0

This is the problem statement:

A link has a nominal rate: 4 kbps for the sender and 1 kbps for the receiver. A propagation delay of 20 ms, ACK size is 40 bytes. For what range of frame sizes does stop-and-wait give an efficiency of at least 50%?

I'm not asking you to solve this problem, but there is something not logical.

  1. How both ends are supposed to correctly interpret the received frames (and ACKs) if they don't have the same bit-rate or clock speed?
  2. Does the network adapter's buffer play a role in this situation?

Both ends have different clock speeds and therefore different bit-rates. I conclude that the clocks of both ends are not synchronized. I further conclude that the synchronization of the the signal sent from one end to another must be done on the physical layer so that the data link layer has the right interpreted bits.

Enlighten me.

MyNameIsUser
  • 119
  • 9

1 Answers1

0

Layering.

The answer was as simple as Layering. The physical layer always provides a consistent medium for communication between two NICs of different clock rates (and therefore digital bandwidth). This means that the physical layer defines the appropriate procedure for two different clocks to synchronize correctly and so the two NICs will not interpret the signals into erroneous bits.

Flow control is implemented on top of the physical layer (precisely, the data link layer) and therefore it does not tackle this low level task of synchronizing the two clocks. All what flow control does is: control the flow between the two (or more) NICs in order not to end up with the buffer being full and the the NICs dropping all of the incoming traffic. (each NIC has a temporary data buffer that holds incoming frames)

Ethernet itself has a built-in flow control mechanism, namely: Auto-negotiation. Ethernet is at layers 1 and 2 of the OSI model.

Building up on what I have said, the ACKs will be interpreted normally, and everything will go on just fine.

Community
  • 1
  • 1
MyNameIsUser
  • 119
  • 9