Port-Shaper and LLQ in the Presence of EFPs

The Port-Shaper and LLQ in the Presence of EFPs feature allows network designers to configure port and class policies on ports that contain Ethernet Flow Points (EFPs). These policies support Low Latency Queueing (LLQ) and traffic prioritization across the EFPs.

Restrictions for Port-Shaper and LLQ in the Presence of EFPs

  • Due to NCS 520 hardware limitations, functionality of Egress QoS policy that is applied on the EFP shall not work as expected, if all the conditions mentioned below occur together:

    • Unknown unicast, multicast, and broadcast stream is received for the same EFP in the ingress direction.

    • Unknown unicast or multicast or broadcast traffic and other traffic that is targeted for same EFP, are classified under same class-map of the policy that is configured on the EFP.

    • Combined transmitted traffic rate (#1 and other traffic that is targeted for same EFP) for the EFP is higher than the configured shaper value.

  • For port-shaper in the presence of EFPs, policy on the port must be applied first followed by policy on EFP. Applying policy on EFP first followed by the port does not work.

  • If you configure a class-based policy on the port, then you cannot configure service-policies on Ethernet Flow Points (EFPs).

  • Attaching a service policy to the BDI is not supported.

  • Usage of bandwidth remaining percentage (BRP) in the absence of priority class, allocates the available bandwidth in an iterative way. For example, the bandwidth is allocated for the first BRP class as per the percentage of share that is configured in the respective class-map and the remaining bandwidth is iteratively allocated to all other BRP classes until the bandwidth is exhausted.

  • When the port is down, if you manually change the port speed then it may result in unexpected behavior.

  • You must remove bandwidth statement from the child policy before it is removed from the parent policy.

Information About Port-Shaper and LLQ in the Presence of EFPs

Ethernet Flow Points and LLQ

An Ethernet Flow Point (EFP) is a forwarding decision point in the provider edge (PE) router, which gives network designers flexibility to make many Layer 2 flow decisions within the interface. Many EFPs can be configured on a single physical port. (The number varies from one device to another.) EFPs are the logical demarcation points of an Ethernet virtual connection (EVC) on an interface. An EVC that uses two or more User-Network Interfaces (UNIs) requires an EFP on the associated ingress and egress interfaces of every device that the EVC passes through.

The Egress HQoS with Port Level Shaping feature allows network designers to configure port and class policies on ports that contain EFPs. These policies support Low Latency Queueing (LLQ) and traffic prioritization across the EFPs.

For information on how to configure LLQ, see the QoS Congestion Management Configuration Guide.

How to Configure Port-Shaper and LLQ in the Presence of EFPs

To configure the Port-Shaper and LLQ in the Presence of EFPs feature, you first create either a hierarchical or flat policy map that supports Low Latency Queueing (LLQ), which you then attach to an EFP interface.

Configuring Hierarchical Policy Maps

To configure hierarchical policy maps, you create child policies which you then attach to a parent policy. The parent policy is then attached to an interface.

Procedure

Step 1

enable

Example:


Device> enable

Enables privileged EXEC mode.

  • Enter your password if prompted.

Step 2

configure terminal

Example:


Device# configure terminal

Enters global configuration mode.

Step 3

policy-map policy-map-name

Example:


Device(config)# policy-map child-llq

Creates or modifies the child policy and enters QoS policy-map configuration mode.

  • child-llq is the name of the child policy map.

Step 4

class class-map-name

Example:


Device(config-pmap)# class qos-1

Assigns the traffic class you specify to the policy map and enters QoS policy-map class configuration mode.

  • qos-1 is the name of a previously configured class map and is the traffic class for which you want to define QoS actions.

Step 5

bandwidth percent percent

Example:


Device(config-pmap-c)# bandwidth percent 20

(Optional) Specifies a bandwidth percent for class-level queues to be used during congestion to determine the amount of excess bandwidth (unused by priority traffic) to allocate to nonpriority queues.

Step 6

exit

Example:


Device(config-pmap-c)# exit

Exits QoS policy-map class configuration mode.

