>I would say that less than 7% was devoted to BGP when I took it. Halabi's
>book is excellent and if memory serves me correctly our own Mr.Howard
>Berkowitz was going to be writing a new BGP book, Howard is that still in
>the works?
Yes, with somewhat different focus than Halabi -- the working title
is "Building IP Service Provider Networks." As you might suspect, it
begins with the "What Problem are you trying to solve?" In many
respects, it's the mirror image of my last book, the "WAN Survival
Guide." WAN Survival Guide focuses on the choices the enterprise has
in the WAN, including multihoming, VPNs, and developing their SLA.
The new book deals with how the provider responds to these customer
requirements, and how it builds its infrastructure.
While there may be configuration snippets (which will include Juniper
as well as Cisco, and possibly others), it's not a how-to-configure
book. It's how-to-design and how-to-operate. For example, I will
spend much more time on the applications of communities (including
extended communities) than community commands.
Completion is scheduled for November, but it's flowing pretty quickly
and I hope to have it done before then. It would be nice if it could
be out late this year, but early 2002 is more likely.
Incidentally, I was just part of a panel on Internet routing system
scalability at the Internet Society INET meeting in Stockholm. Since
there has been confusion about the ISOC site, Sue Hares of Nexthop
put them on her site at www.nexthop.com. Nexthop makes the
commercial version of GateD.
You might get some insight from another document,
http://www.ietf.org/internet-drafts/draft-berkowitz-bgpcon-01.txt
on single-router BGP convergence. That version has coauthors from
Nortel (me), Cisco (Alvaro Retana), and NextHop (Sue Hares and Padma
Krishnaswamy). It's becoming a standards track document in the
Benchmarking Methodology Working Group, and will split into two
documents, one on terminology and one on the basic test. The new
versions will also have a Juniper coauthor (Marianne Lepp), so the
procedures should represent a pretty good consensus.
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=8887&t=8810
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
