NFV Just Made OSS Hot

It's hard to imagine that anyone would ever say "OSS is sexy," but right now it's hard to identify any issue that's more critical to the future of communications networking than management and network orchestration (MANO) for NFV.

I've been loitering with intent to understand the OSS sector for the past 15 years, attending the annual TM Forum events in Nice and Dublin and engaging with the various developers of OSS and BSS systems (large and small). In that time I've watched the sector blossom from a specialist niche -- "It's the back office… nothing to do with us" -- inhabited by specialists to one that is now front and center in the transformation of networking strategies, populated by the biggest names in telecom and (tellingly) IT, and at the heart of service providers' customer experience management (CEM) efforts.

For an example of the importance of OSS to CEM stategies, take today's announcement about how Etisalat Egypt is going to target resources and investments in its mobile network based on the information generated by a network planning and optimization system sourced from JDSU (Nasdaq: JDSU; Toronto: JDU). (See Etisalat Egypt Deploys ariesoGEO.)

What is weighing even heavier than ongoing CEM efforts is quite how network operators, so used to managing static networks with traditional OSS tools and a team of telco engineers, are going to control dynamic networks that are a broader mix of telco and IT systems, and which incorporate some virtualized functions. It's a different ballgame: Telco operations teams will essentially be tasked with running an ever-shifting connected cloud infrastructure. And some of them won't want to.

The challenge is enormous -- that's why alarm bells are ringing about MANO issues. Here, in essence, is the current state of affairs, as stated in Carol Wilson's excellent recent article, Analysts Warn of Major NFV Gaps:

    Caroline Chappell, principal analyst, Cloud and NFV, at Heavy Reading, and Tom Nolle, president and founder, CIMI Corp., both say there still is no clear definition of what orchestration functionality virtual network functions (VNFs) need to run within the network nor any clear assignment of management duties among the different layers of the NFV architecture. Without these key elements defined, network operators who want to deploy NFV will be reliant on vendor-proprietary solutions to plug those gaps.

And proprietary solutions, along with silos, are not what this industry needs, as stated today in a guest column on Light Reading. (See Let's Not Kill SDN & NFV With Silos.)

Need to know more about the management of network assets and applications in an SDN and NFV environment? Then check out the agenda for OSS in the Era of SDN & NFV: Evolution vs. Revolution, November 5, at the Thistle Marble Arch Hotel, London.

A good place to start, then, is to figure out what needs to be addressed, and learn from some of the companies that have started that process and, in some cases, are already learning from some early real-world experiences. (See Colt Preps Next-Gen OSS for NFV, SDN and Meet the Phantom Network for NFV.)

That's why the upcoming OSS in the Era of SDN & NFV: Evolution vs. Revolution one-day conference in London (November 6) -- hosted by the aforementioned Caroline Chappell from Heavy Reading and featuring key presentations from the likes of BT Group plc (NYSE: BT; London: BTA) and Colt Technology Services Group Ltd (which is unveiling its next-gen OSS strategy at the event) -- is going to require a great deal of focus and attention.

OSS is hot -- it's official. Don't let anyone tell you otherwise.

— Ray Le Maistre, Circle me on Google+Follow me on TwitterVisit my LinkedIn profile, Editor-in-Chief, Light Reading

patricknmoore 10/29/2014 | 2:27:55 PM
Re: OSS Great question on if the Traditional OSS sector is capable.  I, personally, have my doubts.  They have so much invested in the old, static methods of managing everything that I think their play will be to try and evolve current solutions.  The problem is current solutions have a core that is not very adaptable in most cases.  Many solutions floating around out there still have core technologies that are pre-web, even though they do an admirable job of hiding things like Powerbuilder and such that remain by deploying on thin client environments like Citirx to make it seem "web based".

I think the answer is that existing solutions live on as long as legacy networks are around, which will be quite a while longer.  New solutions, possibly from legacy vendors IF they realize they need to start from scratch.  To make it palatable to the customer base this can't be a forklift situation where new solution means the old goes away, at least not right away.  I think there has to be a transition plan that includes:
  • Implementation of Agile OSS in parallel with SDN/NFV element introduction - this must include solid abstraction of the network, including discovery equivalence, into the OSS that is as close to real time as possible
  • Federation with legacy systems that won't go away anytime soon - some of these legacy systems may live on for a long time and be relegated to only managing the physical aspects of the network that will be the supporting infrastructure for NFV in the future
  • The idea of a Network API that gives all OSS a single point of contact with the network, and talks directly to the network using things like OpenFlow and NETCONF instead of mixing in the EMSs and CLI interfaces to the NEs.
  • Orchestration that understands NFV and doesn't try to apply old-school methods to managing those new-school elements.
  • Blends IT centric and Network centric methodologies, meaning you are selling a new paradigm to the customers, into a new way of managing the network.

This is early thought on my part, but something my company is seriously looking at right now so we can guide our clients appropriately.
[email protected] 10/29/2014 | 2:00:13 PM
Re: OSS Absolutely right -- so the big question is, where does this new OSS come from?

Is the traditional OSS sector capable of delivering such capabilities?
patricknmoore 10/29/2014 | 1:01:21 PM
OSS A key thing to remember, in addition to the need to have strong and consistent links between OSS and BSS, is that traditional OSS is going to have to evolve as the network does.  This doesn't mean just adding fields to a database to accomodate the new SDN and NFV environments. OSS is going to have to become agile and dynamic, just as the networks are evolving to be.  Otherwise the all too common issue today of the data in your OSS not being in sync with your network is going to become even more of an issue.  For example, your inventory can't be a static thing when your network become elastic and extremely dynamic in operation.  OSS may be hot, but Real Time OSS is going super-nova...or should be.
RitchBlasi 10/28/2014 | 4:24:57 PM
OSS/BSS OSS will never be sexy, but yes, combined with BSS, it will be a marriage made in heaven as carriers begin to migrate away from boxes to software. Itz important that both are linked so that the services offered can then be billed to customers and also have a mechanisim for potentialy sharing revenue on the back end. It aint sexy, but complicated and something carriers need to move on to get beyond the pricing wars used for differentiation.
Sign In