability Server 2
High-Performance Server

Cisco MCS-7845 (All supported models)

7500

Yes

Yes

Cisco MCS-7835 (All supported models)

2500

Yes

No

Cisco MCS-7825 (All supported models)

1000

No

No

Cisco MCS-7815 (All supported models)3

5004

No

No

1 A platform that is not a high-availability server can support a maximum of 500 IP Phones in a non-redundant installation.

2 A high-availability server supports redundancy for both the power supplies and the hard disks.

3 MCS-7815 servers support only N+1 redundancy and may not be a member of a cluster.

4 With Cisco Unified CM 5.0, the MCS 7815 server supports a maximum of 300 users.


For the latest information on supported platforms, third-party platforms, and specific hardware configurations, refer to the online documentation at

http://www.cisco.com/go/swonly


Note The maximum supported non-redundant installation on platforms that are not highly available is 500 IP phones.


Cisco Unified CM Capacity Tool

The Cisco Unified CM Capacity Tool requires various pieces of information to provide a calculation of the minimum size and type of servers required for a system. The information includes the type and quantity of devices, such as IP phones, gateways, and media resources. For each device type, the capacity tool also requires the average BHCA and the average utilization time. For example, if all IP phones make an average of 5 calls per hour and the average call lasts 3 minutes, then the BHCA is 5 and the utilization is 0.25. (Five calls of 3 minutes each equals 15 minutes per hour on the phone, which is 0.25 hour.)

In addition to the device information, the capacity tool also requires information regarding the dial plan, such as route patterns and translation patterns. When all the details have been entered, the capacity tool will calculate how many primary servers of the desired server type are required, as well as the number of clusters if the required capacity exceeds a single cluster.

The capacity tool replaces the previous mechanisms known as device weights, BHCA multipliers, call type multipliers, and dial plan weights.

When you use the Cisco Unified CM Capacity Tool, the results will indicate the highest capacity of the various resources calculated for the configuration. The resources include hard limits, memory, processor utilization, and disk I/O activity. If you add more devices, it might look as if no more capacity is being used because the additional configuration is using resources that are below the current highest capacity indicated.

For example, if 2500 third-party controlled SCCP IP phones were added to the Cisco Unified CM Capacity Tool, it would show that a single MCS 7845 subscriber is required and is at 100% capacity. If an additional 1000 phones are added, the results still indicate a single MCS 7845 subscriber is required and is at 100% capacity. This result is due to the fact that CTI capacity for this server is at 100%, and additional IP phones do not add CTI requirements. Until the number of other devices exceeds one of the other limits, it will appear as if no additional capacity is being used. This is normal and expected behavior for the Cisco Unified CM Capacity Tool.

Use the guidelines in the following sections to provide information for the capacity calculator.

Phone Calculations

The main Telephony section (not the Contact Center section) of the Capacity Tool lists the following phone types:

SCCP Phones (Non-secure)

Secure SCCP Phones

SIP Phones (non-secure)

Secure SIP Phones

For each phone type, you can enter quantity, BHCA, utilization, and line appearance values, which are further explained in the following sections.

Quantity

This value is the total number of IP phones of each type that will be configured in the cluster. The quantity includes all Cisco 7900 Series IP Phones, VG248 ports, VG224, IP Communicator, and other third-party SCCP endpoint devices. The quantity must include all configured phones even if they are not active.

BHCA

This value is the average BHCA of all phones of each type. If you have shared lines across multiple phones, the BHCA should include one call for each and every phone that shares that line. Shared lines across multiple phone types will affect the BHCA for each type. (That is, one call to a shared line will be calculated as multiple calls, one to each phone that rings.) If you have different groups of phones that generate different BHCAs, use the following method to provide a BHCA value for use in the Cisco Unified CM Capacity Tool.

For example, assume there are two classes of user with the following characteristics:

100 phones at 20 BHCA = 2,000 BHCA total

5000 phones at 4 BHCA = 20,000 BHCA total

The total BHCA for all phone devices in this case is 22,000.

Then divide the total BHCA by the total number of phone devices, to yield:

Average BHCA per phone device = 22,000 / 5,100 = 4.31 BHCA

Utilization

This value is the average call utilization per phone, and it includes all the time that a call is on the phone. The actual utilization of a phone can exceed 100% if that phone allows multiple calls per line or has multiple line appearances that also are frequently utilized. Utilization is measured as a percentage of an hour. For example, a phone that is on a call for 3 minutes in the busiest hour is considered to be 5% utilized. The same method for calculating BHCA can also be used to calculate the average utilization for various groups of phone. If the phone has shared lines, then only the expected utilization for that phone should be calculated and not the actual utilization for the line that is shared. The Cisco Unified CM Capacity Tool currently accepts an average utilization for all phones.

