> Number of LSPs in the entire network: 140 * 20 * 32 * 200/1497=12000 LSPs
>
> The performance of IGP has always been affected by the size of the entire
> network's LSDB,
> and even if the periodic flooding time is reduced, there will still be
> convergence issues.


How total number of LSPs in the entire network matter for convergence ?

When link goes down adjacent nodes sitting on this link flood.

If node goes down adjacent nodes to the failed node flood.

Practically the large numbers look indeed scary on slides but have no real
meaning in real operation.

I also do not agree with your assertion that IGP performance is a function
of entrie network's LSDB. Lot's of optimizations went into decent
implementations of IGPs that total size of LSDB does not really matter.

Regards,
R.
_______________________________________________
Lsr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lsr

Reply via email to