[email protected]
[Top] [All Lists]

Re: LFAs and ECMP

Subject: Re: LFAs and ECMP
From: Alex Zinin
Date: Tue, 15 Feb 2005 13:50:27 -0800
Alia,

  When adding text on this to the document, please make sure the difference
  between using ECMP for load-balancing BF and using ECMP and LFAs for
  protection AF is emphasized.

  Another question we may want to think about is load-balancing AF via multiple
  LFAs.
  
-- 
Alex
http://www.psg.com/~zinin

Monday, February 14, 2005, 12:21:37 PM, Alia Atlas wrote:
> At 12:15 PM 2/14/2005, mike shand wrote:
>>>So describe the above example as a case of how ECMP doesn't guarantee 
>>>protection - and then up to the implementation/operator to decide what 
>>>the preference between protection type and cost is?
>>
>>Don't we already have to discuss the same sort of issues with respect to 
>>LFAs? ie. you may be able to get a viable LFA where you couldn't otherwise 
>>by relaxing the protection type.

> We do.  The draft describes the selection of an alternate for each primary 
> next-hop, so it's really part of that...  I can just add in an advisory bit 
> with the example.

>>But if it were a choice between using an ECMP which may not actually 
>>protect against the failure, and using an LFA which is guaranteed to 
>>protect against any failure, then I think I would probably choose the 
>>latter. It is only in cases where there was no "perfect" LFA that I might 
>>be tempted to use a less than unconditionally OK ECMP (or even LFA).

> Right.  The draft says one should pick an alternate that gives the best 
> protection possible.

>>But as you say, to some extent it is an operator/implementation choice.

> So I'll clarify a bit with the example - and leave it as the should.

> Alia


> _______________________________________________
> Rtgwg mailing list
> [email protected]
> https://www1.ietf.org/mailman/listinfo/rtgwg


_______________________________________________
Rtgwg mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/rtgwg

<Prev in Thread] Current Thread [Next in Thread>