|
|
This chapter describes the ATM routing and Private Network-Network Interface (PNNI) protocol implementation within the LightStream 1010 ATM switch. In order to place calls between ATM End Systems, signaling needs to consult an ATM routing protocol. LightStream 1010 provides the following routing protocols:
The following major sections describe PNNI implementation:
The following major sections provide procedures for ATM routing and PNNI configuration:
For a complete description of the commands mentioned in this chapter, refer to the LightStream 1010 ATM Switch Command Reference publication.
PNNI provides dynamic ATM routing with QOS support. The PNNI protocol is used as "the routing protocol" for the global ATM internetwork; and therefore has been specified as a hierarchical routing protocol. The number of hierarchical levels can vary from network to network.
Note The initial LightStream 1010 implementation of PNNI will support only a single level of hierarchy.
Figure 11-1 shows a network using PNNI and IISP protocols and is used in the following sections to describe PNNI and IISP routing and PNNI components. IISP routing is used to interconnect peer groups at the corporate campus backbone level, but the addresses have been assigned to allow future migration to hierarchical PNNI routing. Specifically, the corporate campus will eventually form a PNNI peer group at level 80.
A Peer Group (PG) is a collection of logical nodes that exchange information with other members of the group. This allows all members of the same peer group to maintain an identical view of the group. For example, Hello Packets, database synchronization, and flooding are carried out among members of the same peer group. The Engineering, Marketing, and Administration buildings are separate peer groups in Figure 11-1. For example, SW2.01 and SW2.02 are both members of the peer group 2.
Note Peer groups are similar to Open Shortest Path First (OSPF) routing protocol areas or to IP routing domains.
SW2.01 is a border node between peer group 2 and peer group 3. The border nodes must translate between PNNI and IISP and vice versa in order to splice the connections. The border nodes are also configured with reachability information to end systems (ESs) outside the PNNI network. The configured ES addresses are then advertised within the private network so that any internal node desiring to connect to the advertised ES routes to the border node. The signaling messages for IISP and PNNI are both supported at the border nodes.
The level of a peer group, including all nodes within the peer group, indicates its position in the PNNI hierarchy. A level with a small numerical value implies greater topological aggregation and has a higher level in the PNNI hierarchy. A level with a larger numerical value implies less topological aggregation and has a lower level in the PNNI hierarchy.
The default assignment of nodes to peer groups is determined by their addresses. For example: the default lowest level peer group ID is determined by the level bit prefix of the node address where the level is configurable.
All lowest level nodes in a peer group should share the same level-bit address prefix and should have unique 13-byte address prefixes. See the section "Configure the ATM Address" for the command line interface (CLI) command used to configure the ATM address.
The VC routing and call-processing system architecture are shown in Figure 11-2.
This is a platform-independent architecture, and is valid for both campus and enterprise switches.
When a connection request is received at a UNI or IISP interface, the Designated Transit List (DTL) or source route needs to be computed. Signaling requests the source route by sending a message to the PNNI router.
Static routing protocols require route configuration. IISP is an example of a static routing protocol. Static routing protocols are unsuitable for larger networks because they require a significant amount of configuration.
Dynamic routing protocols adapt to changing network conditions by advertising reachability and topology status information changes. PNNI is a dynamic routing protocol. The LightStream 1010 PNNI implementation can interoperate with IISP to provide routing between multiple peer groups.
Quality of Service (QOS) routing is the selection of routes or paths that satisfy a set of constraints for a requested connection. The user can specify QOS parameters for such connections as:
Routing must find a path that satisfies the requested QOS for the duration of the connection.
There are basically two techniques to thread a path in a network.
Hop-by-hop routing is well known in datagram switching. Hop-by-hop routing is based on the computation of a table that has one entry for each destination node. The next entry in the table specifies what the next node on the path is to that destination. Routing towards the destination is done by forwarding the datagram (call request) to the next node based on the destination node address and the information in the next-hop routing table. Hop-by-hop routing requires the presence of the address of the destination node in the datagram.
When using PNNI the next-hop routes used by IISP are incorporated into the PNNI routing tables.
Source routing has the entire path specified by the source node. The path information is included in the call setup message, and signaling follows the path accordingly.
The PNNI protocol is a source routing protocol and has the following advantages over a hop-by-hop routing protocol:
A source route in PNNI consists of Designated Transit List (DTL) stacks, each including a sequence of nodes and links at a level of the PNNI hierarchy.
Note The initial LightStream 1010 ATM switch implementation of PNNI supports only one level of hierarchy so each source route consists of only one DTL.
Hop-by-hop routing controls loops by ensuring that all switches see a consistent topology view when computing next hop routing tables. The next hop routing tables have to be computed based on the same metric. In every distributed system there are transient conditions, such as a link failure, when the next-hop routing tables are inconsistent. This leads to looping. This is not a problem in datagram routing because the datagrams contain a mechanism that does not allow indefinite packet looping in the network.
