The MPLS-OPS Archive

Cell Relay Retreat>MPLS-OPS Archive>month:2002-Aug> msg00095



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Re: ISIS versus OSPF as IGP

  • From: fraanro <fraanro@arrakis.es>
  • Date: Thu, 15 Aug 2002 09:10:18 GMT
  • Cc: <mpls-ops@mplsrc.com>
  • Resent-Date: Thu, 15 Aug 2002 06:30:25 -0400
  • To: "alok" <alok.dube@apara.com>
  • X-MIME-Autoconverted: from quoted-printable to 8bit by host.secure4-hosting.net id g7F99XT12531

RSVP itself is independent of the IGP SPF. The RSVP process is going to 
ask the IGP process the best path according to a set of constraints 
(let us imagine only bandwidth). The the IGP is going to find the IGP 
best path based on a CSPF calculation. This path does not necessarily 
is the same that that IGP would calculate with the SPF (that is, 
without any constraint). 
RSVP itself is not going to see any TLV or direct information from the 
IGP, that information is going to be used as part of the CSPF 
calculation, and will have as input the constraints the RSVP connection 
is requesting. The result of that will be an explicit path that RSVP 
will use to signal the LSP request.

Javier.

----- Mensaje Original -----
Remitente: "alok" <alok.dube@apara.com>
Fecha: Miércoles, Agosto 14, 2002 2:04 pm
Asunto: Re: [MPLS-OPS]: ISIS versus OSPF as IGP

> if there are no CSPF constraints other than bandwdith demand 
> ....then.??wot i need to figure out is why cant RSVP be treated 
> independent of the
> IGP's SPF...
> are there  no TLVs to point out to RSVP that there is a better 
> path based on
> "desired constraints", even though its not the IGPs best path?
> rgds
> Alok
> 
> ----- Original Message -----
> From: fraanro <fraanro@arrakis.es>
> To: alok <alok.dube@apara.com>
> Cc: fraanro <fraanro@arrakis.es>; <mpls-ops@mplsrc.com>
> Sent: Wednesday, August 14, 2002 5:21 PM
> Subject: Re: [MPLS-OPS]: ISIS versus OSPF as IGP
> 
> 
> now, if the entire bandwidth between A and B on ab1 is used up,
> > will RSVP be
> > able to use ab2 ? will the results be same for OSPF and IS-IS?
> > please note im not talking about link failure, im talking about
> > using an
> > non-congested path....
> >
> > how will OSPF behave in this scenario? and how will ISIS behave? 
> when> coupled with RSVP??
> 
> Both will have the same behaviour. CSPF will remove from its
> computation any link that does not match the constraints (bandwith,
> color of the link, ..). That will happen regardless you are using ISIS
> or OSPF.
> 
> 
> ----- Mensaje Original -----
> Remitente: "alok" <alok.dube@apara.com>
> Fecha: Miércoles, Agosto 14, 2002 8:15 am
> Asunto: Re: [MPLS-OPS]: ISIS versus OSPF as IGP
> 
> > What do you mean with that? With both ISIS and OSPF you can do TE
> > basedon constraints like bandwith request,link coloring, or even
> > metric,that leads to paths not following the IGP best path.
> >
> >
> > ------>
> >
> > let me elaborate...
> > lets assume i have 2 links between routers A and B ..lets call
> > them link ab1
> > and ab2
> > ab1 has better bandwidth and  is the "chosen" path between A and B
> > by the
> > IGP.
> >
> > now im using RSVP to set up MPLS based LSPs and some of these LSPs
> > traverseacross A-B using ab1
> >
> > now, if the entire bandwidth between A and B on ab1 is used up,
> > will RSVP be
> > able to use ab2 ? will the results be same for OSPF and IS-IS?
> > please note im not talking about link failure, im talking about
> > using an
> > non-congested path....
> >
> > how will OSPF behave in this scenario? and how will ISIS behave? 
> when> coupled with RSVP??
> > rgds
> > Alok
> >
> >
> >
> 
> 
> 
> 
> 
> -------
> The MPLS-OPS Mailing List
> Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
> 

-------
The MPLS-OPS Mailing List
Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
Archive: http://www.mplsrc.com/mpls-ops_archive.shtml