Step 7

class class-map-name

Example:


Device(config-pmap)# class qos-2

Assigns the traffic class you specify to the policy map and enters QoS policy-map class configuration mode.

  • qos-2 is the name of a previously configured class map and is the traffic class for which you want to define QoS actions.

Step 8

bandwidth percent percent

Example:


Device(config-pmap-c)# bandwidth percent 80

(Optional) Specifies a bandwidth percent for class-level queues to be used during congestion to determine the amount of excess bandwidth (unused by priority traffic) to allocate to nonpriority queues.

Step 9

exit

Example:


Device(config-pmap-c)# exit

Exits QoS policy-map class configuration mode.

Step 10

policy-map policy-map-name

Example:


Device(config-pmap)# policy-map parent-llq

Creates or modifies the parent policy.

  • parent-llq is the name of the parent policy map.

Step 11

class class-default

Example:


Device(config-pmap)# class class-default

Configures or modifies the parent class-default class and enters QoS policy-map class configuration mode.

  • You can configure only the class-default class in a parent policy. Do not configure any other traffic class.

Step 12

service-policy policy-map-name

Example:


Device(config-pmap-c)# service-policy child-llq

Applies the child policy to the parent class-default class.

  • child-llq is the name of the child policy map configured in step 1.

Configuring an LLQ Policy Map

Procedure

Step 1

enable

Example:


Device> enable

Enables privileged EXEC mode.

  • Enter your password if prompted.

Step 2

configure terminal

Example:


Device# configure terminal

Enters global configuration mode.

Step 3

policy-map policy-map-name

Example:


Device(config)# policy-map llq-flat

Creates a policy and enters QoS policy-map configuration mode.

Step 4

class class-map-name

Example:


Device(config-pmap)# class qos-group

Assigns the traffic class you specify to the policy map and enters policy-map class configuration mode.

Step 5

priority

Example:


Device(config-pmap-c)# priority

Configures LLQ, providing strict priority queueing (PQ) for class-based weighted fair queueing (CBWFQ).

Step 6

exit

Example:


Device(config-pmap-c)# exit

Exits QoS policy-map class configuration mode.

Step 7

class class-map-name

Example:


Device(config-pmap)# class qos-group

Assigns the traffic class you specify to the policy map and enters QoS policy-map class configuration mode.

Step 8

shape average value

Example:


Device(config-pmap-c)# shape average 200000000

Configures a shape entity with a Comitted Information Rate of 200 Mb/s.

Step 9

exit

Example:


Device(config-pmap-c)# exit

Exits QoS policy-map class configuration mode.

Step 10

class class-map-name

Example:


Device(config-pmap)# class qos-group

Assigns the traffic class you specify to the policy map and enters QoS policy-map class configuration mode.

Step 11

bandwidth percent

Example:


Device(config-pmap-c)# bandwidth 4000000

(Optional) Specifies a bandwidth percent for class-level queues to be used during congestion to determine the amount of excess bandwidth (unused by priority traffic) to allocate to non-priority queues.

Step 12

exit

Example:


Device(config-pmap-c)# exit

Exits QoS policy-map class configuration mode.

Configuring Port Level Shaping on the Main Interface with Ethernet Flow Points

To configure port level shaping on the main interface with EFPS, first you enable the autonegotiation protocol on the interface, then you attach a policy map to the interface and finally you configure the Ethernet service instance.

Procedure

Step 1

enable

Example:


Device> enable

Enables privileged EXEC mode.

  • Enter your password if prompted.

Step 2

configure terminal

Example:


Device# configure terminal

Enters global configuration mode.

Step 3

interface type number

Example:


Device(config)# interface GigabitEthernet 0/0/1

Configures an interface type and enters interface configuration mode.

  • Enter the interface type number.

Step 4

no ip address

Example:


Device(config-if)# no ip address

Disables IP routing on the interface.

Step 5

negotiation auto

Example:


Device(config-if)# negotiation auto

Enables the autonegotiation protocol to configure the speed, duplex, and automatic flow control of the Gigabit Ethernet interface.