Lines Appearances

This value is the average number of lines for all the phones. If a phone has multiple appearances of the same DN in different partitions, it will be counted as multiple line appearances. Shared lines should be counted as one line, but ensure that the correct BHCA and utilization are calculated. Multiple calls per line do not increase the line appearance count, but they do affect the BHCA and utilization calculations. For example, if it is normal for a phone to have two calls, one active and the other on hold for various amounts of time, then the utilization will be higher for that device because two calls are actually connected.

Gateways

Quantity of Gateways

The quantity of gateways is split into several entries because it depends on gateway type, as follows:

MGCP T1/E1 gateways

This value is the total number of gateways that need to be configured in the Unified CM database. For example, a Cisco IOS MGCP gateway can have multiple T1s or E1s, but they would be added as a single gateway. A Cisco WS-6608 module would be added as a gateway per port that is configured as a T1 or E1 gateway, up to the maximum of eight per module.

MGCP analogue gateways

This value is the total number of analogue gateways to be added in the Unified CM database. Generally, an entire module (WS-6624 or Unified CM) or hardware platform (Cisco IOS Router platform) would be added as a single gateway.

H.323 gateways

An entire module or hardware platform would be added as a single gateway. This quantity does not include H.323 gateways that are not defined in Unified CM but are used via H.323 trunks.


Note SIP gateways are added as trunks.


Number of DS0s

This value is the total number of DS0s or analogue ports supported by each type of gateway. The number of DS0s is split between the different types of gateways as follows:

T1 CAS:

24 * (Number of T1 CAS spans)

T1 (E1) PRI:

(23 or 30) * (Total number of PRIs)

H.323 gateways

(Total number of DS0s) / (Number of calls supported on all digital, analogue, or IP interfaces)

BHCA

The BHCA is the average for all DS0s or analogue ports on a gateway during the busiest hour. The method for calculating this average is the same as the method used for the phone BHCA.

Utilization

This value is the average utilization of all DS0s or analogue ports during the busiest hour.

EM Profiles

Extension Mobility (EM) profiles contain line appearances that should also be included in the phone calculations. They will not affect the quantity of devices, but they will increase the average number of line appearances per phone. The BHCA and utilization of an EM user should already have been calculated for the phone they will log into.

H.323 and SIP Trunks

Quantity of Trunks

This value is the total number of trunks that will be configured in the Unified CM database. A gatekeeper controlled trunk is not affected by the possible number of destinations and, therefore, will be counted once for each configured gatekeeper controlled trunk. This situation is also the case for Session Initiation Protocol (SIP) trunks using a SIP proxy. Each SIP gateway that is not connected via a SIP Proxy will require a SIP trunk to be defined in Unified CM.

Quantity of Calls

This value is the total number of concurrent calls expected to be allowed on all trunks. The number of allowed calls will generally be controlled by locations or gatekeeper call admission control. Keep in mind that regions and codecs can also affect the total number of calls allowed.

Utilization

This value is the average utilization of all calls across all trunks. It is a percentage of the busiest hour, where 75% utilization means that calls will be active for 45 minutes per hour.

MTP Requirements

MTP requirements must also be considered separately in the calculator. If you have MTP required on an H.323 or SIP trunk, then an MTP resource is required for every concurrent call on that trunk. RSVP agents should be included in the quantity and will also be based on the maximum number of sessions supported.

CTI Route Points, Ports, and Third-Party Controlled Lines

CTI route point calculations include the total quantity, average BHCA, and utilization values. Any assigned directory numbers should be added to the Dial Plan section of the tool.

CTI ports can be used in a variety of ways by the applications controlling them. For simplicity they have been divided into two groups:

Simple call or redirect call

A simple call is a call either to or from the port to another device, with no supplementary service being invoked. It is the typical call scenario where one party calls another, the called party answers, the call proceeds, and the parties hang up.

A redirect is a call to a CTI port that is not actually answered. Instead, it is forwarded to another destination or redirected. This is even simpler than a simple call because no call is connected, so no media is connected to the CTI port.

Transfer or conference

These call types are typical of an IVR or auto-attendant application.

A transfer type call is characterized as a call that is connected to a CTI port, then at some point the CTI port transfers the caller without consultation (or blind transfer) to another destination.

A conference is a variation of the transfer, whereby a call connected to the CTI port is either transferred with consultation (three-way call) or is conferenced with another party for the duration of the call.

Third-Party Controlled Lines

Each application that requires third-party control or monitoring of a device or line must be counted and entered in this field of the calculator. If more than one application is monitoring or controlling the same device, it will need to be counted more than once.

