UPDATED: Nirvana For Vendors? An NFV reading list…

Updated 22 March, 2013. Finding and linking to good articles about NFV and SDN. Updated with numbers 6-10 on 22 March, 2013.

I’m building up a list of content, articles and blog posts looking at NFV and SDN (yes, I know they’re not the same thing).

The aim is to point to articles that clarify, rather than muddy, the waters on these topics, even if they clearly come with a “vendor angle”. I think we’re grown-up enough to deal with that. Please send me your own finds, or add links in the comments.

1. Just before Mobile World Congress Ericsson’s Review carried this article on Service Provider SDN. It clearly lays out an architecture for service provider SDN, includes some appplication examples and also illustrates service chaining (as also mentioned by head of Ericsson IP Jan Haglund to TMN here). It’s a really good starting point for understanding service provider SDN.

2. Sticking with the major vendors, I thought this blog post from Cisco’s Sanjeev Mervana summed up both where NFV is post-MWC, and Cisco’s angle on it. Specifically of interest is the role of Cisco Quantum within Cisco ONE (Open Network Environment) and its optimisation, analytics and orchestration functions.

3 The good people at Light Reading started this NFV-related LinkedIn Group. It’s already had some really good posts from people in the industry. What I like about this group is that it contains both those questing for knowledge, and some expert commentators as well. A timely introduction.

4. Tom Nolle, President of CIMI Corp, publishes a blog that is never short of an opinion or two. I thought this post, which lays out the importance of giving the established vendors as little wriggle room as possible to offset the cost-saving benefits of NFV was quite revealing. Simply put, Nolle says that it the industry doesn’t nail down an NFV architecture, then vendors will end up “modifying the definition…to establish or protect commercial interests.”

5. I liked this one from 6Wind. It outlines clearly some of the use cases for NFV and makes the case for its own product. But we can forgive that. Interesting, I think, for insight into how good ideas can have deeper consequences than might at first appear. “All the concepts have to come back to reality … Today, standard virtualized architectures that work perfectly well for computing won’t be able to deliver the required network performance.”

What else? Please send me your recommendations for this list, or link to them in the comments, and I’ll add them here, tweet them out and all that sort of thing.
22-3-2013: Five more for the list

6. Camille Mendler , analyst at InformaTM recommended the work of one of her colleagues in this area, Dimitris Mavrakis . In this post, Dimitris takes a look at the main vendor SDN announcements at Mobile World Congress.

7. Sticking with Informa, but over the fence at their editorial operation, the journalists have been having a good look at SDN and are putting together a series of SDN-related interviews and features. Mike Hibberd talks to Alcatel-Lucent’s CTO , who introduces the word nirvana (we got there first, OK?) while James Middleton writes down the thoughts of Juniper Networks’ Mike Marcellin.

8. Light Reading’s Phil Harvey has a post on SDN from OFC/NFOEC 2013, where he gets the off-air views of Stanford University professor Nick McKeown. A sample quote: “A lot of the vendors that say they’re on board with SDN are only adding more code to support SDN-like interfaces and functions to specialized control planes inside of vertically integrated, proprietary routers and switches and gateways…”I’m feeling sorry for the service providers who have to wade through that garbage,” Also, it tees off a bit in the comments, always a sign that this ball is very much in play.

9. Here’s an interesting NFV proof of concept from BT, not really in mobile networks but good all the same to see this stuff being worked out with real world implications.

10. Finally, NSN’s Jane Rygaard pointed me towards this press release from NSN, that (revealingly??) mentions neither SDN nor NFV, but does claim proof of concept for core network virtualisation.