As discussed previously, there are 4 techniques in closed-loop congestion loop control. In this post, we are going to learn about the first technique.
→Backpressure
Here is the post of introduction to congestion if you haven't seen it yet:
Backpressure:
crux points:
- Node to node congestion control.
- Occurs opposite to the direction of data flow.
- Causes upstream node to be more congested.
→upstream node is the node which is the immediate node of the congested node. - Congested node tells the upstream node to slow down the data rate.
Carefully, follow these below points by observing the diagram:
- Data is flowing from source to destination (left to right).
- At node 3, congestion has taken place in the network.
- Now, what this backpressure technique says is, "whenever a node becomes congested, it stops receiving the data from its immediate upstream node which makes the upstream node to be more congested."
Confused? Don't worry, let's move on!
- Here, in this case,node-3 is a congested node, the upstream node to node-3 is node-2. Now,node-3 stops receiving data from node-2. If node-3 doesn't accept the data,node-2 becomes exhausted and it becomes congested node now.
- Similarly, when node-2 becomes congested, it stops receiving data from its immediate upstream node i.e,node-1. Now,node-1 becomes a congested node.
- When node-1 becomes congested, it also again stops receiving data from its immediate upstream node i.e, source node. Now, the source node becomes a congested node.
- When the source node becomes congested, the source finally gets to know that congestion has occurred in a network and slows down the data rate which reduces congestion.
Take your own time to understand the mechanism properly. If not, just remember the crux points of backpressure.
In the next post, we will learn about the remaining techniques of closed-loop congestion control.
Comments
Post a Comment