[c-nsp] OSPF per-prefix LFA

[Adam Vitkovsky]

> -----Original Message-----
> From: cisco-nsp [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of
> Gert Doering
> Sent: 28 May 2015 15:07
> To: George Giannousopoulos
> Cc: cisco-nsp at puck.nether.net
> Subject: Re: [c-nsp] OSPF per-prefix LFA
> 
> Hi,
> 
> On Thu, May 28, 2015 at 04:06:43PM +0300, George Giannousopoulos wrote:
> > As you have probably already noticed, after OSPF timers tuning, the
> > convergence is quite fast even without the LFA.. So why would you bother
> to
> > configure LFA in the fiirst place?
> 
> The interesting bits are in the two letters "L" and "F" here.
> 
> Link-state protocol convergence will cause microloops, LFA avoids those
> (if possible in the topology given).
> 

Right one might argue that on modern kit SPF recalculations are in usec scale.
However other burden on a path to fast convergence is the actual programing of HW which for the same amount of prefixes is in msec scale.
And LFA addresses this as the alternate path is already programed in HW.

adam
---------------------------------------------------------------------------------------
 This email has been scanned for email related threats and delivered safely by Mimecast.
 For more information please visit http://www.mimecast.com
---------------------------------------------------------------------------------------