At 10:29 PM -0400 5/27/02, nrf wrote: >Nothing much better than the RFC's. Cisco has some textbooks, but in my >opinion they really aren't that good. The MPLS/VPN book, for example, is >OK if you really really want to do VPN's, but even then it still isn't that >informative. They also have a frustratingly Cisco-centric approach to MPLS >(yeah yeah, I know what you're going to say, what do you expect if it's from >Ciscopress, but still...). > > >From what I've seen, MPLS gets a lot clearer when you can actually see it in >action. If you guys are implementing an MPLS network somewhere (and I know >quite a few ISP's are doing just that), then that would be the best place to >learn it.
Just as it takes a certain change of mindset to understand what problem BGP is solving, it takes a change of mindset to understand MPLS and the next generation, GMPLS. Many people have problems with the ideas that MPLS does not do away with the needs for routing protocols, and indeed adds new protocols. Also, there tends to be an assumption that MPLS is faster than conventional IP routing, which no longer is the case. There are real values to MPLS, which are more carrier than enterprise oriented. These include traffic engineering/QoS, and extremely flexible failure override techniques that can make link recovery times comparable to SONET. You won't find any Cisco (or other) configuration examples in my new book, but you will see an extensive discussion of functionality--why would you want to use MPLS, and how it changes design approaches (e.g., perhaps no IGP in a large network core). It's a sufficiently rapidly developing field that books will have trouble keeping up with it. The RFCs _and_ Internet-Drafts (see http://www.ietf.org/html.charters/mpls-charter.html ) are still the best references. Don't worry too much about CR-LDP, but pay attention to the traffic engineering and link restoral documents. If you like PowerPoint tutorials, browse around www.nanog.org, and you'll find several good presentations at the last few meetings. > > >""Scott H."" wrote in message >[EMAIL PROTECTED]">news:[EMAIL PROTECTED]... >> Any recommendations for good sources on MPLS? >> >> ""Howard C. Berkowitz"" wrote in message >> [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... >> > At 8:58 PM -0400 5/27/02, Scott H. wrote: >> > >I recently completed the CCIE lab exam (R&S) and have been >contemplating >> the >> > >next mountain. Anybody have any suggestions? I have considered both >> > >Juniper and Nortel but my interest level in those is limited and so is >my >> > >access to equipment. Anybody gone down the road of CCIP yet? Metro >> seems >> > >like a fun topic and I would love to learn MPLS. Any comments (except >> B.S. >> > >cert bashing) welcome. >> > > >> > >Scott >> > >CCIE #9340 >> > >> > I'd encourage studying MPLS, if for no other reason that to >> > understand it, you have to learn new paradigms, which will indirectly >> > improve your knowledge of IP routing. For that matter, it's a key >> > technology for Juniper, Nortel, and metro services. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=45229&t=45176 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