Gatekeeper Design Considerations

A single Cisco IOS gatekeeper can provide call routing and call admission control for up to 100 Unified CM clusters in a distributed call processing environment. Multiple gatekeepers can be configured to support thousands of Unified CM clusters. You can also implement a hybrid Unified CM and toll-bypass network by using Cisco IOS gatekeepers to provide communication and call admission control between the H.323 gateways and Unified CM.

Gatekeeper call admission control is a policy-based scheme requiring static configuration of available resources. The gatekeeper is not aware of the network topology, so it is limited to hub-and-spoke topologies.

The Cisco 2600, 2800, 3600, 3700, 3800, and 7200 Series routers all support the gatekeeper feature. You can configure Cisco IOS gatekeepers in a number of different ways for redundancy, load balancing, and hierarchical call routing. This section considers the design requirements for building a gatekeeper network, but it does not deal with the call admission control or dial plan resolution aspects, which are covered in the chapters on Call Admission Control, page 9-1, and Dial Plan, page 10-1, respectively.

For additional information regarding gatekeepers, refer to the Cisco IOS H.323 Configuration Guide, available at

http://www.cisco.com/en/US/products/sw/iosswrel/ps5207/products_configuration_guide_book09186a00801fcee1.html

Hardware Platform Selection

The choice of gatekeeper platform is based on the number of calls per second and the number of concurrent calls. A higher number of calls per second requires a more powerful CPU, such as a Cisco 3700, 3800, or 7200 Series Router. A higher number of concurrent calls requires more memory. For the latest information on platform selection, contact your Cisco partner or Cisco Systems Engineer (SE).

Gatekeeper Redundancy

With gatekeepers providing all call routing and admission control for intercluster communications, redundancy is required. Prior to Cisco Unified CM Release 3.3, the only method for providing gatekeeper redundancy was Hot Standby Router Protocol (HSRP); however, beginning with Cisco Unified CM Release 3.3, gatekeeper clustering and redundant gatekeeper trunks are also available as methods of providing gatekeeper redundancy. The following sections describe these methods.


Note Cisco recommends that you use gatekeeper clustering to provide gatekeeper redundancy whenever possible. Use HSRP for redundancy only if gatekeeper clustering is not available in your software feature set.


Hot Standby Router Protocol (HSRP)

Hot Standby Router Protocol (HSRP) is the only option for gatekeeper redundancy with Cisco Unified CM prior to Release 3.3. HSRP does not provide the features required to build a redundant and scalable gatekeeper network, therefore it should be used only in Unified CM environments prior to Release 3.3.

The following guidelines apply to HSRP:

Only one gatekeeper is active at a time.

The standby gatekeeper does not process any calls unless the primary fails.

No load balancing features are available.

All gatekeepers must reside in the same subnet or location.

No previous state information is available after failover.

After a failover, the standby gatekeeper is not aware of the calls that are already active, so over-subscription of the bandwidth is possible.

Failover time can be substantial because the endpoints have to re-register with the HSRP standby gatekeeper before calls can be placed. The failover time depends on the settings of the registration timers.

Figure 8-7 show a network configuration using HSRP for gatekeeper redundancy.

Figure 8-7 Gatekeeper Redundancy Using HSRP

Example 8-1 shows the configuration for Gatekeeper 1 and Example 8-2 shows the configuration for Gatekeeper 2 in Figure 8-7. Both configurations are identical except for the HSRP configuration on the Ethernet interface.

Example 8-1 Configuration for Gatekeeper 1

interface Ethernet0/0
 ip address 10.1.10.2 255.255.255.0
  standby ip 10.1.10.1
  standby priority 110
 
gatekeeper
 zone local GK-Site1 customer.com 10.1.10.1
 zone local GK-Site2 customer.com 
 zone prefix GK-Site1 408.......
 zone prefix GK-Site2 212.......
 bandwidth interzone default 160
 gw-type-prefix 1#* default-technology
 arq reject-unknown-prefix
 no shutdown

Example 8-2 Configuration for Gatekeeper 2

interface Ethernet0/0
 ip address 10.1.10.3 255.255.255.0
 standby ip 10.1.10.1 
 
gatekeeper
 zone local GK-Site1 customer.com 10.1.10.1
 zone local GK-Site2 customer.com 
 zone prefix GK-Site1 408.......
 zone prefix GK-Site2 212.......
 bandwidth interzone default 160
 gw-type-prefix 1#* default-technology
 arq reject-unknown-prefix
 no shutdown

The following notes also apply to Example 8-1 and Example 8-2:

Each router has standby commands configured for HSRP and to identify the virtual IP address shared by each. Gatekeeper 1 is configured as the primary with the command standby priority 110.