Source routing is by definition loop free. Using source routing allows every switch to compute the source routes independently using its own selected mechanism. Transient conditions are not crucial to PNNI. In the worst case the call is cranked back to the source and an alternate path selected, but no looping ever occurs. See the section "Crankback Mechanism."
Source routing provides the best control over the path selected in the network. The path never diverges from what the source route specifies, regardless of the route selection criteria. PNNI uses the crankback mechanism to correct failures by systematically routing calls to the destination. See the section "Crankback Mechanism."
PNNI is a Topology State algorithm. It advertises information about the status of links and nodes in the network. The advertised information contains multiple metrics and attributes for links and nodes for each ATM service category. Both topological information and address reachability are advertised. PNNI supports a hierarchical organization of the topology database.
The Hello protocol is used to discover the identity of the adjacent neighbor node. The PNNI Hello protocol was modeled on the Open Shortest Path First (OSPF) protocol with appropriate extensions to support a hierarchical organization of the topological database. Discovering the identity of the neighbor is done via an exchange of hello packets containing appropriate information. If the switches discover they are members of the same peer group, they form an inside link. If they are members of different peer groups, additional information about the hierarchy is exchanged and an outside link is created.
When the Hello protocol has declared the link as functional, the adjacent switches exchange a summary of their database contents. This mechanism is similar to the OSPF database synchronization procedures. The synchronization is governed by a master and slave relationship of switches. Nodes exchange database summary packets which contain header information of all PNNI Topology State Elements (PTSEs) in a node database. After such an exchange differences in the topological databases are updated. When completed, both nodes have consistent topological databases.
PNNI Topology State Packets (PTSP) containing one or more PTSEs are used to disseminate information in the ATM network. PTSPs contain reachability, link and node status information necessary for PNNI to calculate QOS paths in an ATM network.
PNNI allows summarization of multiple ATM addresses into a single summary address prefix. Address summarization and the hierarchical organization of the topology enables PNNI to scale to very large networks.
Reachability information is used as the first step in routing a PNNI signaling request for a virtual connection. The call set up packet will be directed to a node advertising a prefix which matches the leading portion of the destination address. The longest matching reachable address prefix is always used.
This information group describe internal reachable ATM destinations. Internal means known to PNNI to be local. For a node representing a single switch, an internally reachable address represents a summary of end systems attached to the switch, for example, discovered via ILMI address registration. At higher levels of the hierarchy, it summarizes information provided by members of the peer group.
Note Internal static routes can be configured manually to end systems that do not support ILMI. These routes will automatically be advertised as internal reachable addresses (subject to address summarization).
Exterior reachable ATM addresses are similar to internal reachable addresses and describe reachability to a set of ATM destinations. Using an exterior advertisement implies that the reachability information came from elsewhere. This includes cases such as information from other routing domains in which the switch participates or configuration about what is reachable over a specific link. A link connecting to an IISP network is an example of an exterior reachable address.
Being a topology state routing protocol, PNNI advertises detailed information about the status of the links and nodes. The status of the topological entities (links and nodes) is described via metrics and attributes. Metrics are combined along a path. The simplest example of a metric is the administrative weight. The administrative weight of a path is the sum of the weights of links and nodes along the path.
Attributes are treated by PNNI in a different way. If an attribute value for a parameter violates the QOS constraint, then PNNI excludes that topological entity from consideration while making a path selection.
Supported metrics and attributes include the following:
Metrics and attributes supported by PNNI are listed in Table 11-1.
| Parameter | Metric/ Attribute |
Static/ Dynamic |
Unit | Granularity | Range | Ecoding |
|---|---|---|---|---|---|---|
PNNI Metrics and Attributes Supported
PNNI metrics and attributes are specified separately for each parameter in the following service categories:
Administrative Weight (AW) is the main metric used for computation of paths by PNNI. The assignment of administrative weights to links and nodes will influence the way PNNI selects paths in the private ATM network.
Administrative weight indicates the relative preference of a link assigned by the private network owner. For example, it may depend on link capacity or link length.
Administrative weight can also be used to exclude certain links from routing, such as a backup link that needs to be used only when the primary link is full. The administrative weight for a path is simply the sum of the individual weights of the links on the path.
The LightStream 1010 will select paths with the least administrative weight when such paths satisfy the requested QOS of a connection.
Available Cell Rate (AvCR) is the most dynamic metric in PNNI. It reflects the amount of equivalent bandwidth that is available on the link for new connections. AvCR depends on the calls traversing the link and is viewed as the residual capacity left for use by additional calls. Not every change in AvCR will be advertised in the network by PNNI. Only significant changes as defined by the ATM Forum PNNI specification are advertised in the network. PNNI needs the knowledge of AvCR to decide whether a given link or node is suitable to carry a given call. The LightStream 1010 PNNI implementation supports both simple and complex Generic Call Admission Control (GCAC) to make this decision.
