Cisco CCNA Wan Switching

ATM Basics

ATM is a packet-switched technology based on a 53-byte packet called a cell. Each cell is divided into a 5-byte header and a 48-byte payload. The short, fixed length cell reduces delay and jitter, allowing time sensitive information such as voice and video to be transported along with data. There are various transmission media and rates available with bandwidth measured in megabits to gigabits.

ATM Model

ATM is based on Broadband Integrated Services Digital Network, an extension of ISDN. Similar to the OSI model, B-ISDN uses a seven-layer model. ATM redefines the lower 3 layers into the Physical Layer, the ATM Layer and the ATM Adaptation Layer.

• The Physical Medium sub-layer (PMD) interfaces with the physical medium.

• The Transmission Control sub-layer handles cell extraction from the data stream and error checking.

ATM Headers

The ATM Cell has a 5-byte header, with the remaining 48-bytes left for payload (data).

There are two standard headers. Cisco added an additional header type to allow for advanced ATM features.

• User-to-Network Interface (UNI) header-specifies the interface between a user device and a network. Note: A user device is not just a computer interface. It can be a router or switch as well.

• Network-to-Network Interface (NNI) header-specifies interface between two networks. Usually a private ATM network and a public ATM network.

• STI header-used between two Cisco ATM switching nodes to allow advanced network features.

• Flow Control (4-bits)-a UNI field for controlling access and flow control. Usually all zeros, as there is no defined standards.

• Virtual Path Indicator (VPI)-identifies the path to be taken by the ATM cell.

• Virtual Circuit Indicator (VCI)-indicates the circuit number on the path.

• Payload Type Indicator (PTI, 3-bits)-the type of data being carried in the payload. High order bit is 0 for user data and 1 for connection management information, second bit indicates if there was congestion, and the third bit shows if user data is from customer premises equipment.

• Payload Class (4-bits)-STI field indicating classes of service and BPX switch queues.

• Cell Loss Priority bit (CL)-indicates that the cell may be discarded if congestion is encountered.

ATM Addressing

• VCI identifies the circuit or connection.

• VPI identifies the virtual path.

• The path can be seen as a trunk that carries multiple circuits between switches.

• VCI is 16-bits allowing 65, 535 circuit numbers. For the STI header VCI is 8-bits for 256 circuits.

• VPI is 8-bits for the UNI header, 12-bits for the NNI header (4096 paths), and 10-bits for the STI header (1024 paths).

Cisco ATM Addressing

• There are three modes available depending on the hardware being used.

ATM Adaptation Layer

• ATM Adaptation Level (AAL)-used to convert data from various sources and convert it to 48-byte segments that fit the payload of ATM cells.

• Four classifications of source traffic are outlined.

• Consists of Convergence sub-layer (CS) and Segmentation and Reassembly sub-layer (SAR).

• CS receives data from the applications and sends it to the SAR to be segmented into 48-byte blocks.

Quality of Service

These are some of the most important QoS parameters:

• Peak Cell Rate (PCR)-maximum rate a sender can send cells.

• Sustained Cell Rate (SCR)-required cell rate averaged over a long interval.

• Minimum Cell Rate (MCR)-minimum acceptable rate of cells per second.

• Cell Loss Ratio (CLR)-the fraction of cells not delivered or delivered late.

• Cell Transfer Delay (CTD)-time from source to destination (transmit time).

• Cell Error Ratio (CER)-fraction of cells delivered with incorrect bits.

• Cell Delay Variation (CDV or Jitter)-how regularly cells are delivered. Cells from one conversation on a multiplexed connection may be delayed while cells from another conversation are delivered.

• Cell Delay Variation Tolerance (CDVT)-amount of variation in cell transmission times. Specified separately for PCR and SCR.

Service Categories

The ATM Forum specifies four service categories:

• Constant Bit Rate (CBR)-entire connection, from source to destination, including intermediary switches, provides a set amount of bandwidth at all times. Expensive because bandwidth is reserved even if it is not used.