Each Unified CM cluster has a local zone configured on each router to support Unified CM trunk registrations. Note that the IP address defined on the first zone should match the virtual IP address used by HSRP.

A zone prefix is configured for each zone on both routers, allowing inter-zone and inter-cluster call routing.

Bandwidth statements are configured on each router for both sites. Cisco recommends that you use the bandwidth interzone command because the bandwidth total command does not work in some configurations.

The gw-type-prefix command is configured on both routers, allowing all locally unresolved calls to be forwarded to a device registered with a technology prefix of 1# in the local zone. In this example, all Unified CM trunks have been configured to register with a 1# prefix.

The arq reject-unknown-prefix command is configured on both routers to guard against potential call routing loops across redundant Unified CM trunks.

For additional and advanced HSRP information, refer to the online documentation at the following locations:

http://www.cisco.com/en/US/docs/internetworking/case/studies/cs009.html

http://www.cisco.com/en/US/tech/tk648/tk362/technologies_q_and_a_item09186a00800a9679.shtml

http://www.cisco.com/en/US/tech/tk648/tk362/technologies_tech_note09186a0080094afd.shtml

Gatekeeper Clustering (Alternate Gatekeeper)

Gatekeeper clustering (alternate gatekeeper) enables the configuration of a "local" gatekeeper cluster, with each gatekeeper acting as primary for some Unified CM trunks and an alternate for others. Gatekeeper Update Protocol (GUP) is used to exchange state information between gatekeepers in a local cluster. GUP tracks and reports CPU utilization, memory usage, active calls, and number of registered endpoints for each gatekeeper in the cluster. Load balancing is supported by setting thresholds for any of the following parameters in the GUP messaging:

CPU utilization

Memory utilization

Number of active calls

Number of registered endpoints

With the support of gatekeeper clustering (alternate gatekeeper) and Cisco Unified CM Release 3.3 or later, stateful redundancy and load balancing is available. Gatekeeper clustering provides the following features:

Local and remote clusters

Up to five gatekeepers in a local cluster

Gatekeepers in local clusters can be located in different subnets or locations

No failover delay (Because the alternate gatekeeper is already aware of the endpoint, it does not have to go through the full registration process.)

Gatekeepers in a cluster pass state information and provide load balancing

Figure 8-8 shows three sites with Unified CM distributed call processing and three distributed gatekeepers configured in a local cluster.

Figure 8-8 Gatekeeper Clustering

In Figure 8-8, Gatekeeper 2 is the backup for Gatekeeper 1, Gatekeeper 3 is the backup for Gatekeeper 2, and Gatekeeper 1 is the backup for Gatekeeper 3.

Example 8-3 shows the configuration for Gatekeeper 1 (SJC), and Example 8-4 shows the configuration for Gatekeeper 2 (CHC). The configuration for Gatekeeper 3 (NYC) is not shown because it is very similar to the other two.

Example 8-3 Gatekeeper Clustering Configuration for Gatekeeper 1

gatekeeper
 zone local SJC cisco.com 10.1.1.1
 zone local CHC_GK1 cisco.com
 zone local NYC_GK1 cisco.com
!
 zone cluster local SJC_Cluster SJC
  element SJC_GK2 10.1.2.1 1719
  element SJC_GK3 10.1.3.1 1719
 !
 zone cluster local CHC_Cluster CHC_GK1
  element CHC 10.1.2.1 1719
  element CHC_GK3 10.1.3.1 1719
 !
 zone cluster local NYC_Cluster NYC_GK1
  element NYC 10.1.3.1 1719
  element NYC_GK2 10.1.2.1 1719
 !
 zone prefix SJC 40852.....
 zone prefix NYC_GK1 21251.....
 zone prefix CHC_GK1 72067.....
 gw-type-prefix 1#* default-technology
 load-balance cpu 80 memory 80
 bandwidth interzone SJC 192
 bandwidth interzone NYC_GK1 160
 bandwidth interzone CHC_GK1 160
 arq reject-unknown-prefix
 no shutdown

Example 8-4 Gatekeeper Clustering Configuration for Gatekeeper 2

gatekeeper
 zone local CHC cisco.com 10.1.2.1
 zone local SJC_GK2 cisco.com 
 zone local NYC_GK2 cisco.com
!
 zone cluster local CHC_Cluster CHC
  element CHC_GK3 10.1.3.1 1719
  element CHC_GK1 10.1.1.1 1719
!
 zone cluster local SJC_Cluster SJC_GK2
  element SJC  10.1.1.1 1719
  element SJC_GK3 10.1.3.1 1719
 !
  zone cluster local NYC_Cluster NYC_GK2
  element NYC_GK1 10.1.1.1 1719
  element NYC 10.1.3.1 1719
 !
 zone prefix SJC_GK2 40852.....
 zone prefix NYC_GK2 21251.....
 zone prefix CHC 72067.....
 gw-type-prefix 1#* default-technology
 load-balance cpu 80 memory 80
 bandwidth interzone CHC_Voice 160
 bandwidth interzone SJC_Voice2 192
 bandwidth interzone NYC_Voice3 160
 arq reject-unknown-prefix
 no shutdown

The following notes also apply to Example 8-3 and Example 8-4:

Each Unified CM cluster has a local zone configured to support Unified CM trunk registrations.