AvCR is maintained on a per service category basis. Three AvCRs are maintained, one each for CBR, RT-VBR, and NRT-VBR service categories.
Not every change of parameter value is substantial enough to generate an advertisement. The network would be overwhelmed with PNNI advertisement packets if frequently changing parameters were to generate advertisements every time any change in their value occurred. Changes in CDV, MaxCTD or AvCR are measured in terms of a proportional difference from the last value advertised. A proportional multiplier threshold expressed as a percentage provides flexible control over the definition of significant change.
Note For other parameters such as administrative weight, any change in value is considered as significant.
For each parameter, PNNI defines what constitutes a significant change. See the section "Configure Significant Change Thresholds" for configuration information.
Not all switches will have the same mechanism to perform call control for connection admission. PNNI has defined its own mechanisms to determine whether a call with requested Peak Cell Rate (PCR) and Sustainable Cell Rate (SCR) will be admitted on a selected link (node). These mechanisms screen links and nodes for consideration in path computation. Two mechanisms perform this screening function, depending on the number of parameters advertised by each entity:
Simple GCAC requires only AvCR to be advertised.
Complex GCAC provides a more accurate calculation at a price of increased processing complexity. It uses two additional parameters that can be optionally advertised by a PNNI entity. These are the following:
By default the LightStream 1010 PNNI uses simple GCAC. In addition, it operates using complex GCAC for those links and nodes that have advertised CRM and VF.
The ATM Forum PNNI specification defines GCAC for the following service categories:
The crankback concept adapted in PNNI is based on similar mechanisms used in the circuit switching. Crankback folds back the call to the source node in the peer group that created the DTL and the source retries on an alternate path. Crankback pinpoints the link or node to be avoided in the next retrial. A single call can be cranked back to the source many times. After a number of retrials the crankback mechanism declares that it cannot provide the requested QOS path.
Figure 11-3 is an example of a signaling request encountering insufficient available cell rate at the link SW3p3->SW5p1.
Signaling includes a crankback information element (IE) in the call release message indicating the blocked link SW3p3->SW5 and sends it back upstream to the source switch. PNNI computes a new DTL that avoids the blocked link: (SW1p3-> SW2p3-> SW4p2->SW5).
Note Crankback is transparent. It is an automatic mechanism that increases the success probability of a call.
The LightStream 1010 ATM switch PNNI implementation supports the following enhanced PNNI features:
The LightStream 1010 ATM switch PNNI implementation supports autoconfiguration. When the switch initially comes up, an autoconfigured ATM address is assigned. The autoconfigured ATM address provides a unique 13-byte address prefix to each switch used for ILMI address registration and address summarization. All autoconfigured addresses share the same 7-byte address prefix so they belong to the same peer group at level 56. This feature allows you to interconnect multiple switches out of the box without any configuration necessary.
You can modify the ATM address using manual configuration commands. In PNNI, by default the node identifier originates from the ATM address assigned to the switch. The node identifier uniquely identifies the node in PNNI. A change of ATM switch address would normally result in a change of the node identifier, causing all links to go down and thus disallowing the possibility of smooth address migration in the peer group. See the sections "Configure the ATM Address" and "Configure PNNI Node" for ATM address modification procedures.
The LightStream 1010 PNNI implementation allows you to tune to the network conditions using the Command Line Interface (CLI).
Load balancing distributes the traffic throughout the network. As a result of load balancing techniques, you can transmit more traffic across the network. The LightStream 1010 ATM switch uses two types of load balancing:
For configuration information see the section "Configure ATM PNNI Link Selection."
A typical application of VP tunneling using the workgroup and enterprise switch is shown in Figure 11-4. The switches form small private networks over a public cloud. They are interconnected over Permanent Virtual Paths (PVPs) which are essentially logical trunks.
The dashed lines indicate the PVPs which interconnect the ATM switches. Each of these VPs are logical trunks over which signaling (on channel VP=X, VC=5) and PNNI (on channel VP=X, VC=18) can operate normally. Switched Virtual Circuits (SVCs) can be routed and signaled across the VP tunnel as if both endpoints were connected by a physical link.
PNNI address scope allows you to constrain advertised reachability information within configurable bounds. This allows network administrators flexible control of reachability information, in case of network restructuring.
The service scope for a group of addresses is the highest-level peer group containing the group address for anycast purposes. Calling parties will not be allowed to establish connections outside of their highest peer group.
A source node will route to the node that has advertised the longest address prefix that matches the destination. However scope checking takes precedence over longest-prefix match routing.
Note If two destination nodes have the same address prefix, the node closest to the source will be selected.
