Table of Contents

Configuring the Cisco ICS 7750

Many tasks are required for fully configuring the Cisco Integrated Communications System 7750 (Cisco ICS 7750) for data and voice routing. This chapter lists common tasks required to configure the Cisco ICS 7750, gives pointers to Cisco IOS and Cisco CallManager documentation that tells how to perform these tasks, and describes any differences between configuring Cisco IOS or Cisco CallManager software on the Cisco ICS 7750 and configuring Cisco IOS or Cisco CallManager on other platforms.

This chapter contains these sections:

Best Practices for Using the IOS CLI

ICS System Manager is designed to communicate with and to monitor the status of all the components in the chassis. To enable ICS System Manager to perform these functions, a configuration program (ICSConfig) guides you through the configuration process. ICSConfig enables you to change key system parameters, such as the IP addresses of system cards, passwords, destination for syslog messages, and Simple Network Management Protocol (SNMP) community strings.

To enable ICS System Manager to properly function as a system management tool, it is important that you use ICSConfig or ICS System Manager, as appropriate, rather than the IOS command-line interface (CLI), when you enter key system parameters.

Except for the procedures listed in "ICSConfig Tasks," you can enter all IOS CLI commands that are available for use in any IOS software release that is intended for use on the Cisco ICS 7750.

ICSConfig Tasks

You should always use ICSConfig for the following tasks:

---

Note    SNMP community strings and system passwords are case sensitive.

---

The following list includes tasks that should never be configured on the Cisco ICS 7750 by using the IOS CLI under any circumstances:

Saving Configuration Changes

To prevent loss of the ASI or MRP configuration, save the running-config file to the startup-config file by following these steps:

Setting the System Date and Time

This section explains how to set the date and time on Cisco ICS 7750 cards. It contains the following tasks:

Setting the Date and Time on SPE310 Cards

Complete the following steps to set the date and time on Cisco System Processing Engine 310 (SPE310) cards:

Step 2   Use the Client Connection Manager to open a Terminal Services Client connection with the SPE310:

Step 3   Log in as an administrator (user ID administrator), and enter your password (the default is changeme).

Step 4   On the SPE310, choose Start > Settings > Control Panel > Date/Time.

The Date/Time Properties dialog opens.

Step 5   Fill in the necessary fields. Click OK to close the Date/Time Properties dialog box.

Step 6   Repeat Step 2 through Step 6 for any additional SPE310s, if present.

Setting the Date and Time on SSP, MRP, and ASI Cards

The system switch processor (SSP) card, multiservice route processor (MRP) cards, and analog station interface (ASI) cards each have a system clock that begins to run from the point at which the card starts up. The system clock keeps track of the date and time. The SSP stores its configuration data in Flash-simulated NVRAM. The MRP300, MRP3-8FXS, and MRP3-16FXS cards store their configuration data in NVRAM. The MRP200, ASI81, and ASI160 cards obtain their configuration data from the SPE310 running System Manager when they boot.

When you set the date and time, the setting remains accurate until the next card restart.

Note   When changing the date and time settings on SSP, MRP, and ASI cards, open a Telnet session from the PC, not from the SPE310. If you open a Telnet session from the SPE, the changes you make to the card configuration are not saved.

Complete the following steps to set the date and time on the SSP card and MRP cards:

Step 2   Enter the following command to open a Telnet session:

where IP address is the IP address of the card with which you wish to communicate.

Step 3   Enter your login password.

Step 4   Enter privileged EXEC mode by entering the following command:

Step 5   Enter your enable password.

Step 6   To enter global configuration mode, enter the following command:

Step 7   To set the time zone, enter the following command in global configuration mode:

where:

For example, to set the time to PST, eight hours offset from UTC, enter the following command:

Table 6-1 lists the time zones in North America and their offsets from UTC.

Step 8   To set the clock for a card, enter one of the following IOS commands in privileged EXEC mode:

or

where:

---

Note    This step must be performed after every reboot of the Cisco ICS 7750.