• Variable Bit Rate (VBR)- guarantees an average bit rate over time, but allows a higher peak bit rate, with no cells lost, for a certain amount of time each minute. VBR is broken into two subclasses real time (rt) and non-real time (nrt).
  

  1. rt-VBR is used for connections that require end-to-end timing, such as voice or video connections using compression or noise reduction.

  2. nrt-VBR can be used when timing relationships are not required, but QoS is still needed.
     

• Available Bit Rate (ABR)-bit rate varies according to network conditions up to a stated maximum. Source adjusts transmission rate due to network information received from the network.

• Unspecified Bit Rate (UBR)-no set bit rate guarantee. Allows a transmission rate up to a certain maximum, with no guarantee of cell loss or delivery.

Frame Relay Basics

Frame relay

• Defines a connection between user equipment and a WAN, not the interface or protocols of the WAN itself. Usually describes a LAN-to-LAN connection.

• A standard optimized for the transport of protocol-oriented data.

• Defines network paths using statistical multiplexing (creates virtual circuits).

• Does not allocate bandwidth until data needs to be transmitted.

• Combines packet switching and port sharing with time division multiplexing circuit switching to allow multiple point-to-point permanent virtual circuits over a single physical interface.

• Operates only on the Data Link and Physical layers of the OSI model. Routing is general handled by the Network layer protocol.

Components of Frame Relay

• Committed Information Rate (CIR)-rate of data transfer under normal operations. Generally the contracted rate from the public service provider. CIR should never be set higher than the speed of the slowest physical connection on a VC. CIR of 0=best effort.

• User-Network Interface (UNI)-describes a connection between user equipment and a frame relay network. Usually, a router (DTE) and the service provider.

• Network-Network Interface-describes connection between frame relay networks.

• Local Management Interface (LMI)-monitors the status of DLCIs by periodically polling the network. Can be used to exchange status information between frame devices and the network. Also supports multicasting, global addressing and flow control. Cisco switches support LMI.

• Bandwidth on Demand-capacity is provided as a PVC requires it up to the maximum CIR.

• Flag-first and last byte of a frame, indicating the beginning and the end. There may be only one flag between frames with the end flag of one frame used as the start flag of the next.

• Data Link Connection Identifier (DLCI)-10-bit routing address of the PVC at a particular UNI or NNI. Some DLCIs are reserve for signaling, management and future use so only 16 to 1007 are used to address virtual circuits.

• Command/Response (C/R)-not used. Always set to 0.

• Extended Address (EA)-allows the header to be lengthened to 3 or 4 bytes enabling a DLCI longer than 10-bits (more VC addresses).

• Forward Explicit Congestion Notification (FECN)-set to 1 by the frame network when congestion is occurring on the packet forwarding direction of the frame.

• Backward Explicit Congestion Notification (BECN)- set to 1 by the frame network when congestion is occurring on the reversed packet forwarding direction of the frame.

• Discard Eligibility (DE)-set by the end node to indicate frames to be discarded if congestion occurs.

• Data-also called the information field. May be as large as 4,520 bytes, but is more efficient with a size of less than 4K.

• Frame Check Sequence (FCS)-standard cyclic redundancy check that detects bit errors in the frame.

General Installation

Tools Required

• Box knife

• Standard blade screwdriver set

• Phillips screwdriver set

• Wrench set, including adjustable wrench and/or channel locks

• Allen key set

• Grounding wrist strap

Installation Outline

Preliminary Steps

1. Prepare the site for equipment installation
   

   1. Location should have restricted access.

   2. Sufficient area should be available to access the equipment.

   3. The proper power requirements need to be met.

   4. Confirm that the UPS meets Cisco requirements

   5. Airflow and cooling must be suitable for the equipment being installed.
      

2. Verify that all parts ordered are present and in good condition. Record all part numbers and serial numbers.

3. Unpack and install the equipment in the desired rack (may be provided by Cisco or third party). Follow the instructions in the supplied install guide.