A cluster is defined for each local zone, with backup zones on the other gatekeepers listed as elements. Elements are listed in the order in which the backups are used.

A zone prefix is configured for each zone to allow inter-zone and inter-cluster call routing.

The gw-type-prefix command allows all locally unresolved calls to be forwarded to a device registered with a technology prefix of 1# in the local zone. In this example, all Unified CM trunks have been configured to register with a 1# prefix.

The load-balance cpu 80 memory 80 command limits CPU and memory usage. If the router hits either limit, all new requests are denied and the first backup in the list is used until utilization drops below the threshold.

Bandwidth statements are configured for each site. Cisco recommends that you use the bandwidth interzone command because the bandwidth total command does not work in some configurations.

The arq reject-unknown-prefix command guards against potential call routing loops across redundant Unified CM trunks.

All gatekeepers in the cluster display all Unified CM trunk registrations. For trunks that use the gatekeeper as a primary resource, the flag field is blank. For trunks that use another gatekeeper in the cluster as their primary gatekeeper, the flag field is set to A (alternate). Having all endpoints registered as primary or alternate allows all calls to be resolved locally without having to send a location request (LRQ) to another gatekeeper.

Example 8-5 shows the output from the show gatekeeper endpoints command on Gatekeeper 1 (SJC).

Example 8-5 Output for Gatekeeper Endpoints

                    GATEKEEPER ENDPOINT REGISTRATION
                    ================================
CallSignalAddr  Port  RASSignalAddr   Port  Zone Name         Type        Flags
--------------- ----- --------------- ----- ---------         ----        -----
10.1.1.12       1307  10.1.1.12       1254  SJC               VOIP-GW
    H323-ID: SJC-to-GK-trunk_1
10.1.1.12       4422  10.1.1.12       4330  SJC               VOIP-GW    
    H323-ID: SJC-to-GK-trunk_2
10.1.2.12       4587  10.1.2.12       4330  CHC_GK1           VOIP-GW     A
    H323-ID: CHC-to-GK-trunk_1
10.1.3.21       2249  10.1.3.21       1245  NYC_GK1           VOIP-GW     A
    H323-ID:  NYC-to-GK-trunk_1
Total number of active registrations = 4

Directory Gatekeeper Redundancy

You can implement directory gatekeeper redundancy by using HSRP or by configuring multiple identical directory gatekeepers. When a gatekeeper is configured with multiple remote zones using the same zone prefix, the gatekeeper can use either of the following methods:

Sequential LRQs (default)

Redundant remote zones (matching zone prefixes) are assigned a cost, and LRQs are sent to the matching zones in order based on the cost values. Using sequential LRQs saves WAN bandwidth by not blasting LRQs to all matching gatekeepers.

LRQ Blast

LRQs are sent to redundant zones (matching zone prefixes) simultaneously. The first gatekeeper to respond with an Location Confirm (LCF) is the one that is used.

Cisco recommends that you use multiple active directory gatekeepers with sequential LRQs, thus allowing directory gatekeepers to be placed in different locations. Using HSRP requires both directory gatekeepers to be located in the same subnet, and only one gatekeeper can be active at any time.

Figure 8-9 illustrates a Unified CM distributed call processing environment with two active directory gatekeepers.

Figure 8-9 Redundant Directory Gatekeepers

Example 8-6 and Example 8-7 show the configurations for the two directory gatekeepers in Figure 8-9.

Example 8-6 Configuration for West Directory Gatekeeper

gatekeeper
 zone local DGKW customer.com 10.1.10.1
 zone remote SJC customer.com 10.1.1.1
 zone remote CHC customer.com 10.1.2.1
 zone remote NYC customer.com 10.1.3.1
 zone prefix SJC 408.......
 zone prefix CHC 720.......
 zone prefix NYC 212.......
 lrq forward-queries
 no shutdown

Example 8-7 Configuration for East Directory Gatekeeper

gatekeeper
 zone local DGKE customer.com 10.1.12.1
 zone remote SJC customer.com 10.1.1.1
 zone remote CHC customer.com 10.1.2.1
 zone remote NYC customer.com 10.1.3.1
 zone prefix SJC 408.......
 zone prefix CHC 720.......
 zone prefix NYC 212.......
 lrq forward-queries
 no shutdown

The following notes also apply to Example 8-6 and Example 8-7:

Both directory gatekeepers are configured exactly the same.