---

Step 9   To exit global configuration mode, enter the following command:

Step 10   To save your configuration, enter the following command:

Step 11   To verify your settings, enter the following command:

Step 12   Repeat Step 3 through Step 12 for additional cards, as necessary.

Step 13   Close the Telnet session by typing exit at the prompt.

Configuring the SSP

The SSP is an eight-port switching module in the Cisco ICS 7750. It has two external ports for connecting to external network devices and has six internal ports for connecting to the other cards in the Cisco ICS 7750.

The SSP serves the following purposes:

Features

The SSP provides the following features:

---

Note    Hot-swapping the SSP will result in system downtime since all the cards in the ICS 7750 chassis will lose connectivity during the swap.

---

SSP Configuration Tasks

The SSP is, in the default configuration, network-ready. In most network configurations, the SSP will not require any additional configuration. However, many settings on the SSP are configurable. Table 6-2 lists tasks that you may need or want to perform in order to configure the SSP. In addition, Table 6-2 gives pointers to the locations in the Catalyst 2900 XL and Catalyst 3500 Software Configuration Guide that provide instructions for performing those tasks. The Catalyst 2900 XL and Catalyst 3500 Software Configuration Guide is available at the following URL:

http://www.cisco.com/univercd/cc/td/doc/product/lan/c2900xl/29_35wc/sc/index.htm

Configuring MRPs and ASIs

This section explains how to configure MRP and ASI cards and contains the following sections:

System Card Overview

This section lists the key features of the MRP200, MRP300, ASI81, ASI160, MRP3-8FXS, and MRP3-16FXS system cards.

Key Features of MRP200 and MRP300 Cards

MRP200 and MRP300 cards have the following features:

Key Features of ASI81, ASI160, MRP3-8FXS, and MRP3-16FXS Cards

The key features of the ASI81, ASI160, MRP3-8FXS, and MRP3-16FXS cards are delineated below:

Supported WICs, VICs, and VWICs

For a list of the WICs, VICs, and VWICs that are supported in MRP200, MRP300, MRP3-8FXS, and ASI81 cards, refer to the Cisco ICS 7750 System Description . For information about valid combinations of WICs, VICs, and VWICs on MRP and ASI cards, see "PVDM Requirements."

Codec/DSP Overview

VICs and VWICs installed in MRP cards or ASI cards might require additional digital signal processors (DSPs) for processing heavier voice traffic. Each DSP can perform a maximum of 100 million instructions per second (MIPS).

You can install up to two packet voice/data modules (PVDMs) on each MRP or ASI card. PVDMs contain DSP chips that give MRP and ASI cards more processing power.

Voice Compression Algorithms (Codecs)

The Cisco ICS 7750 supports several options for voice-compression algorithms. These algorithms are commonly called codecs. The word codec is a combination of the words coder and decoder. Coding is the process of encoding a digitized signal into a more efficient form for transmission or storage. Decoding is the process of restoring the coded signal to the original form.

Codecs differ in terms of voice quality, compression rate and bandwidth, ability to carry dual-tone multifrequency (DTMF) and modem traffic, and number of channels (calls) that a single DSP can support. The more DSP channels, the greater the number of calls that an MRP or ASI card can support. The number of channels supported also depends on whether the DSP is running a digital image or it is running an analog image. (Digital T1 and E1 VWICs process digital signals, and analog VICs process analog signals.)

As Table 6-3 shows, some codec compression techniques require more processing power than others. Multiple DSP firmware images are available for use on MRP and ASI cards. High-complexity images support fewer calls than medium-complexity images.

G.711

G.711 performs pulse code modulation (PCM) and is the standard digital channel used in the public telephone network. PCM provides no compression and therefore no opportunity for bandwidth savings. Any services that operate over the public network should operate with similar performance over a Cisco ICS 7750 PCM channel (although the Cisco ICS 7750 connection might have more delay).

G.723.1

G.723.1 is a compression technique that uses multi-pulse, multi-level quantization (MP-MLQ) or code excited linear prediction (CELP) coding to compress speech or audio signal components at 5.3 or 6.3 kilobits per second (kbps), respectively. G.723.1, which is part of the H.324 family of standards, can be used for compressing speech or audio signal components at very low bit rates. G.723.1 Annex-A provides built-in voice activity detection (VAD) and Comfort Noise Generation (CNG).