Configure scope for path selection using the following rules:
Anycast connections use scope to make point-to-point connections to a group address. The anycast route to one or any of the group is determined by the longest prefix address.
Note Only members of the group that are reachable within the indicated connection scope are considered.
The UNI design of the membership scope control considers how it is used in the real network applications. Most applications would like to control reachability according to human organization hierarchies (for example, intra-network or intra-site).
Table 11-2 lists the UNI scope number and its corresponding hierarchy.
Table 11-2 UNI Scope Hierarchy
| Scope Number | Hierarchy |
|---|---|
The LightStream 1010 switch uses the UNI scope numbers to display the UNI scope value.
As autoconfigured, LightStream 1010 switches form a peer group at level 56 of the PNNI hierarchy. Suitable UNI scope-to-PNNI level mapping is provided automatically and is listed in Table 11-3, "UNI Default Scope, Automatic." The mapping table is recomputed if the peer group level is modified and the scope mode is configured as automatic.
Note No manual modification of the scope map is allowed if scope mode is configured as automatic. See the section "Configure ATM PNNI Scope Mode" to change scope mode from automatic to manual.
If you reconfigure an entry and then use the no scope mode command to disable scope mode, the entry value will revert to the ATM Forum defaults specified in Table 11-3.
Note When you change scope mode from automatic to manual, no subsequent changes in the scope mapping table entries will occur unless you manually modify the entries according to your needs. When changing scope mode from manual to automatic, the table will be modified automatically to suitable recomputed values.
Table 11-3 lists the PNNI routing level indicators and the corresponding UNI default scope values.
Table 11-3 Scope Mapping Table
| PNNI Routing Level Indicator | UNI Default Scope | Description | |
|---|---|---|---|
| Automatic | ATM Forum | ||
By changing the PNNI routing levels to values that are greater than the node level, the reachability information scope becomes smaller than the node scope and the reachability cannot be advertised. The automatic UNI default scope mapping used by the LightStream 1010 is based on the ATM Forum UNI default scope.
For example, in Figure 11-1, all of the corporate campus switches should be configured with a low organization scope of 1 and a high organization scope of 10, except the two edge switches SW1.01 in peer group 1 and the corresponding switch SW1.02 in peer group 4. The two edge switches should be configured with a low organization scope of 1 and a high organization scope of 15. This would allow them to advertise their network reachability globally, and would restrict the reachability traffic of the other switches.
The LightStream 1010 scope default values are set as follows:
The PNNI scope associated to the ATM address and prefix can be configured as automatic or manual. see the section "Configure ATM PNNI Scope Mode" for configuration steps. To manually configure the scope mode see the sections "Configure Static Routes" and "Scope of Addresses" in this chapter.
This section describes the following LightStream 1010 ATM switch PNNI global configuration:
During the initial startup, the LightStream 1010 generates an ATM address using the defaults described in the section "ATM Address Configuration" in the chapter "Initially Configuring the LightStream 1010 ATM Switch." The switch's ATM address uses a hierarchical addressing model similar to the Open System Interconnect (OSI) network service access point (NSAP) addresses. PNNI uses this hierarchy to construct ATM peer groups. ILMI uses the first 13-bytes of this address as the switch prefix that it registers with end systems.
To configure a new ATM address that replaces the previous ATM address and is used to generate a new PNNI node ID and peer group ID, see the section "Configure PNNI Node."
Multiple addresses can be configured for a single switch and this configuration can be used during ATM address migration. ILMI registers end systems with multiple prefixes during this period until an old address is removed. PNNI automatically summarizes all of the switch's prefixes in its reachable address advertisement.
To configure an additional ATM address manually, use the following command using the no form of this command to disable.
| Task | Command |
|---|---|
At the privileged EXEC prompt, enter configuration mode from the terminal. |
|
| 1This command is documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example adds the ATM address prefix 47.0091.8100.5670.000.0ca7.ce01 and the ellipses (...) add the default MAC address as the last six bytes.
Use the show atm addresses command to display the ATM address configuration.
To display the ATM address configuration, perform the following task in user EXEC mode:
This example displays the ATM address configuration using the show atm addresses command from user EXEC mode:
The ATM routing software can be restricted to operate in static mode. In this mode the cell routing is restricted to only the static configuration of ATM routes. This disables running of any dynamic ATM routing protocols, such as PNNI.
The ability to configure the routing mode obviates the need for two separate software images for PNNI and IISP as in Release 11.1. The default mode is dynamic, in which PNNI functionality is available on all interfaces.
The atm routing-mode command is different than deleting all PNNI nodes using the node command and affects ILMI auto-configuration. If the switch is configured as static routing mode for each interface, the switch ILMI variable atmfAtmLayerNniSigVersion is set to IISP. This causes ILMI auto-configuration on interfaces between two switches to determine an interface type of IISP or if the switch on the other side indicates NNI signaling protocol is not supported.