A local zone is configured for the directory gatekeeper.

Remote zones are configured for each remote gatekeeper.

Zone prefixes are configured for both remote zones for inter-zone call routing. The wildcard (*) could be used in the zone prefix to simplify configuration, but the use of dots (.) is more specific. Calls are not routed to the DGK zone, so a prefix is not required for it.

The lrq forward-queries command allows the directory gatekeeper to forward an LRQ received from another gatekeeper.


Note Directory gatekeepers do not contain any active endpoint registrations and do not supply any bandwidth management.


Example 8-8, Example 8-9, and Example 8-10 show the configurations for Gatekeepers 1 to 3 in Figure 8-9.

Example 8-8 Configuration for Gatekeeper 1 (SJC)

zone local SJC customer.com 10.1.1.1
 zone remote DGKW customer.com 10.1.10.1
 zone remote DGKE customer.com 10.1.12.1
 zone prefix SJC 408.......
 zone prefix DGKW ..........
 zone prefix DGKE ..........
 bandwidth remote 192
 gw-type-prefix 1# default-technology
 arq reject-unknown-prefix
 no shutdown

Example 8-9 Configuration for Gatekeeper 2 (CHC)

gatekeeper
 zone local GK-CHC customer.com 10.1.2.1
 zone remote DGKE customer.com 10.1.12.1
 zone remote DGKW customer.com 10.1.10.1
 zone prefix CHC 720.......
 zone prefix DGKE ..........
 zone prefix DGKW ..........
 bandwidth remote 160
 gw-type-prefix 1# default-technology
 arq reject-unknown-prefix
 no shutdown

Example 8-10 Configuration for Gatekeeper 3 (NYC)

gatekeeper
 zone local NYC customer.com 10.1.3.1
 zone remote DGKE customer.com 10.1.12.1
 zone remote DGKW customer.com 10.1.10.1
 zone prefix NYC 212.......
 zone prefix DGKE ..........
 zone prefix DGKW ..........
 bandwidth remote 160
 gw-type-prefix 1# default-technology
 arq reject-unknown-prefix
 no shutdown

The following notes also apply to Example 8-8, Example 8-9, and Example 8-10:

Each Unified CM cluster has a local zone configured to support Unified CM trunk registrations.

Remote zones are configured for each directory gatekeeper.

Zone prefixes are configured for the local zone and both remote zones for inter-zone call routing. Both directory gatekeeper prefixes are 10 dots. Sequential LRQs are used by default when matching zone prefixes are configured. The gatekeeper sends an LRQ to the directory gatekeeper with the lowest cost; if there is no response, the gatekeeper tries the second directory gatekeeper.

The bandwidth remote command is used to limit bandwidth between the local zone and any other remote zone.

The gw-type-prefix command allows all locally unresolved calls to be forwarded to a device registered with a technology prefix of 1# in the local zone. In this example, all Unified CM trunks have been configured to register with a 1# prefix.

The arq reject-unknown-prefix command guards against potential call routing loops across redundant Unified CM trunks.

Interoperability of Unified CM and Unified CM Express

This section explains the requirements for interoperability and internetworking of Cisco Unified CM with Cisco Unified Communications Manager Express (Unified CME, formerly called Cisco IOS Telephony Services, or ITS) using H.323 or SIP protocol in a multisite IP telephony deployment. This section highlights the recommended deployments between phones controlled by Unified CM and phones controlled by Unified CME.

Cisco Unified CME 3.4 adds the ability to configure Cisco IP SIP Phones 7905, 7912, 7940, and 7960 in addition to the currently supported Cisco IP SCCP Phones 7902, 7905, 7910, 7912, 7920, 7935, 7936, 7940, 7960, 7970, 7971, and Cisco IP Communicator. If you use Unified CME with SIP phones, the WAN interface must be SIP. Unified CME SCCP phones are supported with either an H.323 or SIP WAN interface.

All call signaling is sent through Unified CME, regardless of the endpoint being used. However, with SCCP endpoints on the same Unified CME, the media can flow around Unified CME; whereas with SIP endpoints on the same Unified CME, media always flows through Unified CME.