G.726

G.726 performs adaptive differential pulse code modulation (ADPCM) coding. G.726 reduces network bandwidth requirements for transmitting voice by encoding 64 kbps voice channels as 32, 24, or 16 kbps ADPCM (that is, 32 kbps provides 2:1 compression, 24 kbps provides 3:1 compression, and 16 kbps provides 4:1 compression). Generally, there is a trade-off between the amount of compression and voice quality. ADPCM-encoded voice can be interchanged between packet voice, PSTN, and private branch exchange (PBX) networks if the PBX networks are configured to support ADPCM.

G.729

G.729 performs CELP coding, where voice is coded into 8-kbps streams. There are two variations of this standard (G.729 and G.729 Annex A [G.729a]) that differ mainly in terms of their computational complexity; both provide speech quality similar to 32-kbps ADPCM. G.729 is a high complexity algorithm, and G.729a is a medium complexity variant of G.729 with slightly lower voice quality. G.729a performs conjugate structure algebraic code excited linear predictive (CS-ACELP) coding, providing speech quality similar to 32-kbps ADPCM. G.729a offers the best compression rate (8:1), but it does not typically carry modem traffic, and it degrades DTMF and music signals somewhat. Depending on the type of traffic, using G.729a can produce cost savings of 40 percent, relative to using G.711. Other algorithms in the G.729 family include G.729 Annex-B, a high complexity algorithm, and G.729a Annex-B, a medium-complexity variant of G.729 Annex-B with slightly lower voice quality. The difference between the G.729 and G.729 Annex-B codecs is that G.729 Annex-B provides built-in VAD and CNG.

Codec Interoperability

Codec interoperability is the ability of one codec to decode another codec. If a DSP is configured with a certain codec, the DSP should be able to decode the voice codec using any codec with which the DSP is interoperable.

The following G.729 codec combinations interoperate:

The following G.723.1 codec combinations interoperate:

Delay

Delay is the time it takes for packets to travel between two endpoints. In traditional data networking, delay can be tolerated with little or no impact on network users; however, in networks carrying voice traffic, delay is potentially quite significant because it can affect the ability of users to carry on a telephone conversation. For example, delay can introduce pauses or gaps in the conversation, increasing the likelihood that one person will start talking before the other person has finished.

Because of the speed of network links and the limited processing power of many devices, some delay is expected. Telephone users normally accept up to about 150 milliseconds (ms) of delay without noticing problems. You can measure delay by using ping tests at various times of the day with different network traffic loads. If network delay is excessive, reduce it before deploying a network that carries Voice over IP (VoIP) traffic.

The two types of delay most commonly found in today's telephony networks are propagation delay and handling delay. Propagation delay is caused by the characteristics of the speed of light traveling via a fiber-optic-based or copper-based medium. Handling delay (sometimes called serialization delay) is caused by the devices that handle voice information. Handling delays have a significant impact on voice quality in a packetized network. Codec-induced delays are considered a handling delay.

Table 6-4 shows the delay that is introduced by different codecs.

DSP Groups

ASIs and MRPs handle calls based on the grouping of the DSPs. The DSPs are located on PVDMs. There can be up to five DSPs on a single PVDM. Each PVDM corresponds to one DSP group. MRP200 and MRP300 cards each have two PVDM slots and, therefore, can have a maximum of two DSP groups. Each DSP group serves either an analog port or a T1 port on the VIC. Therefore, one analog VIC and one T1 VWIC make up two groups, and two T1s with two different clock sources (regardless of whether they are on the same VWIC) also make up two groups.

DSP Group Serving a T1 Port

Each DSP group that serves a T1 port can support as many DSPs as there are in the PVDM.

A DSP has a maximum capacity of 100 MIPS to handle a particular number of simultaneous calls. One G.729a call requires 25 MIPS, and one G.711 call requires 12.5 MIPS. The number of calls on a DSP is determined by the total MIPS used reaching 100 on that DSP. The DSP resource manager rejects a call if it cannot find a DSP with required unused MIPS for the selected codec.

