nerdexam
Cisco

352-001 · Question #220

Cyclic congestion spikes are causing your Telnet users to experience delays. Traffic analysis shows minimal use of UDP. Which technology can you deploy to mitigate the problem?

The correct answer is B. Weighted RED. Cyclic congestion spikes with mostly TCP traffic indicate TCP global synchronization, where multiple flows back off and ramp up in unison. WRED breaks this synchronization by randomly dropping packets before the queue is full.

Design Considerations

Question

Cyclic congestion spikes are causing your Telnet users to experience delays. Traffic analysis shows minimal use of UDP. Which technology can you deploy to mitigate the problem?

Options

  • ACommitted Access Rate
  • BWeighted RED
  • CDeficit Round Robin
  • DClass Based Weighted Fair Queuing

How the community answered

(24 responses)
  • A
    4% (1)
  • B
    79% (19)
  • C
    4% (1)
  • D
    13% (3)

Why each option

Cyclic congestion spikes with mostly TCP traffic indicate TCP global synchronization, where multiple flows back off and ramp up in unison. WRED breaks this synchronization by randomly dropping packets before the queue is full.

ACommitted Access Rate

Committed Access Rate is a traffic policing mechanism that enforces rate limits but does not address TCP synchronization or smooth cyclic queue oscillation.

BWeighted REDCorrect

Weighted RED (WRED) proactively drops packets from individual flows as the queue depth rises, causing those TCP flows to reduce their transmission rate at different times rather than all simultaneously. This prevents the synchronized sawtooth pattern of global TCP back-off and ramp-up that produces cyclic congestion spikes, which directly addresses the delay Telnet users experience.

CDeficit Round Robin

Deficit Round Robin is a scheduling algorithm that fairly distributes bandwidth across flows but does not perform early random dropping and therefore does not mitigate TCP global synchronization.

DClass Based Weighted Fair Queuing

Class-Based Weighted Fair Queuing provides bandwidth guarantees per class and tail-drop queuing but does not break TCP synchronization - tail drop is actually a contributing cause of global synchronization.

Concept tested: WRED resolving TCP global synchronization congestion

Source: https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/qos_conavd/configuration/xe-16/qos-conavd-xe-16-book/qos-conavd-oview.html

Topics

#WRED#congestion avoidance#TCP optimization#QoS

Community Discussion

No community discussion yet for this question.

Full 352-001 Practice