What do you think?
Rate this book
MPLS-Enabled Applications : Emerging Developments and New Technologies by Ina Minei, Julian Lucek published by Wiley-Blackwell
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 2 of 2 reviews
November 6, 2011
I dabble in MPLS design and troubleshooting, but I didn't have much theoretical understanding of the reasons why MPLS networks were built the way they were. MPLS-enabled applications came to my attention when I was trying to figure out how point-to-multipoint LSPs would work in the context of IP multicast packet delivery, and it certainly delivered a solid theoretical basis for them.
I was surprised at how strongly the authors come down in favor of RSVP - LDP really comes across as an ugly hack. I was also surprised that while the book had lots of mentions of "some vendors do X and some do y", it didn't say which vendors did which, nor did it discuss the practical maturity of the respective approaches. While the RSVP bias makes sense the way they present it (it certainly comes across as a lot more powerful - LDP's only advantage that comes across is simplicity and IGP synchronization, while RSVP has TE and FRR capabilities. It would have been great to have seen "the rsvp implementation in JunOS works the way we discussed for the last N years, and Cisco introduced in in XR versions blah blah blah" or vice versa. Examples of configuration snippets in JunOS, IOS, IOS-XR or others would have been tremendously useful, and it is unfortunate that those were not included.
They also come down strongly in favor of the route-reflector+BGP autodiscovery approach to vpn signaling, and the force of their argument is very strong - it changes the work involved with configuration and deployment of a new PE, feature, or VPN from O(N^2) to O(N) - that is, a new PE only needs the RR configuration to be updated, instead of requiring that every other PE also be configured.
All in all, this was a great book on MPLS theory, but without much practical implementation detail.
I was surprised at how strongly the authors come down in favor of RSVP - LDP really comes across as an ugly hack. I was also surprised that while the book had lots of mentions of "some vendors do X and some do y", it didn't say which vendors did which, nor did it discuss the practical maturity of the respective approaches. While the RSVP bias makes sense the way they present it (it certainly comes across as a lot more powerful - LDP's only advantage that comes across is simplicity and IGP synchronization, while RSVP has TE and FRR capabilities. It would have been great to have seen "the rsvp implementation in JunOS works the way we discussed for the last N years, and Cisco introduced in in XR versions blah blah blah" or vice versa. Examples of configuration snippets in JunOS, IOS, IOS-XR or others would have been tremendously useful, and it is unfortunate that those were not included.
They also come down strongly in favor of the route-reflector+BGP autodiscovery approach to vpn signaling, and the force of their argument is very strong - it changes the work involved with configuration and deployment of a new PE, feature, or VPN from O(N^2) to O(N) - that is, a new PE only needs the RR configuration to be updated, instead of requiring that every other PE also be configured.
All in all, this was a great book on MPLS theory, but without much practical implementation detail.
September 11, 2016
This plus MPLS Fundamentals from Cisco Press and you'll be a MPLS expert.
Displaying 1 - 2 of 2 reviews