Table 6-5 provides some examples of the number of calls that can be supported on a single DSP, depending on the codec used. Table 6-6 lists some of the combinations of calls that can be handled on a single DSP.

Note   The examples provided in Table 6-5 and Table 6-6 are based on the assumption that you are using a medium-complexity digital image.

DSP Group Serving Analog Ports

Each DSP group that serves analog ports requires the following:

Choosing Codecs

This section provides information that can help you choose the DSP image that is best suited for a particular type of traffic. The following are some common scenarios:

Choosing DSP Firmware

When you choose DSP firmware, it is important to consider the following factors:

DSP firmware is included with each IOS release for the Cisco ICS 7750. Five DSP firmware images are available for use on ASI and MRP cards. Two of the DSP firmware images are intended for MRP200 and MRP300 cards (which contain analog VICs) and for MRP3-8FXS and ASI81 cards (which contain FXS ports); two images are intended for digital trunks (such as T1 CAS and T1/E1 PRI); and one image is intended for transcoding.

Each DSP firmware image supports a particular set of codecs. High-complexity DSP firmware supports more codecs than medium-complexity firmware supports. However, in order to support more codecs, the number of voice channels supported by the firmware has to be reduced.

Table 6-7 lists the number of channels supported by the DSP firmware images.

Note   The abbreviations FIXHC, FIXMC, FLEX6, FLEX8, and XCODE represent the names of the firmware images. These abbreviations appear in the output from the show voice dsp command.

Determining How Many DSPs Are Needed

The number of DSPs needed for each voice interface depends on the following two factors:

Table 6-8 shows how to calculate the number of DSPs needed for each channel. For example, with a medium-complexity analog image and a G.726 codec, 1 DSP is needed for 4 voice interfaces.

Note   For additional information on PVDM selection, refer to the "PVDM Requirements" appendix in the Cisco ICS 7750 Hardware Installation Guide .

Note   Table 6-8 does not address configuration rules for transcoding. See the "Determining How Many DSPs Are Needed for Transcoding" section.

Note   On MFT-T1 and MFT-E1 cards, medium-complexity firmware is not recommended, because this firmware restricts echo cancellation coverage to 16 ms.

Note   The codec complexity for ASI cards defaults to medium-complexity but can be changed to high-complexity with sufficient DSPs (see Table 6-8).

Transcoding

Because some hardware and software currently support only G.711 (uncompressed) connections, transcoding is available on MRP and ASI cards. MRP and ASI cards are considered packet-to-packet gateways because they have DSPs that transcode between voice streams using different compression algorithms. For example, when a user on a Cisco IP Phone at a remote location calls a user at the central location, Cisco CallManager can be configured so that it causes the remote IP phone to use compressed voice (G.729a) for the WAN call. However, if the called party at the central site is unavailable, the call potentially could be routed to an application that supports only G.711. In this case, the MRP or ASI card transcodes the G.729a voice stream to G.711 so that a voice message is stored by the G.711-compliant voice-messaging server.

Transcoding is required when a compressed voice stream is used to save WAN bandwidth and when the local device does not support the codec. The transcoding service compresses and decompresses voice streams to match the capabilities of the endpoint device.

A transcoder is a device that takes the output stream of one codec and transcodes (converts) it from one compression type to another compression type. For example, a transcoder could take an output stream from a G.711 codec and transcode (convert) it in real time to a G.729 input stream accepted by a G.729 codec.

Transcoding is supported under the following conditions:

Deciding When to Use Transcoding

Transcoding is needed when the calling and called parties cannot use the same codec type. Codec incompatibility may result from of a lack of support for a particular codec. For example, some unified messaging systems support only G.711, while Cisco IP Phones support G.711 and G.729. (Note that Cisco Unity supports both G.729a and G.711.) Codec incompatibility could also be caused by a failure when negotiating a common codec. For example, in a lab, two Cisco voice gateways (such as an MRP or an ASI card) can be forced to use different codecs so that transcoding is required for them to communicate.

