Tom Anderson, CSE/EE 461 Winter 2002.

These are the rules of the game by which we simulate TCP:

• Let there be one sender, one receiver, and one bottleneck router (which has a infinitely fast link in and a slow link out).
• Latency of first hop is 1 time step; latency of second hop is 0
• Bandwidth of first hop is infinite; bandwidth of second hop is 1 packet per time step
• Let post-its be packets; #s to indicate sequence #’s.
• Need approximately 30 post-its
• Let each packet be acked immediately; no delayed acks.
• Let the instructor be the clock that drives the simulation. Graph/observe any quantities at the end of each clock tick.
• Queuing in the router is represented by slots on the whiteboard – need space for four packets in each direction
• Note that we can also check the congestion window by seeing how many packets are "out there in the network" including any lost packets.

Simulate this as a warm-up exercise.

Graph:

• router queuing versus time, with an X on top when there is packet loss.
• number of successful sends over time (could be: acks versus time as a proxy for throughput (since sequence number is too hard to track).

Notice

• low link utilization (1 out of 4 clock ticks)
• negligible queuing at the router.

Blow out the queue until there is loss. Each ack causes congestion window to be increased by one – i.e., 2 packets to be sent.

Add graph for

• the congestion window versus time

Notice that

• many packets can be lost.
• bursts in transmission of at most 2x the bottleneck rate
• self-clocking effect in the acks.

Go through a couple of rounds of the sawtooth. Each successful ack triggers a send; every time a complete window is sent the window increases by one (ack triggers two sends). After a loss, congestion window drops by factor of 2. Carry a special object to signal each RTT round and hence congestion window increase.

Notice

• delay between loss and congestion window reduction.
• single packet loss
• transmission pause after loss is observed.