Light Reading

Carriers Say SDN Won't Save Capex

Carol Wilson
10/30/2013
50%
50%
Repost This

SAN JOSE — Digital Disruption 2013 — Contrary to popular opinion, software-defined networking will not produce major capex savings for network operators, according to the two men responsible for virtualization strategy at CenturyLink and at Sprint.

Speaking on a Tuesday afternoon panel here on the impact of virtualization, James Feger, vice president, network strategy and development for CenturyLink, and Fred Feisullin, senior network architect in the CTO's office of Sprint, said there are advantages to deploying SDN and network functions virtualization (NFV), including getting new services to market quickly and, at some future point, opex savings. But capex savings isn't something they are expecting.

"When it comes to capex, I'd say it's a wash," Feger said. "Can I buy a virtual router today that is cheaper than buying a regular router? Sure. But transforming the telco environment into an SDN/NFV environment initially is going to cost more."

Telco central offices weren't built to house datacenter equipment and will need to be outfitted to do that, making the total cost of ownership (TCO) of a virtualized network about the same as today's capex budgets, he said.

In Sprint's case, the initial move to virtualization will take place in the evolved packet core, which isn't where most of the wireless network operator's costs lie, Feisullin explained.

"The core is a fraction of our capex, our costs are in the radio access network, mostly in radios, and those aren't going to be virtualized right now," he said. "The bigger gains are the new revenue sources that can be generated, then followed on by lower opex which will take much longer to be realized, then capex, maybe."

Like Feger, Feisullin sees higher opex associated with the initial deployment of new equipment to "get the infrastructure to work."

There may be more capex savings at the edge of the network and in CPE, such as set-top boxes, Feger said.

One further challenge in the early days of SDN/NFV is that creating virtual functions running on commercial off-the-shelf hardware creates complex operations challenges in the era when each function may be managed by a different siloed management system, Feger said.

New services acceleration and the ability to "fail fast" -- i.e., try something and then shut it down quickly if it doesn’t work -- are key advantages to a virtualized approach, according to both men. They balked at a suggestion from fellow panelist Jeff Edlund, CTO of CMS-Enterprise Solutions at Hewlett Packard, however, that network operators are willing to accept a lower level of overall reliability when they move more functions to COTS gear that is not built for the five nines of proprietary telecom boxes.

"It's not that we are tolerant of fewer nines or less reliability," Feger said. In a virtualized network, the reliability burden may be shared among multiple boxes that are logically connected rather being built into a more expensive individual component.

Service providers are willing to move ahead of standards development, however, because they see virtualization as important to future business models and may, in fact, be influencing standards by what they deploy and use, Feger said.

Feisullin sees "as many different strategies out there as there are service providers," when it comes to what to virtualize first and how much legacy gear continues to function for some time to come.

He also credited ETSI with getting the ball rolling on NFV, regardless of where its standards process ends up.

"ETSI is already a success because it has torn down walls between service providers and vendors and between vendors and that has shaved years the deployment cycles," Feisullin said. "Are we there yet? No, we have a long way to go."

— Carol Wilson, Editor-at-Large, Light Reading

(18)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View
Page 1 / 2   >   >>
TomNolle
50%
50%
TomNolle,
User Rank: Light Sabre
11/22/2013 | 9:24:11 AM
Re: Service providers may not benefit that much
I agree; there is a major risk that application awareness becomes stateful behavior that we already know doesn't scale in a cost-efficient way, particularly wrt opex.  I think that the SDN community will need to prove in an operations strategy, and to do that they'll also have to prove in a holistic approach to SDN--is it a complete strategy or a limited-area solution that still has to fit inside a bigger picture like IP.  If the former, how do we make it scale.  If the latter, how can it make enough of a difference to matter?
varkonyib
50%
50%
varkonyib,
User Rank: Light Beer
10/31/2013 | 4:18:26 AM
Service providers may not benefit that much
Centralization we had already with PDH/SDH, or ATM. Centralization of PNNI has failed miserably, although we still have some of those ATM switches in operations... :-)

The real issue would be a standard flow-through provisioning. But this is still far away. A standard interface between the switch and the controller does not help too much, when for each vendor's controller we need to develop to a different API.

At the end of the day all comes down to prices and technology architectures are secondary. SDH could not conquer the world because of pricing, not because it could not solve all technical problems. Recently, it has provided fully dynamical bandwidth control and common provisioning with other layers.

The real motivation for carriers is in OPEX savings in most cases. Managers are typically measured by EBITDA, and rarely on cash-flow. So CAPEX savings are virtual, and controllers never accept it, because it is too easy to falsify.