Suppose that an application is communicating with a G.711-only voice-mail system over a WAN link. To conserve bandwidth, the caller on one side of WAN link uses G.729, while the called party voice-mail system recognizes only G.711. This is a situation that would require transcoding.

Transcoding is not required if all the called parties (except those on a voice-mail system) are on the same LAN. You can configure the calling and called parties so that they must negotiate a common codec when possible.

Here are some additional transcoding guidelines:

Choosing a DSP Firmware Image for Transcoding

When a DSP is reserved for transcoding, a special DSP firmware image is downloaded to the DSP. At present, the DSP firmware supports transcoding between G.723.1/G.729 and G.711 U/a-law, as well as between G.711 U-law and G.711 a-law. Transcoding between low-bit-rate codecs, such as between G.723.1 and G.729, is not supported.

Determining How Many DSPs Are Needed for Transcoding

Before an MRP or ASI can act as a transcoder, DSP resources must be reserved for transcoding. Unlike other Cisco gateways, the MRP or ASI provides the flexibility to choose the number of channels that should be reserved for transcoding. One DSP is required for every two transcoding channels (full duplex).

Understanding How DSPs Are Allocated for Transcoding

When the MRP or ASI boots, DSP resources are statically allocated first for analog VICs and the VIC-2BRI-NT/TE. These DSP resource allocations cannot be changed. In the show voice dsp command output, these DSPs are represented with a value of FIXMC or FIXHC in the Image field, depending on whether high- or medium-complexity DSP firmware is being used. The remaining DSP resources can be allocated to T1 VWICs, to E1 VWICs, or to transcoding, as needed.

For T1 VWICs or E1 VWICs, DSPs are reserved by defining a ds0-group or pri-group under the individual T1 or E1 controller. A DSP is reserved if it hosts a signaling channel for the T1/E1 VWIC. Such a reserved DSP has a non-zero value in the D-sig Allocate field, which can be seen in the show voice dsp command output.

Configuring Fast Ethernet Ports

ASI and MRP cards have Fast Ethernet interfaces that can be configured.Depending on your own requirements and the protocols you plan to route, you might need to enter additional configuration commands. For more information about basic configuration, including enabling the interface and specifying IP routing on Fast Ethernet interfaces, see the section "Configuring Ethernet, Fast Ethernet, or Gigabit Ethernet Interfaces" in the Cisco IOS Interface Configuration Guide, Release 12.2.

Note   See the "Configuring Dial Plans" section for a sample configuration to configure a FastEthernet interface on an ASI or MRP card.

Note   Use ICSConfig to assign or modify the IP address of an ASI or MRP card, as necessary. Do not use the CLI.

Configuring WAN Interfaces

You can configure an MRP or ASI card with a WIC or VWIC installed for access to the WAN. For example, if you are using a serial interface, you can configure Frame Relay, Point-to-Point Protocol (PPP), and High-Level Data Link Control (HDLC) over that serial interface.

Information about the various types of connections is provided in the sections that follow:

Table 6-9 lists tasks you might need to perform in order to configure WAN interfaces on MRP or ASI cards and gives pointers to the location in Cisco IOS documentation set that provides additional instructions on performing those tasks. The various Cisco IOS configuration guides for version 12.2 are available at the following URL:

http://www.cisco.com/univercd/cc/td/doc/product/software/ios122/122cgcr/index.htm

Configuring Asynchronous/Synchronous Serial WICs

You can configure the serial interfaces on your asynchronous/synchronous serial WIC (WIC-1T, WIC-2T, or WIC-2A/S) by entering IOS commands at the ASI or MRP command prompt, in configuration mode.

Note   See the "Configuring Synchronous Serial WICs" section for a sample configuration.

Table 6-10 lists the half-duplex timer commands.

Table 6-11 through Table 6-13 list clock rate settings in bits per second (bps) for specific interfaces.

