Question: In the slow start algorithm, the size of the congestion window grows until _____.

Answer:

In the slow start algorithm, the size of the congestion window grows exponentially until either the congestion window size reaches the value of the slow start threshold (ssthresh), or congestion is detected by the occurrence of packet loss.
During slow start, the congestion window size starts at one maximum segment size (MSS) and doubles for each successful transmission. This means that the congestion window size grows exponentially as more packets are successfully transmitted without congestion. However, if congestion is detected (e.g. due to packet loss or receipt of a congestion indication signal from the network), the congestion window size is reduced and the congestion control algorithm transitions to congestion avoidance mode.
The slow start algorithm is used by TCP to initially ramp up its sending rate after a connection is established or after a period of inactivity. It is intended to maximize network utilization without causing congestion or excessive packet loss.

MCQ: In the slow start algorithm, the size of the congestion window grows until _____.

Correct Answer: A. exponentially; it reaches a threshold

In the slow start algorithm, the size of the congestion window grows exponentially until either the congestion window size reaches the value of the slow start threshold (ssthresh), or congestion is detected by the occurrence of packet loss.
During slow start, the congestion window size starts at one maximum segment size (MSS) and doubles for each successful transmission. This means that the congestion window size grows exponentially as more packets are successfully transmitted without congestion. However, if congestion is detected (e.g. due to packet loss or receipt of a congestion indication signal from the network), the congestion window size is reduced and the congestion control algorithm transitions to congestion avoidance mode.
The slow start algorithm is used by TCP to initially ramp up its sending rate after a connection is established or after a period of inactivity. It is intended to maximize network utilization without causing congestion or excessive packet loss.