 

 

 

 
TomNolle
50%
50%
TomNolle,
User Rank: Light Sabre
10/30/2013 | 6:05:41 PM
Re: And the operators say...
I would hope that the IETF group thinks about some of the deeper issues (beyond simply deploying VNFs) but I can't say whether that will be what comes out.  If you look at service chaining, or any NFV element, it's kind of two-dimensional.  One dimension is how you deploy it and manage it, and the other is how the functionality itself has to be structured in order to achieve your service quality and availability goals.  It's not clear to me whether the IETF is fully engaged on either point but I think they do have a more general model for deployment and management than the NFV ISG does (because they have a broader scope of interest).
Dredgie
100%
0%
Dredgie,
User Rank: Light Sabre
10/30/2013 | 5:38:27 PM
Re: Vendor messaging
I can't talk for data center SDN, but suppliers should absolutely talk capex reductions when it comes to carrier SDN (as described superbly by dwx*). With on-demand provisioning, you can dramatically increase path efficiencies and reduce oversubscription. Throw-in an additional low-end edge layer – a whitebox MPLS switch – you can even reduce the overall traffic hitting costly LERs. All while keeping your core optics and switches unchanged.

 

*Full disclosure: I have no idea who dwx is – but adding to my xmas card list!**

**Fuller disclosure: I don't actually have an xmas card list.
DOShea
50%
50%
DOShea,
User Rank: Blogger
10/30/2013 | 4:45:03 PM
Vendor messaging
I think there is at least some vendor messaging out there that mentions capex savings as a benefit, though I have never seen it mentioned prominently. Maybe some companies are hard-wired to list capex savings as a benefit of anything they do.
sam masud
50%
50%
sam masud,
User Rank: Light Sabre
10/30/2013 | 3:37:28 PM
Too early
With all due respect, I don't think we can predict whether SDN will produce major or moderate savings in capex (and/or in opex) for the simple reason that we're much too early in terms of adoption of SDN. There will be a shakeout/consolidation in this market and vendor pricing models will change as the market matures. Also, I would not be surprised if opex savings from SDN aren't as expected because a more dynamic network that supports an increasing number of apps and let's service providers tailor services on a per user basis could impact opex.
dwx
50%
50%
dwx,
User Rank: Lightning
10/30/2013 | 3:30:19 PM
Re: That's their story and they're sticking to it
But Google is not using or have any plans on using open flow on their much larger backbone network. They are looking at hybrid central/distributed control using a PCE. The reality though is carriers have been using things like RSVP-TE to run links at 80-90 percent for 10+ years now on high capacity backbone networks. Where SDN may help those networks is at the edge and finding ways to control the spigots in and out of the network traditionally hard to control. Or optical+IP integration to turn up circuits dynamically or collapse layers. But you still need all the infrastructure in place to do those things, so capex isn't lowered dramatically. SDN adds some tweaks to that but the overall network doesn't change and the economics of using cheaper hardware in core networks doesn't really pan out. Like was said you are better off issuing a RFP and beating vendors up for better pricing.
gleavieboy
100%
0%
gleavieboy,
User Rank: Moderator
10/30/2013 | 3:03:32 PM
Re: That's their story and they're sticking to it
Good post Carol and kudos to the panelists.  But this shouldn't be a shocker.  ONF's original whitepaper on "SDN: The new norm for networks" doesn't sell Cap-Ex as the driver.  It's all about centralizing management and control, improving automation, increasing programmability, speeding innovation and offering more granular network control -> needs driven by today's east-west traffic patterns, big data, access from any device anywere etc...

Savvy operators also understand that SDN creates the flexible network architectures to support NFV (which in turn brings its own series of benefits). CFOs should be looking at SDN/NFV as a way of designing their networks for more contemporary needs - and establishing the elastic, programmable, scalable tenets required for future prosperity and competitiveness.  For those a tad more blinkered, there are plenty of vendors out there that will help them wring a few extra dollars out of cap-ex when doing like for like box replacements or upgrades. But work those deals now -because the vendor landscape for the new global network is going to look very different.
Dredgie
100%
0%
Dredgie,
User Rank: Light Sabre
10/30/2013 | 2:59:47 PM
Re: And the operators say...
Re. your last point, Tom: Hence, I assume, the energy that is being put behind the Service Function Chaining (Network Service Chaining as was) initiative in the IETF. Momentum that surprised even the Area Chairs. 16 drafts in progress with only one BoF behind them. One more BoF at IETF88 (next week) before the charter is approved and working group status is achieved - but if the meeting was a few week later, even that might have been negated.
TomNolle
100%
0%
TomNolle,
User Rank: Light Sabre
10/30/2013 | 2:08:26 PM
And the operators say...
...at least at the meetings I've recently had, that capex reduction is not the compelling driver of either SDN or NFV.  They say operations savings and service velocity, which is what they said in the panel Carol is quoting.  The problem is that operators spend only about 17 cents of every revenue dollar on capex lf all kinds, and neither SDN nor NFV could hope to impact ALL of that.  As one operator said, they could bring about a 20% reduction in equipment spending by beating up Huawei.

