[c-nsp] ASR9k Bundle QoS in 6.0.1

[Robert Williams]

Hi,

Got to the bottom of this at last!

[The new command enables a feature which works as follows]

- Whenever a policy is applied on the bundle member – first a ratio can be calculated based on total bundle bandwidth to bundle member’s bandwidth as follows: (ratio = bundle bandwidth/member bandwidth)
- For example if the bundle bandwidth is 20 Gbps with two bundle member bandwidth 10 Gbps, then the reduction will be 2/1 (0.5 * policy rate) for both members.
- With bundle bandwidth 50G (with 10G+40G members) the ratios become 5/1 and 5/4 respectively.
- The feature will then automatically recalculate the member-port QoS rate whenever a change among bundle occurs. i.e. bundle member active/down/added/removed.
- If traffic is load balanced well among the bundle members, then in aggregate the bundle-ether traffic will be shaped to the policy rate, matching the QoS policy configuration (instead of being members*rate).

[CLI for Aggregated Bundle QoS Mode]

Enable the mode:  (config)# hw-module all qos-mode bundle-qos-aggregate-mode
Disable the mode: (config)# no hw-module all qos-mode bundle-qos-aggregate-mode

- The above commands take effect chassis wide. (I don't believe you can enable/disable it per bundle).
- When the aggregated bundle mode changes, QoS polices on bundle intfs/subintfs are modified automatically.
- Reloading line card is not required.

So essentially it will now automatically do what we have been doing manually by recalculating the 'rate' based on the desired rate (div) No. members.

From what I can see (at least in our scenario) the pros/cons are:

[Advantages]
- Is proportional to unbalanced members (40G+10G) or (10G+1G).
- Automatically responds to addition/loss of a member, LC failure etc. dynamically.
- Doesn't require future engineers to remember to make changes to the base rate when modifying bundle members.

[Disadvantages]
- Assumes good balance, will over-police if traffic is not evenly distributed between member ports.
- Still doesn't fix the problem of an egress rate limit on a BVI being applied (multiplied by) the number of ingress NPs when the destination port is a single interface.

Hope that may be of some use to anyone looking for this in the future!

Cheers,


Robert Williams
Custodian Data Centre
Email: Robert at CustodianDC.com
http://www.CustodianDC.com