egress_strict_priority_scheduler_with_bursty_traffic_test - openconfig/featureprofiles GitHub Wiki
This test verifies the behavior of an egress strict priority scheduler under bursty traffic conditions. By configuring multiple priority queues with specific traffic classes and generating bursty traffic that exceeds the interface capacity in short durations, we will validate that the scheduler maintains strict priority order, prioritizing the transmission of higher-priority traffic even during bursts, potentially leading to drops in lower-priority traffic.
-
DUT has 2 ingress ports and 1 egress port with the same port speed. The interface can be a physical interface or LACP bundle interface with the same aggregated speed.
| | ---- | ATE Port 1 | [ ATE Port 3 ] ---- | DUT | | | | | ---- | ATE Port 2 | -
Traffic classes:
- We will use 6 traffic classes NC1, AF4, AF3, AF2, AF1 and BE1.
-
Traffic types:
- All the traffic tests apply to both IPv4 and IPv6 and also MPLS traffic.
-
Queue types:
- NC1/AF4/AF3/AF2/AF1/BE1 will have strict priority queues (be1 - priority 6, af1 - priority 5, ..., nc1 - priority 1)
-
Test results should be independent of the location of interfaces. For example, 2 input interfaces and output interface could be located on
- Same ASIC-based forwarding engine
- Different ASIC-based forwarding engine on same line card
- Different ASIC-based forwarding engine on different line cards
-
Test results should be the same for port speeds 100G and 400G.
-
Counters should be also verified for each test case:
- /qos/interfaces/interface/output/queues/queue/state/transmit-pkts
- /qos/interfaces/interface/output/queues/queue/state/dropped-pkts
- transmit-pkts should be equal to the number of Rx pkts on Ixia port
- dropped-pkts should be equal to diff between the number of Tx and the number Rx pkts on Ixia ports
-
Latency:
- Should be < 100000ns
-
Forwarding Classes: Configure six forwarding classes (be1, af1, af2, af3, af4, nc1) based on the classification table provided.
-
Classification table
IPv4 TOS IPv6 TC MPLS EXP Forwarding Group 0 0-7 0 be1 1 8-15 1 af1 2 16-23 2 af2 3 24-31 3 af3 4,5 32-47 4,5 af4 6,7 48-63 6,7 nc1 -
Egress Scheduler: Apply a multi-level strict-priority scheduling policy on the desired egress interface. Assign priorities to each forwarding class as below: * be1 - priority 6 * af1 - priority 5 * af2 - priority 4 * af3 - priority 3 * af4 - priority 2 * nc1 - priority 1
-
Traffic Generation:
-
Generate the IPv4 traffic as below
- Interface Port 1:
Forwarding Group Traffic linerate (%) Frame size Expected Loss % be1 12 512 100 af1 12 512 100 af2 15 512 50 af3 12 512 0 af4 30 512 0 nc1 1 512 0 - Interface Port 2:
Forwarding Group Traffic linerate (%) Frame size Burst size Inter-pkt gap inter-burst gap Expected Loss % be1 | 20 | 256 | 50000 | 12 | 100 | 100 af1 | 13 | 256 | 50000 | 12 | 100 | 100 af2 | 17 | 256 | 50000 | 12 | 100 | 50 af3 | 10 | 256 | 50000 | 12 | 100 | 0 af4 | 20 | 256 | 50000 | 12 | 100 | 0 nc1 | 10 | 256 | 50000 | 12 | 100 | 0
-
-
Verification:
- Loss Rate: Capture packet loss for every generated flow and verify that loss for each flow does not exceed expected loss specified in the tables above.
- Telemetry: Utilize OpenConfig telemetry parameters to validate that per queue dropped packets statistics corresponds (with error margin) to the packet loss reported for every flow matching that particular queue.