Note The atm routing-mode command differs from other CLI commands. It takes effect only after the next reload of the software. The switch continues to operate in the current mode, until the software is reloaded.
To configure the ATM routing mode to static, use the following command. Use the no form of this command to return the switch to dynamic routing mode (PNNI) operation.
| Task | Command |
|---|---|
At the privileged EXEC prompt, enter configuration mode from the terminal. |
|
Write the running configuration to the startup configuration. |
|
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example uses the atm routing-mode static command to restrict the switch operation to static routing mode:
If the switch is operating is static mode, the following example resets the switch operation back to PNNI:
Use the show running-config command to display the ATM routing mode configuration.
To display the ATM routing mode configuration, perform the following task in user EXEC mode:
This example displays the ATM routing mode configuration using the show running-config command from user EXEC mode:
Use the atm route command to configure a static route. A static route attached to an interface allows all ATM addresses matching the configured address prefix to be reached through that interface.
Note Two PNNI peer groups may be connected using the IISP protocol. This requires a static route to be configured on the IISP interfaces, allowing connections to be set up across the IISP link(s).
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
Following is an example of the atm route command configuring a static route to the 13-byte-switch-prefix, 47.00918100000000410B0A1081 to ATM interface 0/0/0.
Following is an example of the atm route command configuring a static route to the 13-byte-switch-prefix, 47.00918100000000410B0A1081 to ATM interface 0/0/0 configured with an E.164 address 1234567.
See the section "Configuring a Static Route with an E.164 Address" for detailed configuration information.
Following is an example of the atm route command configuring a static route to the 13-byte-switch-prefix, 47.00918100000000410B0A1081 to ATM interface 0/0/0 configured with a scope 1 associated.
Note The scope mode must be configured as manual using the scope mode manual command before ATM static routes may be configured with a scope setting. See the section "Configure ATM PNNI Scope Mode."
Use the show atm route command to display the ATM static route configuration.
To display the ATM static route configuration, perform the following task in user EXEC mode:
| Task | Command |
|---|---|
This example displays the ATM static route configuration using the show atm route command from user EXEC mode:
This example displays the ATM PNNI scope configuration using the show atm pnni scope command from user EXEC mode:
Use the atm route command to configure a static route to an E.164 address. A static route attached to an interface allows all ATM addresses matching the configured address prefix, to be reached through that interface and routed to an E.164 address.
Configuring a static route with an E.164 address is a two-step process and is described in this section:
Step 2 Configure the E.164 address to an interface
One E.164 address can be configured per ATM port. Signaling uses E.164 addresses in called and calling party information elements (IEs). Signaling uses NSAP address in the called and calling party subaddress IEs and are used if a network supports E.164 addressing.
E.164 addresses have the following properties:
Figure 11-5 is an example of an E.164 address switch static route configuration.
The network service access point (NSAP) address is used to initiate the call at the ingress to the public network. The public network routes the call based on the E.164 address. NSAP subaddress are carried through the public network in the subaddress field then the NSAP address is used at egress of the public network to complete the call.
The following commands are used to configure E.164 address static routes:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
Following is an example of the atm route command configuring a static route to the 13-byte-switch-prefix, 47.00918100000000410B0A1081 to ATM interface 0/0/0 configured with an E.164 address 1234567.
To complete the E.164 address static route configuration proceed to the section "Configure ATM E.164 Address on an Interface."
Use the show atm route command to display the E.164 address configuration.
To display the E.164 address configuration, perform the following task in user EXEC mode:
This example displays the E.164 address configuration using the show atm route command from user EXEC mode:
Use the atm e164 address command to configure an E.164 address to a specific interface.
One E.164 address can be configured per ATM port. Signaling uses E.164 addresses in called and calling party information elements (IEs). Signaling uses NSAP address in the called and calling party subaddress IEs if a network supports E.164 addressing.
This section describes configuration of an E.164 address on a per interface basis. Use the no form of the following commands to disable:
| Task | Command |
|---|---|
At the privileged EXEC prompt, enter configuration mode from the terminal. |
|
| 1This command is documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example configures the E.164 address, 7654321, on ATM interface 0/0/1:
Proceed to the following section to confirm the interface configuration.
To show the E.164 configuration, use the following EXEC mode command:
| Task | Command |
|---|---|
The following example displays the E.164 address configuration for ATM interface 0/0/1:
Use the show atm pnni command to display the ATM PNNI configuration.
