Scribd
Upload
26 views

Week 8

The document describes two scenarios of a distributed vector (DV) routing algorithm. In scenario 1 the DVs converge immediately, while in scenario 2 it takes 23 steps for the DVs of nodes X and Y to converge. The algorithm is inefficient in some scenarios. To improve efficiency, each node should track which neighbors use it as their first node out, and not update its DV based on those neighbors if a link cost changes. This would allow the DVs in scenario 2 to converge in 4 steps instead of 23.

Uploaded by

Anonymous 2ZSSleW
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
26 views

Week 8

The document describes two scenarios of a distributed vector (DV) routing algorithm. In scenario 1 the DVs converge immediately, while in scenario 2 it takes 23 steps for the DVs of nodes X and Y to converge. The algorithm is inefficient in some scenarios. To improve efficiency, each node should track which neighbors use it as their first node out, and not update its DV based on those neighbors if a link cost changes. This would allow the DVs in scenario 2 to converge in 4 steps instead of 23.

Uploaded by

Anonymous 2ZSSleW
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Scenario 1:

t0:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 1 0 1 Y 30 0 1

Z 31 1 0 Z 31 1 0

t1:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 1 0 1 Y 1 0 1

Z 31 1 0 Z 2 1 0

t2:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 1 0 1 Y 1 0 1

Z 2 1 0 Z 2 1 0
Scenario 2:

t0:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 32 0 1 Y 30 0 1

Z 31 1 0 Z 31 1 0

t1:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 32 0 1 Y 32 0 1

Z 31 1 0 Z 33 1 0

t2:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 34 0 1 Y 32 0 1

Z 33 1 0 Z 33 1 0

2.

It will take 23 steps for x and y’s DVs to converge.

t21:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 49 0 1 Y 49 0 1

Z 48 1 0 Z 50 1 0
t22:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 51 0 1 Y 49 0 1

Z 50 1 0 Z 50 1 0

t23:

DV Y X Y Z DV Z X Y Z

X 0 30 31 X 0 30 31

Y 51 0 1 Y 51 0 1

Z 50 1 0 Z 50 1 0

3.

This algorithm is inefficient in some scenarios.

4.

Have each node keep track of what neighbours use itself as their first node out. Then if a link-

cost change occurs the node will know not to update based on any nodes that use itself as the

first out node. So in scenario 2; node Y would know that node Z’s first out node for it’s path to

X is Y, so Y wouldn’t use node X to update its DV. Thus the estimate from Y to X would be

updated to 59. This way our DV’s would converge in 4 steps.

You might also like