The following sections provide guidelines for achieving interoperability between Unified CM and Unified CME:

Multisite IP Telephony Deployments Using Unified CM and Unified CME with SIP Trunks

Multisite IP Telephony Deployments Using Unified CM and Unified CME, with H.323 Trunks and an IP-to-IP Gateway

For more information about Unified CME, refer to the product documentation for Cisco Unified CME, available online at

http://www.cisco.com

Multisite IP Telephony Deployments Using Unified CM and Unified CME with SIP Trunks

Unified CM can communicate directly with Unified CME using a SIP interface. Figure 8-10 shows an IP Telephony deployment with Unified CM networked directly with Cisco Unified CME using a SIP trunk WAN interface.

Figure 8-10 Multisite IP Telephony Deployment Using Unified CM and Unified CME with SIP Trunks

Best Practices

Follow these guidelines and best practices when using the deployment model illustrated in Figure 8-10:

Configure a SIP Trunk Security Profile with Replaces header acceptance.

Configure a SIP trunk on Unified CM using the SIP Trunk Security Profile created, and also specify a ReRouting CSS. The ReRouting CSS is used to determine where a SIP user (transferor) can refer another user (transferee) to a third user (transfer target) and which features a SIP user can invoke using the SIP 3XX Redirection Response and INVITE with Replaces.

For SIP trunks there is no need to enable the use of media termination points (MTPs) when using SCCP endpoints on Unified CME. However, SIP endpoints on Unified CME require the use of media termination points on Unified CM to be able to handle delayed offer/answer exchanges with the SIP protocol (that is, the reception of INVITEs with no Session Description Protocol).

Use the Unified CM dial plan configuration (route patterns, route lists, and route groups) to send calls to the SIP trunk that connects to Unified CME.

Use Unified CM device pools and regions to configure a G.711 codec within the site and the G.729 codec for remote Unified CME sites.

Configure the allow-connections sip to sip command under voice services voip on Unified CME to allow SIP-to-SIP call connections.

For SIP endpoints, configure the mode cme command under voice register global, and configure dtmf-relay rtp-nte under the voice register pool commands for each SIP phone on Unified CME.

For SCCP endpoints, configure the transfer-system full-consult command and the transfer-pattern .T command under telephony-service on Unified CME.

Configure the SIP WAN interface voip dial-peers to forward calls, destined for Unified CM, with session protocol sipv2 and dtmf-relay sip-notify rtp-nte on Unified CME.


Note When multiple PSTN connections exist (one for Unified CM and one for Unified CME), fully attended transfer between a Unified CM endpoint and a Unified CME endpoint to a PSTN endpoint will fail. The recommendation is to use blind transfer when using multiple PSTN connections, and it is configured under telephony-service as transfer-system full-blind.


Example 8-11 lists a sample configuration for Unified CME using this SIP deployment model.

Example 8-11 Configuration for Cisco Unified CME 3.4 with SIP

voice service voip
    allow-connections sip to sip
    sip
	registrar server
dial-peer voice 1 voip       /* To Unified CM endpoints */
    destination-pattern xxxx
    session protocol sipv2
    session target ipv4:10.10.10.20
    session transport udp    /* tcp can be used here also */
    dtmf-relay rtp-nte
    codec g729r8             /* Voice class can also be used */
    no vad
voice register global
    mode cme
    source-address 10.10.10.21 port 5060
voice register pool 1
    id mac 0007.0E8B.5777
    type 7940
    number 1 dn 1
    codec g729r8	      /* Voice class can also be used */
    dtmf-relay rtp-nte
telephony-service
 ip source-address 10.10.10.22 port 2000
 create cnf-files
 keepalive 45
 max-conferences 8 gain -6
 moh music-on-hold.au
 transfer-system full-consult /* full-blind can also be used */
 transfer-pattern .T

Using the guidelines presented here ensures that basic calls can be completed between Unified CM and Unified CME phones. By using SIP trunk interfaces instead of H323, the need for MTPs is eliminated if you are using only SCCP endpoints on Unified CME. If SIP endpoints are used on Unified CME, the configuration of an MTP on Unified CM is required.

Multisite IP Telephony Deployments Using Unified CM and Unified CME, with H.323 Trunks and an IP-to-IP Gateway

An IP-to-IP gateway is a separate router that provides a proxy (or front end) for a system that does not support H.450, such as Unified CM. The IP-to-IP gateway can be placed between Unified CM and a Unified CME router, and it provides H.323-to-H.323 call connections to terminate and re-originate calls for transfers and forwards to endpoints that do not support H.450.

Unified CME can be directly integrated with Unified CM using an H.323 interface without the IP-to-IP gateway. However, because Unified CM does not support H.450 specifications, supplementary services such call transfer and call forward initiated from Unified CM phones will require the media to be hair-pinned (over the WAN in some cases) through Unified CM even though the Unified CM phones might not be involved in the call. The IP-to-IP gateway ensure that this media hair-pinning does not occur over the WAN.