The thing is, service velocity is as facile an answer as capex.  Velocity without direction isn't progress it's just movement.  Nothing is going to make bits profitable no matter how fast you can shuffle them or arrange them to support new stuff.  The "new stuff" has to be created above the network of today.  That means that a big part of "service velocity" has to come from assembling features into new services faster and more effectively.  That's not an SDN story, and it may not even be an NFV story.  If NFV is just about virtualizing existing network functions then it's not about creating truly new services.
Page 1 / 2   >   >>
Flash Poll
LRTV Documentaries
Cable Eyes Big Technology Shifts

4|16|14   |   03:02   |   (3) comments


US cable engineers are facing a lot of heavy lifting in the coming years, notes Light Reading Cable/Video Practice Leader Alan Breznick.
LRTV Custom TV
Maximizing Customer Experience & Assuring Service Delivery in an IP World

4|15|14   |   4:57   |   (0) comments


Steven Shalita, VP of Marketing, NetScout Systems, Inc., discusses the challenges cable/MSO operators face in assuring the delivery of new IP-based services. Key points include the value of proactively managing performance, and using rich analytics and operational intelligence to better understand service and usage trends, make smarter business decisions and ...
LRTV Documentaries
Bye-Bye DVD: Consumers Embrace Digital Video

4|10|14   |   04:17   |   (6) comments


Veteran video analyst Colin Dixon, founder and principal analyst of nScreenMedia, says research shows 56% are using digital video already.
LRTV Documentaries
Video: TW Cable Puts Multicast Gateways to the Test

4|8|14   |   04:13   |   (1) comment


Tom Gonder, a chief architect at Time Warner Cable, explains how its trial of multicast gateways is impacting IP-based video plans.
LRTV Custom TV
Managing & Monetizing Big Data in Operator Environments

4|7|14   |     |   (1) comment


At Mobile World Congress, Gigamon's Director of Service Provider Solutions, Andy Huckridge, and Heavy Reading Analyst Sarah Wallace discuss the 'big data' issues facing carriers and operators today.
LRTV Huawei Video Resource Center
Data Center Energy – Build Your Data Center in a Modular Way

4|7|14   |   2:13   |   (0) comments


Dr. Fang Liangzhou, VP Network Energy Product Line, shared his thoughts about the challenges for data centers during CeBIT 2014.
LRTV Huawei Video Resource Center
Agile Network Solution – An Overview of Huawei's Agile Network Solution

4|7|14   |   2:31   |   (0) comments


Ajay Gupta, Director of Product Marketing, Networking Product Line, gives an overview of the Agile Network Solutions during CeBIT 2014.
LRTV Huawei Video Resource Center
Huawei’s eLTE Voice Trunking, Video and Data Applied for Railways

4|7|14   |   1:38   |   (0) comments


Gottfried Winter is the Sales Director at Funkwerk, a German specialist in GSM-r terminals and a long-time partner of Huawei. At CeBIT 2014, Winter talks to Light Reading about this partnership and the integration of enhanced voice trunking, video and data functions.
LRTV Huawei Video Resource Center
LeaseWeb Speaks Highly of Huawei's Datacenter Products

4|7|14   |   1:37   |   (0) comments


Rene Olde Olthof, Operations Director LeaseWeb, talks about the next data center transformation during CeBIT 2014.
LRTV Documentaries
Comcast: Reshaping the Cable Network Architecture

4|3|14   |   07:11   |   (8) comments


Shamim Akhtar, Comcast's architect and senior director of network strategy, explains why the cable company is moving to a more distributed network architecture.
LRTV Custom TV
VMware CEO Pat Gelsinger at Mobile World Congress

4|1|14   |   3:41   |   (0) comments


VMware CEO Pat Gelsinger speaks to Heavy Reading about the value of virtualization spanning from the data center to service provider networks to mobile devices.
LRTV Huawei Video Resource Center
Analysts' Impressions of Huawei SoftCOM at ONS 2014

4|1|14   |   1:11   |   (0) comments


After visiting the Huawei booth at ONS, Lee Doyle of Doyle Research gives his appraisal of Huawei's SoftCOM solution.
Hot Topics
BlackBerry's Chen: We're Not Dumping Devices
Dan Jones, Mobile Editor, 4/10/2014
AT&T's Going to Carolina With 1 Gig
Sarah Reedy, Senior Editor, 4/10/2014
Cisco, Juniper Treating Gear Against Potential Heartbleed
Dan O'Shea, Managing Editor, 4/11/2014
Cisco & VMware Are Apple & Google of SDN
Mitch Wagner, West Coast Bureau Chief, Light Reading, 4/14/2014
T-Mobile Petitions Operators to Kill Overages
Sarah Reedy, Senior Editor, 4/14/2014
Like Us on Facebook
Twitter Feed