Reliable Data Transport

• sending side
  • rdt_send(data) -> packet

    called from above(e.g. by application). Passed data to deliver to receiver upper layer

  • udt_send(packet)

    called by RDT, to transfer packet over unrealiable channel to receiver

• receiving side
  • rdt_rcv() -> packet

    called when packet arrives on receive-side of channel

  • deliver_data(packet) -> data

    called by RDT to deliver data to upper layer

What to do

• develop sender, receiver sides of reliable data transfer protocol
• consider only unidirectional data transfer
  • control information will flow on both directions
• Use Finite State Machines(FSM) to specify sender, receiver

RDT1.0

Reliable transfer over a reliable channel

• underlying perfectly reliable channel
  • no bit errors
  • no loss of packets
• separate FSMs for sender, receiver
  • sender sends data into underlying channel
  • receiver reads data from underlying channel

RDT2.0

Reliable transfer over a channel with bit error

Fatal Flaw

ACK NAK may be corrupted

RDT2.1

Tag ACK NAK with flag 0 or 1

RDT2.2

a NAK-free Protocol

Tag packet with segment number (i++)

RDT3.0

RDT over a lossy channel with bit error

Strategy

• retransmit if no ACK received(timeout)

Scenario

• if a packet just delayed not lost:
  • retransmission will be duplicated(solved by seq_number)

Drawback

Terrible Performance

1Gbps link,15ms e-e prop.delay,1KB packet

$$T_{transmit}=\frac{8\times {10}^{3}b/pkt}{1\times {10}^{9}b/s}=8\times {10}^{-6}s$$$$RTT=15ms\times 2=3\times {10}^{-2}s$$

1KB every ~30ms $\Rightarrow$ 33KB/s over a 1Gbps link