4. Install and connect the required power supply. Most models are available in either AC or DC versions.

5. Install the cards in the equipment. Most cards consist of a front and back card pair. If cards are not matched, alarm conditions will occur. All Cisco WAN switches have specific slots reserved for certain cards. These reservations must be followed. Proper grounding (wrist strap) must be followed prior to handling any cards.

6. Install the cable management system and connect the cables to the required ports.

7. Connect a management terminal or Cisco WAN Manager workstation to the console port.

Installation troubleshooting

If there is no power to the switch

• Check the power cables.

• Check the circuit breaker.

• Check for card short circuit (pull and reseat all cards).

• Check power supply fault.

No LED lit on a single card

• Remove and reseat card.

• Check card fuse if applicable.

• Replace card. If replacement does not fix problem, check the backplane.

Alarm Indicator on Card

• Get card status with dspcds.

• Reset card with resetcd f.

• Remove and reseat card.

Command Line Interface (CLI)

Connect to an IGX or BPX with a PC using a terminal emulator (hyperterm), or a terminal connected via the control port, a LAN connection, or remotely using a modem connected to the auxiliary port.

• Once connected, hit enter to bring up the command screen.

• Enter user name at the logon prompt

• Enter password at the password prompt.

• The bye command ends the session.

Use vt <nodename> to create a virtual terminal session on a remote node .

• The bye command returns user to local session.

Command Screen Format

• Top line displays node name, current user, software revision, date, time and time zone.

• The middle of the screen shows information returned by executed commands.

• The bottom of the screen displays prompt for current command or next command.

Basic commands

• . (period) -displays the last twelve commands. Enter a command number to re-use a command.

• help or ? –displays a help menu. If typed with a command brings up help for that command.

• bye –ends local or remote session.

• clrscrn –clears the display.

• prtscrn –sends current display to printer.

• redscrn –redraws display (useful if connected via modem).

• vt –used to start virtual terminal session with remote nodes.

Numbering Conventions

Trunks, Lines and Channel Numbers

Viewing Node Configuration

Display commands are also useful for troubleshooting.

• Display cards in shelf-dspcds.

• Display status, serial number and revision of a card-dspcd.

• Display redundant cards (y-pairs)-dspcdred.

• Display LAN port configuration-dsplancnf.

• Display power supply status and cabinet temperature-dsppwr.

• Display control port and auxiliary port configuration-dsptermcnf.

• Display function of control port and auxiliary port-dsptermfunc.

Privilege Levels

• There is one super user privilege level (level 0), and six user levels (levels 1 through 6).

• Each privilege level has access to commands of the lower levels (level 4 can access commands for levels 4-6).

Managing Users

• Add users use the adduser command. (You can only create a user for a lower privilege level than you are logged in as.)

• Change password use cnfpwd.

• Remove users with deluser.

Management Workstation

On the Switch

• Set up the LAN port with cnflan.

• Configure the SNMP get and set (read and write) community names with cnfsnmp.

• Configure the IP for the statistics manager, if you are using Cisco WAN Manager Statistics Collection Manager, using cnfstatmast (cnfstatmgr on the MGX switches).

Alarms

• Use dspnw to show a list of network nodes and trunk alarm status.

• Use dspnds to list name, type and alarm status of nodes on the network.

• dspalms displays a summary of all alarms on the current node.

• dsplogs to view log of errors.

Redundancy

• Set up card redundancy-addyred (or addcdred).

• Switch between active and redundant cards-resetcd.

• Switch between active and standby processor-switchcc.

