400-007 · Question #127
How must the queue sizes be designed to ensure that an application functions correctly?
The correct answer is D. The sum of the queuing delay of all devices plus serialization delay in the chain must be less than. Application delay tolerance must be compared against the total accumulated delay across every device in the forwarding path, not against any single device in isolation.
Question
How must the queue sizes be designed to ensure that an application functions correctly?
Options
- AEach individual device queuing delay in the chain must be less than or equal to the application
- BThe queuing delay on every device in the chain must be exactly the same to the application
- CThe default queue sizes are good for any deployment as it compensates the serialization delay.
- DThe sum of the queuing delay of all devices plus serialization delay in the chain must be less than
How the community answered
(25 responses)- B4% (1)
- C4% (1)
- D92% (23)
Why each option
Application delay tolerance must be compared against the total accumulated delay across every device in the forwarding path, not against any single device in isolation.
Requiring each individual device delay to be below the full application tolerance ignores the cumulative nature of delay and would make per-device budgets far too generous.
Queuing delays do not need to be uniform across devices; only their aggregate sum relative to the application budget matters for correct design.
Default queue sizes are tuned for generic traffic patterns and can introduce excessive bufferbloat or starvation depending on interface speed and traffic mix, making them unsuitable as a universal baseline.
Each network device in the path adds its own queuing delay, and serialization delay is incurred at every link; these values are additive. The combined sum of all queuing and serialization delays along the entire chain must stay below the application's maximum one-way delay budget (for example, 150 ms for interactive voice) to ensure correct application behavior.
Concept tested: End-to-end cumulative queuing delay budget design
Source: https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/qos_nque/configuration/xe-16/qos-nque-xe-16-book/qos-nque-lowlat-queuing.html
Topics
Community Discussion
No community discussion yet for this question.