The IP-to-IP gateway can also act as the PSTN gateway for the remote Cisco Unified CME systems as well as for Unified CM. A separate PSTN gateway is not required in such cases.

The IP-to-IP gateway must run a Cisco IOS release that is compliant with Cisco Unified CME 3.4 and that supports H.450, such as Cisco IOS Release 12.4(6)T or later with the IP VOICE feature set.

Figure 8-11 shows an IP Telephony deployment with Unified CM connected to Cisco Unified CME through an IP-to-IP gateway.

Figure 8-11 Multisite IP Telephony Deployment with Unified CM, Unified CME, and an IP-to-IP Gateway

Best Practices

Follow these guidelines and best practices when using the deployment model illustrated in Figure 8-11:

Configure a gatekeeper controlled H.225 trunk between Unified CM and the IP-to-IP gateway, with Media Termination Point Required checked and Wait For Far End H.245 Terminal Capability Set unchecked.

In Unified CM, set the service parameter Send H225 user info message to H225 info for Call Progress Tone.

Use the Unified CM dial plan configuration (route patterns, route lists, and route groups) to send calls to the H.225 trunk that connects to the IP-to-IP gateway.

Register Cisco Unified CME and the IP-to-IP gateway as H.323 gateways on the gatekeeper.

A media termination point (MTP) is required for H.450 to perform Empty Capability Set (ECS) signaling conversion. The MTP should be configured on the IP-to-IP gateway and registered to Unified CM. An MTP can cause media to be hair-pinned over the WAN in certain rare cases. For more details on media termination points, see the chapter on Media Resources, page 6-1.

Configure the allow-connection h323 to h323 command on the IP-to-IP gateway to allow H.323-to-H.323 call connections. Remote Unified CME routers need not have this command enabled.


Note When using H.323 trunks between Unified CME and Unified CM, configure the allow-connection h323 to h323 command on Unified CME to allow H.323-to-H.323 hair-pinned call connections. The Unified CME auto-detection feature (starting with Unified CME 3.1) disables all H.450-based communications on the interface connected to Unified CM. To complete the call transfer or forward between Unified CM and Unified CME phones, an H.323-to-H.323 connection is required.


Define VoIP dial peers on the IP-to-IP gateway to route calls to the Unified CM and Unified CME endpoints.

Define VoIP dial peers on Unified CME to forward calls, destined for Unified CM endpoints, to the IP-to-IP gateway.

Supplementary services such as transfer and call forward will result in calls being media hair-pinned when the two endpoints reside in the same Unified CME branch location.


Note When multiple PSTN connections exist (one for Unified CM and one for Unified CME), fully attended transfer between a Unified CM endpoint and a Unified CME endpoint to a PSTN endpoint will fail. The recommendation is to use blind transfer when using multiple PSTN connections, and it is configured under telephony-service as transfer-system full-blind.


Example 8-12 lists a sample configuration for the IP-to-IP gateway.

Example 8-12 Configuration for IP-to-IP Gateway

voice service voip
    allow-connections h323 to h323
    supplementary-service h450.2
    supplementary-service h450.3
    supplementary-service h450.12
    h323
	emptycapability
    h225 id-passthru
    h225 connect-passthru
    h245 passthru tcsnonstd-passthru
dial-peer voice 1 voip        /* To Unified CM endpoints */
    destination-pattern xxxx
    session target ipv4:y.y.y.y
    dtmf-relay h245-alphanumeric
    codec g729r8
    no vad
dial-peer voice 1 voip        /* To Unified CM endpoints */
    destination-pattern zzzz
    session target ras        /* "ras" if gatekeeper is used, otherwise "ipv4:a.b.c.d" */
    dtmf-relay h245-alphanumeric
    codec g729r8
    no vad

Example 8-13 lists a sample configuration for the in this H.323 deployment model.

Example 8-13 Configuration for Cisco Unified CME 3.4 with H.323

voice service voip
    h323
interface FastEthernet0/1
    ip address 10.10.10.23 255.255.255.0
    h323-gateway voip interface
    h323-gateway voip id cme ipaddr 10.10.10.30 1719 
dial-peer voice 1 voip       /* To Unified CM endpoints */
    destination-pattern xxxx
    session target ras
    session transport tcp   
    codec g729r8             /* Voice class can also be used */
    no vad
telephony-service
 ip source-address 10.10.10.22 port 2000
 create cnf-files
 keepalive 45
 max-conferences 8 gain -6
 moh music-on-hold.au
 transfer-system full-blind /* Used with multiple PSTN connections */
 transfer-pattern .T