BPX 8600 Series

The BPX 8600 series switches are high capacity, standards based broadband ATM switches that support:

• Backbone ATM

• IP+ATM services

• Multi Protocol Label Switching (MPLS)

Card Types

• Four types of BPX cards are available
  

  1. Common Core

  2. Network Interface

  3. Service Interface

  4. Power Supplies
     

• It is important to match the proper front card with the proper back card.

Installing BPX Cards

• The card shelf has 15 slots numbered from left to right when viewed from the front.

• When installing cards, follow ESD precautions.

• Slot 7 is reserved for BCC card. Redundant nodes have a BCC card in slots 7 and 8.

• Slot 15 is for the ASM/LM-ASM card pair.

• Slots 1-6 and 10-15 are for all interface cards.

19.2 Gbps Operation

The BPX switches support a 9.6 or a 19.2 Gbps backplane. For 19.2 Gbps support, you need:

• 19.2 Gbps backplane identified by white slot fuses on bottom rear of backplane (or result of “Word #2=0001” from dspbpnv command).

• Backplane NOVRAM that indicates the backplane is 19.2 Gbps capable

• BCC-4 or later controller.

• Switch software release 8.1.18 or later.

• At least one BXM card.
  

Initial Configuration

1. Set up the node
   

   1. Set node name with cnfname command.

   2. Time zone-cnftmzn.

   3. Date-cnfdate.

   4. Time-cnftime.

   5. LAN interface-cnflan.

   6. Configure console or aux ports to support any external devices using cnfprt (configure printing), cnfterm (configure tx parameters), and conftermfunc (configure port functions).
      

2. Set up the trunks (use dspcds to ensure proper cards are installed)
   

   1. Up trunk at each node-uptrk.

   2. Configure trunk parameters-cnftrk.

   3. Add trunks at each node-addtrk.

   4. Configure redundancy (if require)-addyred.
      

3. Configure IGX interface shelf (if present).
   

   1. Set up trunk between BPX and IGX as above.

   2. Contact Cisco customer service to add options.

   3. Add IGX as shelf at BPX-addshelf.
      

4. Add MGX 8220 Shelf (if present).
   

   1. Add MGX as a shelf at BPX-addshelf.
      

5. Set up ATM services.
   

   1. Activate line-upln.

   2. Configure line-cnfln.

   3. Activate ports-upport.

   4. Configure ports-cnfport.
      

6. Configure ATM connections.
   

   1. Add connection-addcon.

   2. Configure connection type-cnfcontyp.
      

7. Set up ATM to Frame Relay (ATF).
   

   1. Add connection-addcon.

   2. Specify connection class-cnfcls.

   3. Add connection group-addcongrp.

   4. Configure Interface Shelf to Frame Relay Connections.
      

8. Connections are added using Cisco WAN manager.
   

   1. Frame Relay connections from IGX to BPX.

   2. ATM or Frame Relay connections from MGX to BPX.

   3. Connections are end-to-end and managed via SNMP.

IGX 8400 Series

• Multi-service switch supports WAN voice, data and video.

• 1 Gbps cellbus with a .2 Gbps backup bus

• Six models available:
  

  1. IGX 8410-8 slot rack mount

  2. IGX 8410-8 slot stand-alone

  3. IGX 8420-16 slot rack mount

  4. IGX 8420-16 slot stand-alone

  5. IGX 8430-32 slot, dual shelf rack mount

  6. IGX 8430-32 slot, dual shelf stand-alone

Card Installation

• IGX 8410 (8 Slot configurations).

• IGX 8420 (16 Slot configurations).

• IGX 8430 (32 Slot, dual shelf configurations).

1. Remove jumper W6 on the system clock module (SCM) card(s) to indicate an IGX 8430. The jumper is left on for 8420s.

2. A maximum of 16 trunk cards can be used on the 8430.

Non-redundant configurations always have the NPM card in slot 1. Redundant configurations have NPMs in slots 1 and 2. The SCM card always goes behind the NPM in slot 1.

The ARM card can go in any slot, but is usually installed in the far right.

Table of IGX Cards