How does TCP implement reliable data transfer using acknowledgments and timeouts?

• A. The sender transmits data and waits for ACKs; missing ACKs trigger retransmission via timeouts or fast retransmit when duplicate ACKs are received.
• B. Data is never retransmitted.
• C. Receiver never acknowledges.
• D. Uses parity bits only.

Answer: (A)

TCP ensures reliable data transfer by pairing each data segment with a sequence number and requiring acknowledgments from the receiver. The sender transmits data and waits for an acknowledgment that confirms receipt up to a certain point. If an acknowledgment doesn’t arrive in time, a timer expires and the sender retransmits the unacknowledged data. TCP also uses fast retransmit: if the sender sees duplicate acknowledgments, it infers that a segment was lost and retransmits it immediately, even before the timeout fires. This combination of timeouts and quick retransmissions keeps the data flowing reliably in the presence of losses. While acknowledgments are typically cumulative and the receiver advertises its available window for flow control, the reliability hinge is on ACKs and timeouts. The other options don’t provide a mechanism to recover lost data—data isn’t kept from retransmission, receivers must acknowledge, and parity bits alone don’t guarantee delivery.