To display the ATM PNNI router configuration, perform the following task in user EXEC mode:
This example displays the ATM PNNI global configuration variables available using the show atm pnni command from user EXEC mode:
The next example displays the ATM PNNI node configuration using the show atm pnni node command for node index number 1 from user EXEC mode:
This section describes the following LightStream 1010 ATM switch PNNI routing configuration:
To enter the ATM router PNNI configuration mode, perform the following task in global configuration mode:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example changes configuration mode to ATM router PNNI mode and displays the variables using the ? help command:
Most calls are routed using precomputed routing trees. To satisfy QOS requirements, multiple background trees, are precomputed. The LightStream 1010 ATM switch supports the following two route selection modes:
The background-routes-enable mode should be enabled in large networks, where it will most likely exhibit less stringent processing requirements and better scalability. Route computation is performed at most every poll-interval, when a significant change in the topology of the network is reported or when significant-threshold changes have occurred since the last route computation.
To configure the route selection mode to use background routes, perform the following task in global configuration mode, using the no form of the background-routes-enable command to return to on-demand mode:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
insignificant-threshold - Insignificant threshold value from 1 to 100.
poll-interval - Background routes poll interval value from 1 to 60 seconds.
See the section "PNNI Topology Description and Distribution" for more information describing ATM PNNI database updates.
The following example enables background-routes and configures the background routes poll interval to 30 seconds:
Use the show atm pnni background status and show atm pnni background-route commands to display the background route configuration.
To display the background route configuration, perform the following tasks in user EXEC mode:
| Task | Command |
|---|---|
The following example displays the ATM PNNI background route configuration:
The following example displays the ATM PNNI background route tables for CBR:
The LightStream 1010 ATM switch route selection algorithm chooses routes to particular destinations using the longest match reachable address prefixes known to the switch. When there are multiple longest match reachable address prefixes known to the switch, the route selection algorithm first attempts to find routes to reachable addresses with types of greatest precedence. Among multiple longest match reachable address prefixes of the same type, routes with the least total administrative weight are chosen first.
Local internal reachable addresses, whether learned via ILMI or as static routes, are given highest precedence; precedence value one. The precedence of other reachable address types is configurable.
To configure the ATM router PNNI precedence, perform the following task in global configuration mode. The no form of this command reverts to the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example configures all PNNI remote exterior routes with a precedence value of four:
Use the show atm pnni precedence command to display the ATM PNNI route determination precedence configuration.
To display the ATM PNNI route determination precedence configuration, perform the following task in user EXEC mode:
The following example displays ATM PNNI route determination precedence configuration:
You can configure administrative weight to indicate the relative desirability of using a link. In addition to the per interface atm pnni administrative-weight command, the ATM router PNNI administrative weight command can be used to change the default administrative weight assignment. For example, assigning equal administrative weights to all links in the network will minimize the number of hops used by each connection. For more information see the section "Administrative Weight."
To configure the ATM router PNNI administrative weight mode, perform the following task in global configuration mode using the no form of this command to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
linespeed—Used to set default administrative weight to values determined by the interface line speed (maximum cell rate).
Note Higher linespeeds (maximum cell rates) have lower administrative weight and are selected first during routing.
uniform—Used to set all interface administrative weights that are not explicitly configured to the uniform value of 5040.
Figure 11-6 is an example of how administrative weight affects call routing.
In Figure 11-6, the top network is configured as uniform causing equal administrative weight to be assigned to each link. In the bottom network configuration the same network is configured as linespeed. The links between SW1 and SW2 (SW1p1->SW2p1) and between SW2 and SW3 (SW2p2->SW3p2) are both faster OC12 connections and have lower administrative weights. PNNI interprets the route over the two OC12 links as being administratively equivalent to a more direct route between SW1 and SW3 using the OC 3 connection.
The following example configures administrative weight for the node as line speed:
Use the show atm pnni node command to display the administrative weight mode configuration for the individual interfaces.
To display the administrative weight configuration, perform the following task in user EXEC mode:
The following example displays the administrative weight (AW) configuration for the node:
Administrative weight (AW) is the main metric used for computation of the paths by PNNI. The assignment of administrative weights to links and nodes impacts the way PNNI selects paths in the private ATM network. For more detailed information see the section "Administrative Weight."
To configure the maximum administrative weight percentage, perform the following task in global configuration mode using the no form of this command to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
Note Configure the maximum administrative weight command only if background route computation is enabled. See the section "Configure On-Demand and Background Path Updates."
The following example configures the node maximum administration weight value as 300:
Use the show atm pnni node command to display the node ATM PNNI maximum administrative weight configuration.
To display the node ATM PNNI maximum administrative weight percentage configuration, perform the following task in user EXEC mode:
The following example displays maximum administrative weight percentage configuration:
The resource management (RM) poll interval specifies how often PNNI polls RM to update the values of link metrics and attributes. Configuration of the resource poll interval allows you to control the trade-off between the processing load and the accuracy of PNNI information. A larger value will probably generate a smaller number of PTSE updates. A smaller value results in greater accuracy in tracking resource information.
