Test: Congestion Control - Computer Science Engineering (CSE) MCQ

In slow-start algorithm, the size of the congestion window increases exponentially until it reaches a threshold. When it reaches the threshold, it stops increasing and continues sending packets through the threshold window thus preventing congestion.

In Virtual circuit networks, each node knows the upstream node from which a flow data is coming. So, it makes possible for the congested node to track the source of the congestion and then inform that node to reduce the flow to remove congestion.

In closed loop congestion control, methods are implemented to remove congestion after it occurs. Some of the methods used are backpressure and choke packet.

In Selective Repeat, the sender side uses a searching algorithm to find the packets which need to be retransmitted based on the negative acknowledgements received and then resends only those packets thus saving bandwidth.

Open loop congestion control techniques are used to prevent congestion before it even happens by enforcing certain policies. Retransmission policy, window policy and acknowledgement policy are some policies that might be enforced.

Choke packet is sent by a node to the source to inform it of congestion. Two choke packet techniques can be used for the operation called hop-by-hop choke packet and source choke packet.

In this closed loop congestion control technique, the congested node propagates in the opposite direction of the data flow to inform the predecessor node to reduce the flow of packets. This is why this technique is called a node-to-node congestion control technique.

The discarding policy adopted by the routers mainly states that the routers discard sensitive or corrupted packets that it receives, thus controlling the integrity of the packet flow. The discarding policy is adopted as an open loop congestion control technique.

Retransmission refers to the sender having to resend the packet to the receiver. It needs to be done only when some anomaly occurs with the packet like when the packet is lost or corrupted.

Open loop congestion control techniques are used to prevent congestion before it even happens by enforcing certain policies. Closed loop congestion control techniques are used to treat congestion after it has happened.