Flow control and congestion control are favourite topics in GATE, university exams and networking interviews because they mix concepts with short calculations. This set covers stop-and-wait efficiency, Go-Back-N and Selective Repeat window sizes, the bandwidth-delay product, TCP slow start, congestion avoidance, fast retransmit and traffic shaping, with a worked explanation for every answer.
How to practise: read each question and tap the option you think is correct. You will see right away whether you got it, along with a short explanation, and your score is tracked at the bottom of the screen. This set has 15 questions.
Topics covered in this set
- Stop-and-wait ARQ and link utilisation
- Go-Back-N and Selective Repeat window sizes
- Bandwidth-delay product and pipelining
- TCP slow start, congestion avoidance and AIMD
- Fast retransmit, fast recovery and timeouts
- Leaky bucket, token bucket and Nagle's algorithm
Computer Networks MCQs with answers and explanations
Q1. A stop-and-wait link has a frame transmission time of 1 ms and a one-way propagation delay of 2 ms. Ignoring ACK transmission time, what is the link efficiency?
a) 20%
b) 33.3%
c) 50%
d) 25%
Q2. Go-Back-N uses 3-bit sequence numbers. What is the maximum sender window size?
a) 8
b) 4
c) 7
d) 3
Q3. Selective Repeat uses 4-bit sequence numbers. What is the maximum window size?
a) 15
b) 16
c) 4
d) 8
Q4. What is the receiver window size in Go-Back-N ARQ?
a) 1
b) Same as the sender window
c) 2^n - 1
d) Half of the sequence space
Q5. A Go-Back-N sender transmits frames 0 to 6. Frame 3 is lost and all other frames reach the receiver. How many frames does the sender retransmit after the timeout for frame 3?
a) 1
b) 3
c) 4
d) 7
Q6. A 1 Mbps link has a one-way propagation delay of 20 ms and uses 1000-byte frames. What is the minimum window size (in frames) for 100% utilisation?
a) 3
b) 5
c) 6
d) 11
Q7. TCP starts with cwnd = 1 MSS and ssthresh = 16 MSS. Assuming no losses and that cwnd changes once per RTT, what is cwnd after 6 RTTs?
a) 32 MSS
b) 18 MSS
c) 16 MSS
d) 64 MSS
Q8. A TCP Tahoe connection has cwnd = 20 MSS when a retransmission timeout occurs. What are the new ssthresh and cwnd?
a) ssthresh = 10, cwnd = 10
b) ssthresh = 20, cwnd = 1
c) ssthresh = 10, cwnd = 1
d) ssthresh = 5, cwnd = 1
Q9. What triggers TCP fast retransmit?
a) A retransmission timeout
b) Receiving three duplicate ACKs
c) A zero receive window
d) Receiving a SYN with a new sequence number
Q10. How does TCP Reno react to three duplicate ACKs, compared with Tahoe?
a) Reno halves cwnd and continues with fast recovery instead of dropping to 1 MSS
b) Reno closes the connection
c) Reno doubles cwnd
d) Reno behaves exactly like Tahoe
Q11. In TCP, how much unacknowledged data can a sender have in flight?
a) cwnd only
b) rwnd only
c) max(cwnd, rwnd)
d) min(cwnd, rwnd)
Q12. Which traffic shaping algorithm allows bursts up to a fixed size while limiting the long-term average rate?
a) Leaky bucket
b) Token bucket
c) Stop-and-wait
d) Choke packet
Q13. A token bucket has capacity 1 MB and a token rate of 2 MBps. The maximum output rate is 10 MBps. For how long can the host transmit at the full 10 MBps?
a) 100 ms
b) 500 ms
c) 200 ms
d) 125 ms
Q14. What is piggybacking in data link and transport protocols?
a) Sending an acknowledgement inside an outgoing data frame
b) Sending duplicate frames for reliability
c) Combining two connections into one
d) Retransmitting only damaged frames
Q15. Nagle's algorithm is mainly used to address which problem?
a) Count-to-infinity in routing
b) Hidden terminal problem
c) Sender-side silly window syndrome caused by many tiny segments
d) IP fragmentation
Quick revision notes
- Stop-and-wait efficiency = 1 / (1 + 2a), where a = propagation delay / transmission time.
- Window needed for full utilisation = 1 + 2a frames.
- Go-Back-N: sender window up to 2^n - 1, receiver window 1. Selective Repeat: both windows up to 2^(n-1).
- Slow start doubles cwnd every RTT until ssthresh; congestion avoidance then adds 1 MSS per RTT (AIMD).
- Timeout: ssthresh = cwnd / 2 and cwnd = 1 MSS. Three duplicate ACKs (Reno): ssthresh = cwnd / 2 and fast recovery.
- Token bucket burst time S = C / (M - p); leaky bucket gives a constant output rate.
Frequently asked questions
What is the difference between flow control and congestion control?
Flow control stops a fast sender from overrunning a slow receiver and is driven by the receiver's advertised window. Congestion control stops senders from overloading the network and is driven by the congestion window, which TCP adjusts based on losses and delays.
Why does Selective Repeat need a smaller window than Go-Back-N?
Because the Selective Repeat receiver buffers out-of-order frames. If the window were larger than half the sequence space, a retransmitted old frame could carry the same number as a new frame the receiver is expecting, and it would be accepted by mistake.
Is slow start actually slow?
No. The name refers to starting from a small window, but the window doubles every round trip, so growth is exponential until it reaches the slow start threshold.
Comments
Post a Comment