1. What is the difference between a. and a. Why does the maximum packet lifetime have to be large enough to ensure that not only the packet but also its acknowledgments have disappeared? Answer No: 1 The maximum packet lifetime needs to be large enough to ensure that both the packet and its acknowledgments have disappeared because TCP acknowledgments are cumulative. This means that when a packet is acknowledged, it implies that all previous packets have been received successfully. If the acknowledgment for a packet is lost and the packet lifetime expires, the sender might retransmit the packet, assuming it was lost, which could lead to duplicate packets being received by the receiver. Therefore, ensuring that both the packet and its acknowledgments have disappeared helps in…show more content… Answer No: 2 Nagle's algorithm optimizes TCP transmission by buffering small amounts of data until a full segment can be sent, reducing overhead. It suggests the sender waits for 1 RTT before transmitting, ensuring efficient use of network resources. However, in congested networks, this delay can exacerbate latency issues as data accumulates for transmission, potentially impacting real-time applications. This can lead to increased perceived latency and reduced throughput, especially if the network remains congested for prolonged periods. Therefore, Nagle's algorithm might not be ideal for applications sensitive to latency in heavily congested network conditions. 3. What is the difference between a'smart' and a'smart'? Give two examples of cases where TCP sends data-less packets on an established connection (which is not being torn down). Answer No: 3 Two examples of cases where TCP sends data-less packets on an established connection (which is not being torn down) are: Keep-alive packets: These are used to maintain the connection between two hosts by periodically checking if the other end is still
