cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1073
Views
0
Helpful
2
Replies

N9K-C93240YC-FX2 NXOS queuing question

brian.holmes
Level 1
Level 1

With the four queue model setup like this:

 

policy-map type queuing Outbound
class type queuing c-out-8q-q7
priority level 1
class type queuing c-out-8q-q6
bandwidth remaining percent 30
class type queuing c-out-8q-q5
bandwidth remaining percent 20
class type queuing c-out-8q-q-default
bandwidth remaining percent 50

 

Do the "bandwidth remaining" statements determine the queue size percentages also?  or is that defined somewhere else?

 

i.e.  if we are seeing the Queue Depth increase past 0 on a show queuing output, would adjusting the bandwidth remaining statements give it more breathing room and stop the Q Depth Byts  from building as frequently?

 

 

 

 

Brian Holmes
Verizon
2 Replies 2

brian.holmes
Level 1
Level 1

or should I not be very concerned with the queue depth as long as drops are not incrementing also?

Brian Holmes
Verizon

Sergiu.Daniluk
VIP Alumni
VIP Alumni

Hi Brian,

For Nexus 9K switches, there are two bandwidth commands:

 

    Bandwidth percent:
    switch(config-pmap-c-que)# bandwidth percent {percent}

    Bandwidth remaining percent:
    switch(config-pmap-c-que)# bandwidth remaining percent {percent}

The first one, "bandwidth percent" command assigns a minimum rate of the interface bandwidth to an output queue as the percentage of the underlying interface link rate.

 

The "bandwidth remaining" command assigns the percentage of the bandwidth that remains (not already assigned) to this queue.

 

You can basically assign a minimum rate of the interface bandwidth or assign the percentage of the bandwidth that remains.

Now, why should I use one or another, you might ask? That is because if you do not configure a priority queue, the system-defined egress pq queues behave as normal queues. When a priority queue is configured, the other queues can only use the remaining bandwidth in the same policy map. That's why you see the default config like this:

policy-map type queuing default-8q-out-policy
  class type queuing c-out-8q-q7
   priority level 1
  class type queuing c-out-8q-q6
   bandwidth remaining percent 0
  class type queuing c-out-8q-q5
   bandwidth remaining percent 0
  class type queuing c-out-8q-q4
   bandwidth remaining percent 0
  class type queuing c-out-8q-q3
   bandwidth remaining percent 0
  class type queuing c-out-8q-q2
   bandwidth remaining percent 0
  class type queuing c-out-8q-q1
   bandwidth remaining percent 0
  class type queuing c-out-8q-q-default
   bandwidth remaining percent 100

"If there are strict-priority queues, however, the strict-priority queues receive their share of the bandwidth first. The remaining bandwidth is shared in a weighted manner among the class configured with a bandwidth percent. For example, if strict-priority queues take 90 percent of the bandwidth, and you configure 75 percent for a class, the class will receive 75 percent of the remaining 10 percent of the bandwidth".

 

For more details about how bandwidth is being allocated on Nexus 9K platform, I would recommend the following config guide:

https://www.cisco.com/c/en/us/td/docs/switches/datacenter/nexus9000/sw/93x/qos/configuration/guide/b-cisco-nexus-9000-nx-os-quality-of-service-configuration-guide-93x/b-cisco-nexus-9000-nx-os-quality-of-service-configuration-guide-93x_chapter_01000.h...

 

Stay safe,

Sergiu