-
Traffic Generation:
-
Generate the IPv6 traffic as below
- Interface Port 1:
Forwarding Group Traffic linerate (%) Frame size Expected Loss % be1 12 512 100 af1 12 512 100 af2 15 512 50 af3 12 512 0 af4 30 512 0 nc1 1 512 0 - Interface Port 2:
Forwarding Group Traffic linerate (%) Frame size Burst size Inter-pkt gap inter-burst gap Expected Loss % be1 | 20 | 256 | 50000 | 12 | 100 | 100 af1 | 13 | 256 | 50000 | 12 | 100 | 100 af2 | 17 | 256 | 50000 | 12 | 100 | 50 af3 | 10 | 256 | 50000 | 12 | 100 | 0 af4 | 20 | 256 | 50000 | 12 | 100 | 0 nc1 | 10 | 256 | 50000 | 12 | 100 | 0
-
-
Verification:
- Loss Rate: Capture packet loss for every generated flow and verify that loss for each flow does not exceed expected loss specified in the tables above.
- Telemetry: Utilize OpenConfig telemetry parameters to validate that per queue dropped packets statistics corresponds (with error margin) to the packet loss reported for every flow matching that particular queue.
-
Traffic Generation:
-
Generate the MPLS traffic as below
- Interface Port 1:
Forwarding Group Traffic linerate (%) Frame size Expected Loss % be1 12 512 100 af1 12 512 100 af2 15 512 50 af3 12 512 0 af4 30 512 0 nc1 1 512 0 - Interface Port 2:
Forwarding Group Traffic linerate (%) Frame size Burst size Inter-pkt gap inter-burst gap Expected Loss % be1 | 20 | 256 | 50000 | 12 | 100 | 100 af1 | 13 | 256 | 50000 | 12 | 100 | 100 af2 | 17 | 256 | 50000 | 12 | 100 | 50 af3 | 10 | 256 | 50000 | 12 | 100 | 0 af4 | 20 | 256 | 50000 | 12 | 100 | 0 nc1 | 10 | 256 | 50000 | 12 | 100 | 0
-
-
Verification:
- Loss Rate: Capture packet loss for every generated flow and verify that loss for each flow does not exceed expected loss specified in the tables above.
- Telemetry: Utilize OpenConfig telemetry parameters to validate that per queue dropped packets statistics corresponds (with error margin) to the packet loss reported for every flow matching that particular queue.
The below yaml defines the OC paths intended to be covered by this test.
paths:
## Config paths
### Classifiers
/qos/classifiers/classifier/config/name:
/qos/classifiers/classifier/config/type:
/qos/classifiers/classifier/terms/term/config/id:
/qos/classifiers/classifier/terms/term/conditions/ipv4/config/dscp-set:
/qos/classifiers/classifier/terms/term/conditions/ipv6/config/dscp-set:
/qos/classifiers/classifier/terms/term/conditions/mpls/config/traffic-class:
/qos/classifiers/classifier/terms/term/actions/config/target-group:
### Forwarding Groups
/qos/forwarding-groups/forwarding-group/config/name:
/qos/forwarding-groups/forwarding-group/config/output-queue:
### Queue
/qos/queues/queue/config/name:
### Interfaces
/qos/interfaces/interface/input/classifiers/classifier/config/name:
/qos/interfaces/interface/output/queues/queue/config/name:
/qos/interfaces/interface/output/scheduler-policy/config/name:
### Scheduler policy
/qos/scheduler-policies/scheduler-policy/config/name:
/qos/scheduler-policies/scheduler-policy/schedulers/scheduler/config/priority:
/qos/scheduler-policies/scheduler-policy/schedulers/scheduler/config/sequence:
/qos/scheduler-policies/scheduler-policy/schedulers/scheduler/config/type:
/qos/scheduler-policies/scheduler-policy/schedulers/scheduler/inputs/input/config/id:
/qos/scheduler-policies/scheduler-policy/schedulers/scheduler/inputs/input/config/input-type:
/qos/scheduler-policies/scheduler-policy/schedulers/scheduler/inputs/input/config/queue:
## State paths
/qos/interfaces/interface/output/queues/queue/state/name:
/qos/interfaces/interface/output/queues/queue/state/transmit-pkts:
/qos/interfaces/interface/output/queues/queue/state/transmit-octets:
/qos/interfaces/interface/output/queues/queue/state/dropped-pkts:
/qos/interfaces/interface/output/queues/queue/state/dropped-octets:
rpcs:
gnmi:
gNMI.Set:
gNMI.Subscribe:- MFF - A modular form factor device containing LINECARDs, FABRIC and redundant CONTROLLER_CARD components
- FFF - fixed form factor