Step 6

service-policy output policy-map-name

Example:


Device(config-if)# service-policy output parent-llq

Specifies the name of the policy map to be attached to the input or output direction of the interface.

  • You can enter the name of a hierarchical or a flat policy map.

Step 7

service instance id ethernet

Example:


Device(config-if)# service instance 1 ethernet

Configures an Ethernet service instance on an interface and enters service instance configuration mode.

Step 8

encapsulation dot1q vlan-id

Example:


Device(config-if-srv)# encapsulation dot1q 100

Defines the matching criteria to map 802.1Q frames' ingress on an interface to the service instance.

Step 9

bridge-domain bridge-domain-id

Example:


Device(config-if-srv)# bridge-domain 100

Binds the bridge domain to the service instance.

Step 10

exit

Example:


Device(config-if-serv)# exit

Exits service instance configuration mode.

Step 11

service instance id ethernet

Example:


Device(config-if)# service instance 2 ethernet

Configures an Ethernet service instance on an interface and enters service instance configuration mode.

Step 12

encapsulation dot1q vlan-id

Example:


Device(config-if-srv)# encapsulation dot1q 101

Defines the matching criteria to map 802.1Q frames' ingress on an interface to the service instance.

Step 13

bridge-domain bridge-domain-id

Example:


Device(config-if-srv)# bridge-domain 101

Binds the bridge domain to the service instance.

Step 14

exit

Example:


Device(config-if-srv)# exit

Exits QoS policy-map class configuration mode.

Step 15

end

Example:


Device(config-if)# end

(Optional) Exits interface configuration mode.

Example: Configuring Hierarchical QoS

The router supports hierarchical QoS policies with up to three levels:

  • Port policy—Uses class-default only and should be applied on physical interface as initial configuration.

  • EFP policy—Uses a nested policy with parent policy matching on class-default and child policy matching on QoS-group values.

The following example displays the hierarchical QoS configuration using port policy and EFP policy: 


policy-map parent1
class class-default
shape average 1000000000 
!
policy-map EFP30
class class-default
shape average 500000000 
service-policy pq_test
!
end

Use the show run policy-map to verify the configuration: 


router#sh run policy-map pq_test
Building configuration...

Current configuration : 590 bytes
!
policy-map pq_test
class qg1
priority level 1 percent 20
queue-limit percent 10
class qg0
priority level 2 percent 10
queue-limit percent 10
class qg2
bandwidth remaining percent 15 
queue-limit percent 10
class qg3
bandwidth remaining percent 20 
queue-limit percent 10
class qg4
bandwidth remaining percent 25 
queue-limit percent 10
class qg5
bandwidth remaining percent 10 
queue-limit percent 10
class qg6
bandwidth remaining percent 25 
queue-limit percent 10
class class-default
bandwidth remaining percent 5 
queue-limit percent 10
!
end

Additional References

Related Documents

Related Topic

Document Title

Cisco IOS commands

Cisco IOS Master Commands List, All Releases

QoS commands: complete command syntax, command modes, command history, defaults, usage guidelines, and examples

Cisco IOS QoS Command Reference

Policing and shaping

"Policing and Shaping Overview" module

Class maps

"Applying QoS Features Using the MQC" module

Policy maps

"Applying QoS Features Using the MQC" module

Low Latency Queueing

QoS Congestion Management Configuration Guide

Standards and RFCs

Standard

Title

No new or modified standards are supported, and support for existing standards has not been modified.

--

MIBs

MIB

MIBs Link

No new or modified MIBs are supported, and support for existing MIBs has not been modified.

To locate and download MIBs for selected platforms, Cisco IOS XE software releases, and feature sets, use Cisco MIB Locator found at the following URL:

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

Technical Assistance

Description

Link

The Cisco Support and Documentation website provides online resources to download documentation, software, and tools. Use these resources to install and configure the software and to troubleshoot and resolve technical issues with Cisco products and technologies. Access to most tools on the Cisco Support and Documentation website requires a Cisco.com user ID and password.

http://www.cisco.com/cisco/web/support/index.html