Packet Queuing and Dropping in Routers

Routers are critical networking equipment that control data flow across networks. They contain limited buffer memory for temporarily storing packets before forwarding them to their destinations. When the arrival rate of packets exceeds the rate at which packets can be processed and transmitted, packet queuing and potentially packet dropping occurs.

This situation arises because routers must balance finite memory resources with varying traffic loads. Effective queuing disciplines are essential for managing packet flow, minimizing delays, and ensuring fair resource allocation among different traffic types.

Router Packet Queuing Process P1 P2 P3 Router Queue Buffer P7 P8 P9 P10 P11 Buffer Nearly Full New packets may be dropped P7 P4 Dropped

Causes of Queue Congestion

Queue congestion occurs when router buffer capacity is exceeded. Common causes include:

  • Input rate exceeding output rate − Packets arrive faster than they can be transmitted on the output link.

  • Multiple input aggregation − Combined traffic from all input interfaces exceeds total output capacity.

  • Processing limitations − Router CPU cannot handle forwarding table lookups fast enough for incoming packet volume.

Queuing Disciplines

First-In, First-Out (FIFO)

FIFO is the simplest and most common queuing discipline. Packets are processed in arrival order using a single queue. When the buffer becomes full, newly arriving packets are dropped (tail drop).

While simple to implement, FIFO provides no differentiation between traffic types, making it unsuitable for applications requiring guaranteed service levels like VoIP or video streaming.

Priority Queuing (PQ)

Priority Queuing divides router memory into multiple queues based on traffic priority levels (typically High, Medium, Low). Higher priority queues are always serviced before lower priority ones, ensuring critical traffic receives preferential treatment.

However, this approach can lead to starvation of lower priority traffic during periods of sustained high-priority traffic load.

Weighted Fair Queuing (WFQ)

WFQ creates separate queues for different traffic flows and allocates bandwidth proportionally based on assigned weights. For example, three active queues might receive 20%, 30%, and 50% of available bandwidth respectively.

This discipline prevents any single flow from monopolizing resources while providing fair access to network capacity across all active flows.

Comparison of Queuing Disciplines

Discipline Complexity Fairness Priority Support Starvation Risk
FIFO Low Equal treatment No None
Priority Queuing Medium Priority-based Yes High for low priority
Weighted Fair Queuing High Proportional Configurable Low

Conclusion

Packet queuing and dropping are fundamental router operations for managing network congestion and resource allocation. The choice of queuing discipline significantly impacts network performance, with each approach offering different trade-offs between simplicity, fairness, and quality of service guarantees.

Updated on: 2026-03-16T23:36:12+05:30

2K+ Views

Kickstart Your Career

Get certified by completing the course

Get Started
Advertisements