The MPLS WG Archive

[Eric Osborne <eosborne@cisco.com>]

On Tue, Jul 02, 2002 at 05:47:46PM +0000, Tissa Senevirathne wrote:
> Why are we saying ECMP hash is to pick a number between 1 and 6 ? Why not 
> 64 ?
> 

"...between 1 and N, where N in this example is 6".  Happy?
IETF should not be defining limits for things like these, and I don't
believe that's what we're talking about.  There happen to be both
implementation limits that may constrain N and also real-world limits
where the odds of having 64 equal-cost paths to a destination is
pretty low.  But it's purely a local matter.



eric

> In my opinion ECMP is entirely a local implementation issue. And at WG 
> requirement level one sentence to say
> " Any load balancing policies implemented MUST not mis order packets"
> is adequate
> 
> 
> >From: Hummel Heinrich <Heinrich.Hummel@icn.siemens.de>
> >To: "'erosen@cisco.com'" <erosen@cisco.com>,        Shahram Davari 
> ><Shahram_Davari@pmc-sierra.com>
> >CC: "'Eric Osborne'" <eosborne@cisco.com>,        George Sheng 
> ><george_s97@hotmail.com>, scullptor@yahoo.com,        mpls@UU.NET
> >Subject: AW: your mail Date: Tue, 2 Jul 2002 16:25:37 +0200
> >
> >It  gets weirder  and weirder: The MPLS WG invents a label stack, but does 
> >not care at all that any of the
> >deeper nested labels may ever be signalled/carried by means of a 
> >Label-TLV/object (no public comments to my
> >Hierarchical LSPs, draft-hummel-mpls-hierarchical-lsp-01.txt).
> >
> >However it is ok, to "steal" the deepest nested label and call it 
> >"VC-label" for indexing
> >some VRF at some remote PE. Furthermore it is ok, to "steal" the very same 
> >label for hashing.
> >
> >Independent from what is clean and what is quick&dirty, I cannot detect 
> >any sense in giving a hash-meaning to the bottom label at all.
> >
> >
> >
> >Heinrich Hummel
> >Siemens
> >
> >-----Ursprüngliche Nachricht-----
> >Von: Eric Rosen [mailto:erosen@cisco.com]
> >Gesendet: Dienstag, 2. Juli 2002 16:06
> >An: Shahram Davari
> >Cc: 'Eric Osborne'; George Sheng; scullptor@yahoo.com; mpls@UU.NET
> >Betreff: Re: your mail
> >
> >
> >Shahram> To do ECMP, you need to  assign hashed values to egress ports. 
> >This
> >Shahram> assignment is the hash state that I said you need.
> >
> >A hash is  a function that maps  some sequence of octets into  a sequence 
> >of
> >integers.  If you want to split  traffic over six paths, while ensuring 
> >that
> >all packets  with the  same bottom label  travel the  same path, you  need 
> >a
> >function that maps  a 20 bit quantity  into a number from 1-6.   There is 
> >no
> >need to maintain state of any kind.
> 
> 
> 
> 
> _________________________________________________________________
> Chat with friends online, try MSN Messenger: http://messenger.msn.com