Enabling Eight Queues on ATM Interfaces
By default, IQ, MPC, and DPC interfaces on M120, T320, T640,
T1600, TX Matrix, and TX Matrix Plus routers, and MIC or MPC interfaces
on MX Series routers, are restricted to a maximum of four egress queues
per interface. You can enable eight egress queues by including the max-queues-per-interface statement at the [edit chassis
fpc slot-number pic pic-number] hierarchy level:
[edit chassis fpc slot-number pic pic-number] max-queues-per-interface (4 | 8);
The numerical value can be 4 or 8.
If you include the max-queues-per-interface statement,
all ports on the PIC use the configured maximum.
When you include the max-queues-per-interface statement
and commit the configuration, all physical interfaces on the PIC are
deleted and re-added. Also, the PIC is taken offline and then brought
back online immediately. You do not need to manually take the PIC
offline and online. You should change modes between four queues and
eight queues only when there is no active traffic going to the PIC.
When you are considering enabling eight queues on an ATM2 IQ interface, you should note the following:
ATM2 IQ interfaces using Layer 2 circuit trunk transport mode support only four CoS queues.
ATM2 IQ interfaces with MLPPP encapsulation support only four CoS queues.
You can configure only four RED profiles for the eight queues. Thus, queue 0 and queue 4 share a single RED profile, as do queue 1 and queue 5, queue 2 and queue 6, and queue 3 and queue 7. There is no restriction on EPD threshold per queue.
The default chassis scheduler allocates resources for queue 0 through queue 3, with 25 percent of the bandwidth allocated to each queue. When you configure the chassis to use more than four queues, you must configure and apply a custom chassis scheduler to override the default. To apply a custom chassis scheduler, include the
scheduler-map-chassisstatement at the[edit class-of-service interfaces at-fpc/pic/*] hierarchy level. For more information about configuring and applying a custom chassis scheduler, see Applying Scheduler Maps to Chassis-Level Queues.
Example: Enabling Eight Queues on ATM2 IQ Interfaces
In Figure 1, Router A generates
IP packets with different IP precedence settings. Router B is
an M320 router or a T Series router with two ATM2 IQ interfaces. On
Router B, interface at-6/1/0 receives traffic from
Router A, while interface at-0/1/0 sends traffic to
Router C. This example shows the CoS configuration for Router B.
On Router B:
[edit chassis]
fpc 0 {
pic 1 {
max-queues-per-interface 8;
}
}
fpc 6 {
pic 1 {
max-queues-per-interface 8;
}
}
[edit interfaces]
at-0/1/0 {
atm-options {
linear-red-profiles {
red_1 queue-depth 1k high-plp-threshold 50 low-plp-threshold 80;
red_2 queue-depth 2k high-plp-threshold 40 low-plp-threshold 70;
red_3 queue-depth 3k high-plp-threshold 30 low-plp-threshold 60;
red_4 queue-depth 4k high-plp-threshold 20 low-plp-threshold 50;
}
scheduler-maps {
sch_red {
vc-cos-mode strict;
forwarding-class fc_q0 {
priority high;
transmit-weight percent 5;
linear-red-profile red_1;
}
forwarding-class fc_q1 {
priority low;
transmit-weight percent 10;
linear-red-profile red_2;
}
forwarding-class fc_q2 {
priority low;
transmit-weight percent 15;
linear-red-profile red_3;
}
forwarding-class fc_q3 {
priority low;
transmit-weight percent 20;
linear-red-profile red_4;
}
forwarding-class fc_q4 {
priority low;
transmit-weight percent 5;
linear-red-profile red_1;
}
forwarding-class fc_q5 {
priority low;
transmit-weight percent 10;
linear-red-profile red_2;
}
forwarding-class fc_q6 {
priority low;
transmit-weight percent 15;
linear-red-profile red_3;
}
forwarding-class fc_q7 {
priority low;
transmit-weight percent 20;
linear-red-profile red_4;
}
}
sch_epd {
vc-cos-mode alternate;
forwarding-class fc_q0 {
priority high;
transmit-weight percent 5;
epd-threshold 1024;
}
forwarding-class fc_q1 {
priority low;
transmit-weight percent 10;
epd-threshold 2048;
}
forwarding-class fc_q2 {
priority low;
transmit-weight percent 15;
epd-threshold 3072;
}
forwarding-class fc_q3 {
priority low;
transmit-weight percent 20;
epd-threshold 4096;
}
forwarding-class fc_q4 {
priority low;
transmit-weight percent 5;
epd-threshold 2048;
}
forwarding-class fc_q5 {
priority low;
transmit-weight percent 10;
epd-threshold 3072;
}
forwarding-class fc_q6 {
priority low;
transmit-weight percent 15;
epd-threshold 4096;
}
forwarding-class fc_q7 {
priority low;
transmit-weight percent 20;
epd-threshold 5120;
}
}
}
}
atm-options {
vpi 0;
}
unit 0 {
vci 0.100;
shaping {
cbr 1920000;
}
atm-scheduler-map sch_red;
family inet {
address 172.16.0.1/24;
}
}
unit 1 {
vci 0.101;
shaping {
vbr peak 1m sustained 384k burst 256;
}
atm-scheduler-map sch_epd;
family inet {
address 172.16.1.1/24;
}
}
}
at-6/1/0 {
atm-options {
vpi 0;
}
unit 0 {
vci 0.100;
family inet {
address 10.10.0.1/24;
}
}
unit 1 {
vci 0.101;
family inet {
address 10.10.1.1/24;
}
}
}
[edit class-of-service]
classifiers {
inet-precedence inet_classifier {
forwarding-class fc_q0 {
loss-priority low code-points 000;
}
forwarding-class fc_q1 {
loss-priority low code-points 001;
}
forwarding-class fc_q2 {
loss-priority low code-points 010;
}
forwarding-class fc_q3 {
loss-priority low code-points 011;
}
forwarding-class fc_q4 {
loss-priority low code-points 100;
}
forwarding-class fc_q5 {
loss-priority low code-points 101;
}
forwarding-class fc_q6 {
loss-priority low code-points 110;
}
forwarding-class fc_q7 {
loss-priority low code-points 111;
}
}
forwarding-classes {
queue 0 fc_q0;
queue 1 fc_q1;
queue 2 fc_q2;
queue 3 fc_q3;
queue 4 fc_q4;
queue 5 fc_q5;
queue 6 fc_q6;
queue 7 fc_q7;
}
interfaces {
at-6/1/0 {
unit * {
classifiers {
inet-precedence inet_classifier;
}
}
}
}
}
[edit routing-options]
static {
route 10.10.20.2/32 {
next-hop at-0/1/0.0;
retain;
no-readvertise;
}
route 10.10.1.2/32 {
next-hop at-0/1/0.1;
retain;
no-readvertise;
}
}
To see the results of this configuration, you can issue the following operational mode commands:
show interfaces at-0/1/0 extensiveshow interfaces queue at-0/1/0show class-of-service forwarding-class