To configure the resource management poll interval, perform the following task in global configuration. The no form of this command to assign the default value:
At the privileged EXEC prompt, enter configuration mode from the terminal. |
|
|---|---|
At the configure prompt, enter ATM router PNNI mode from the terminal. The prompt will change to Switch(config-atm-router)#. |
|
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example configures the RM poll interval to 10 seconds:
To display the RM poll interval configuration, perform the following task in user EXEC mode:
The following example displays the RM poll interval configuration:
The following statistics on routing of ATM connections can be gathered:
To enable ATM PNNI statistics, perform the following task in global configuration mode using the no form of this command to assign the default value:
At the privileged EXEC prompt, enter configuration mode from the terminal. |
|
|---|---|
At the configure prompt, enter ATM router PNNI mode from the terminal. The prompt will change to Switch(config-atm-router)#. |
|
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example enables PNNI ATM statistics gathering:
To display the ATM PNNI statistics, perform the following task in user EXEC mode:
The following example displays the ATM PNNI statistics:
Each LightStream 1010 ATM switch is modeled as a single lowest-level PNNI node (locally identified as node 1). The node command is used to change the level of a node and to disable and enable a node. This causes the node ID and peer group ID of the node to be recalculated based on the level and the first ATM address.
When the node command is entered, the switch changes to node configuration mode.
To change the ATM address of the switch, the level of the node and to recalculate the node ID and peer group ID based on the new ATM address and level, perform the following task in global configuration mode:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example changes the ATM address of the switch from the autoconfigured address 47.0091.8100.0000.0041.0b0a.1081.0041.0b0a.1081.00 to the new address prefix 47.0091.8100.5670.0000.0000.1122.0041.0b0a.1081.00 changes the level of the node to 96, and causes the node ID and peer group ID to be recalculated:
To display the ATM PNNI node configuration, perform the following task in user EXEC mode:
This section describes configuration of the following LightStream 1010 ATM switch PNNI node attributes:
To configure the ATM router PNNI node, perform the following task in global configuration mode:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
Following is an example of entering node level configuration mode and displaying the variables using the ? help command:
To configure the ATM router PNNI node, perform the following task in global configuration mode using the no form of the ATM router node configuration commands to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
Configure the name of the node as eng_1 using the name command, as in the following example:
Use the show atm pnni node command to display the ATM PNNI node name configuration.
To display the ATM PNNI node name configuration, perform the following task in user EXEC mode:
This example displays the ATM node name configuration using the show atm pnni node command from user EXEC mode:
Address summarization allows scalability across multiple networks. By default the node has a summary address equal to the 13-byte address prefix of the ATM address of the switch. This address prefix is advertised into its peer group.
Multiple addresses can be configured for a single switch and this configuration can be used during ATM address migration. ILMI registers end systems with multiple prefixes during this period until an old address is removed. PNNI automatically creates 13-byte summary address prefixes from all of its ATM addresses.
Summary address prefixes can also be manually configured using the summary-address command. A node can have multiple summary address prefixes.
Note The command no auto-summary removes the default summary address(es). The no auto-summary command should be used whenever systems matching the first 13-bytes of this switch's ATM address(es) are attached to different switches. It can also be used for security reasons.
To configure address summarization, perform the following task in global configuration mode:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example removes the default summary address(es) and adds summary address 47.009181005670:
Use the show atm route command to display the ATM PNNI summary address configuration.
To display the ATM PNNI summary address configuration, perform the following task in user EXEC mode:
The following example displays the ATM PNNI summary address configuration:
Addresses and address prefixes used in PNNI can have scopes associated with either their individual addresses or their group address. See the section "Scope of Addresses" for detailed ATM PNNI scope description.
ATM addresses and prefixes may be configured automatically (which is the default setting) or manually. Configuring the ATM address scope as automatic sets all connections on the node as scope 1 to 15.
The individual and group ATM addresses and prefixes may have scopes configured manually using the following commands:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example changes ATM PNNI router scope mode to automatic for node 1:
Use the show atm scope map command to display the ATM PNNI scope configuration.
To display the ATM PNNI address and address prefix scope configuration, perform the following task in user EXEC mode:
The following example displays the ATM PNNI scope mode configuration:
Addresses and address prefixes used in PNNI can have scopes associated with either their individual address or theirgroup address. See the section "Scope of Addresses"for detailed ATM PNNI scope description.
The individual and group ATM addresses and prefixes may have thier scopes configured manually using the following commands:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
Use the no form of the scope map command to return the node to the default scope mapping configuration.
The following example configures scope mapping as low organization scope 1 and high organization scope 15 to PNNI level 56 for node 1:
Use the show atm scope map command to display the ATM PNNI scope configuration.
To display the node redistribution configuration, perform the following task in user EXEC mode:
The following example displays the ATM PNNI scope mapping manual configuration:
Redistribution instructs PNNI to distribute reachability information from non-PNNI sources throughout the PNNI routing domain. The LightStream 1010 ATM switch supports redistribution of static routes, such as those configured on IISP interfaces.