Configuring ISDN BRI WICs

You can use an Integrated Services Digital Network (ISDN) Basic Rate Interface (BRI) WIC to connect MRPs or ASIs with other ISDN routers. ISDN BRI is a dial-up connection. Adding an ISDN BRI connection to the MRP creates a logical dialer interface.

ISDN connections use one or both data channels for the connection to the ISDN service provider. Normally, the ISDN provider is your local telephone company.

This section tells how to configure ISDN BRI WICs.

Note   For information on how to configure ISDN voice interfaces, see the "Configuring ISDN Interfaces for Voice" section.

ISDN BRI WIC Prerequisite Tasks

Before using an MRP with an ISDN BRI WIC, you must order a correctly configured ISDN BRI line from your local telecommunications service provider.

The ordering process varies from provider to provider and from country to country; however, here are some general guidelines:

Table 6-14 lists the ISDN switch types for North America.

ISDN BRI Provisioning by Switch Type

ISDN BRI provisioning refers to the types of services provided by the ISDN BRI line. Although provisioning is performed by your ISDN BRI service provider, you must tell the provider what you want. Table 6-15 lists the provisioning that you should order for switches used in North America.

Defining ISDN SPIDs

Some service providers use service profile identifiers (SPIDs) to define the services subscribed to by the ISDN device that is accessing the ISDN service provider. The service provider assigns the ISDN device one or more SPIDs when you first subscribe to the service. If you are using a service provider that requires SPIDs, your ISDN device cannot place or receive calls until it sends a valid, assigned SPID to the service provider when accessing the switch to initialize the connection.

At present, only the DMS-100 and NI switch types require SPIDs. The AT&T 5ESS switch type may support a SPID, but we recommend that you set up that ISDN service without SPIDs. In addition, SPIDs have significance only at the local access ISDN interface. Remote routers never receive the SPID.

A SPID is usually a seven-digit telephone number with some optional numbers. However, service providers may use different numbering schemes. For the DMS-100 switch type, two SPIDs are assigned, one for each B channel.

To define SPIDs and the local directory number (LDN) for both ISDN BRI B channels, use the following isdn spid commands in interface configuration mode:

Note   Although the LDN is an optional parameter, you might need to enter it so that the MRP or ASI can answer calls made to the second directory number.

For further information on configuring ISDN, refer to the "Configuring ISDN BRI" chapter in the
Cisco IOS Dial Technologies Configuration Guide .

Configuring T1 and Fractional T1 WICs

The 1-port T1 WIC (WIC-1T) and fractional T1 WIC (WIC-1DSU-T1) include an integrated data service unit /channel service unit (DSU/CSU) and can be configured either for full T1 service (1.544 Mbps) or for fractional T1 service (less than 1.544 Mbps). You can configure the interfaces on your T1 WICs by entering IOS commands at the ASI or MRP command prompt, in configuration mode.

The IOS software provides a default configuration for CSU/DSU- and T1-specific parameters. To view the current configuration, enter the show service-module serial slot/port command.

Note   See the "Configuring T1 and Fractional T1 WICs" section to see the default configuration and a sample configuration to configure a new T1 or fractional T1 interface or to change the configuration of an existing interface.

For further information about these commands, refer to the "Configuring Serial Interfaces for CSU/DSU Service Modules" section in the "Configuring Serial Interfaces" chapter in the Cisco IOS Interface Configuration Guide .

Configuring VWICs for Data-Only Transmission

You can configure the multiflex trunk (MFT) interface card as a WIC (for data-only transmission). In the WIC mode, an MRP treats the T1 or E1 as a single serial interface for data. You can specify the number of channels (up to 24 [T1] or up to 30 [E1]) for this connection. On a data T1 or E1, you can configure only one channelized group. The rest of the channels are not used.

In a data-only configuration, an MRP supports the following T1 or E1 configurations:

This section describes basic configuration, including enabling the interface and specifying IP routing. Depending on your own requirements and on the protocols you plan to route, you might need to enter other configuration commands as well.

Note   See the "Configuring VWIC Interfaces for Data" section for a sample configuration to configure a new T1 or E1 VWIC interface or to change the configuration of an existing