To enable redistribution of static routes throughout the PNNI routing domain, perform the following task in global configuration mode using the no form of this command to stop redistribution:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example enables redistribution of static routes:
Use the show atm pnni node command to display the redistribution configuration.
To display the node redistribution configuration, perform the following task in user EXEC mode:
The following example displays the node redistribution configuration:
Transit calls are calls originating from another ATM switch and passing through the switch. Some edge switches may want to eliminate this transit traffic and only allow traffic originating or terminating at this switch.
To configure whether a node allows transit calls, perform the following task in ATM PNNI node level configuration mode using the no form of this command to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The following example enables the transit restricted feature:
Use the show atm pnni node command to display the transit restriction configuration.
To display the ATM PNNI transit restriction configuration, perform the following task in user EXEC mode:
The following example displays the ATM PNNI transit restriction configuration of a node named eng_1:
PTSPs would overwhelm the network if they were transmitted every time any parameter in the network changed. To avoid this, PNNI uses significant change thresholds that control origination of PTSEs.
Note Any change in administrative weight and cell loss ratio is considered significant and triggers origination of a new PTSE instance.
To configure the PTSE significant change thresholds, perform the following task in global configuration mode using the no form of this command to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
The significant change variables are:
For an example of other ptse command keywords, see the section "Configure PNNI Hello, Database Synchronization and Flooding Parameters."
The following example configures a PTSE being sent only if the available cell rate changes 30 percent from the current metric:
Use the show atm pnni resource-info command to display the significant change threshold configuration.
To display the PTSE configuration, perform the following task in user EXEC mode:
The following example displays the significant change threshold configuration:
PNNI uses the Hello protocol to determine the status of neighbor nodes and PTSEs to disseminate topology database information in the ATM network. See the sections "The Hello Protocol," "Database Synchronization," and "Topology Information in PNNI" for more information.
To configure the Hello, Database Synchronization, and flooding parameters, perform the following task in global configuration mode using the no form of this command to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
ack-delay—delay before acknowledging packets
hello-holddown—hello generation hold down time
inactivity-factor—multiplier of hello interval
retransmit-interval—PTSE retransmit interval
The following example configures the PTSE refresh interval at 600 seconds:
The following example configures the retransmission of the Hello timer to 60 seconds:
Use the show atm pnni node command to display the ATM PNNI Hello, database synchronization and flooding configuration.
To display the ATM PNNI Hello, database synchronization, and flooding configuration, perform the following task in user EXEC mode:
The following example displays the ATM PNNI Hello, database synchronization, and flooding configuration:
The following sections describe configuration of the LightStream 1010 ATM switch PNNI interfaces:
Administrative weight is the main metric used for computation of the paths by PNNI. The assignment of administrative weights to links and nodes impacts the way PNNI selects paths in the private ATM network. For more detailed information see the section "Administrative Weight."
To configure the administrative weight for an individual interface, perform the following task in global configuration mode using the no form of this command to assign the default value:
| Task | Command |
|---|---|
At the privileged EXEC prompt, enter configuration mode from the terminal. |
|
Specify an ATM interface and enter interface configuration mode. |
|
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
number—A value from 1 to 1000000.
traffic class—CBR, VBR-RT, VBR-NRT, ABR, UBR, or all.
The following example configures ATM interface 0/0/0 with ATM PNNI administration weight of 7560 for traffic class ABR:
Use the show atm pnni command to display the ATM PNNI interface administrative weight configuration.
To display the ATM PNNI interface administrative weight configuration, perform the following task in user EXEC mode:
The following example displays the administrative weight for interface 0/0/0 configuration:
The Lightstream 1010 ATM Switch Link Selection feature allows you to choose the mode for selecting one specific link among several parallel links.
When multiple parallel links are configured inconsistently, the order of precedence of configured values is as follows:
For example, if any of the links is configured as admin-weight minimize, that is used for the entire link group.
To configure ATM PNNI link selection, perform the following task in global configuration mode using the no form of this command to assign the default value:
| 1These commands are documented in the LightStream 1010 ATM Switch Command Reference publication. |
admin-weight-minimize—Transmits call on the interface with lowest administrative weight.
blocking-minimize—Minimizes subsequent call blocking.
load-balance—Balances calls across all parallel links.
transmit-speed-maximize—Transmits call on highest speed parallel link.
The following example configures ATM interface 0/0/0 to use the transmit-speed-maximize link selection mode:
Use the show atm pnni neighbor command to display the PNNI link selection configuration for all interfaces.
To display the ATM PNNI link selection configuration, perform the following task in user EXEC mode:
The following example displays detailed PNNI link selection configuration:
Posted: Thu Jan 23 21:10:31 PST 2003
All contents are Copyright © 1992--2002 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.
