& cplSiteName &

Bolder Is Better for IP Transformation

Carol Wilson
11/21/2013

Telcos that try to move incrementally into an IP network transformation may have a harder time proving the business case than those that are willing to make bolder moves and combine the transition from TDM to IP with a major upgrade to broadband access and the introduction of new services, according to a vendor closely involved in the process.

Metaswitch Networks has been helping telcos of all sizes move into the VoIP space and explore its revenue potential for some time. The company, now known for a lot more, first made its mark in the US in that area. (See: Metaswitch Formally Joins CloudNFV Effort.)

The one thing that strikes me about my recent conversations with Metaswitch folks and its announcement of ongoing support for telco clients' legacy Class 5 central office switches is that there is no one-size-fits-all approach to how this is done. Each telco is facing its own challenges and opportunities in taking an inevitable step. (See: Metaswitch Offers Legacy Switch Support.)

In fairness, these conversations began because Metaswitch was eager to shoot down claims by Genband Inc. that the business case for IP transformation could be made almost solely on the reduction in commercial power costs from replacing old gear. (See: Genband Plots Funding of TDM Death March.) But the benefit extend much farther, and Kelsyn Rooks, director of product marketing for IP transformation at Metaswitch, shared the overriding conclusions with us.

First, he says, "the overall business case is very specific to the individual carrier." Also, the carriers that have had the greatest success have been willing to adopt a bolder vision of transformation. Not only have they replaced TDM switches, but they've also adopted high-speed broadband and revenue-generating services such as IPTV. That's why Metaswitch believes being able to assemble the business case for IP transformation requires a keen understanding of the carrier's entire cost structure. "It becomes a matter of putting together multiple building blocks."

On the power front, few of the telcos that have worked with Metaswitch can justify the cost of going all-IP based on power savings alone. Those savings vary widely depending on where the carrier is located -- i.e., the hostility of the local environment and the price of commercial power, which also varies -- and even the extent to which the telco has modernized its broadband access facilities, which also consume power.

But, wait, there's more. Many carriers won't buy into a transformation project unless they can see a payback in the first two or three years. A power payoff can take up to 15 years -- if it comes at all. "If you are looking at that kind of timeframe, you need to start thinking about whether you want to start replacing the old gear with hardware-based appliances or, in a 10-year time frame, are we going to be talking about a software-based network?" The risk of launching a project with a 10-year payback is winding up with something quite outdated.

The bad news from the vendor perspective is that there is no single business case or approach that can be adopted everywhere, so vendors have to adopt a consultative approach that delivers some degree of customization. The good news is that many carriers can find savings and benefits somewhere. One telco actually made its transformation business case solely by collapsing central office sites and eliminating multiple leases.

"There is no one size fits all" on power savings or anything else, Rooks said. "You have to take it carrier by carrier."

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

(18)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Page 1 / 2   >   >>
Kevin Mitchell
Kevin Mitchell
11/25/2013 | 5:08:04 PM
bolder vision: ditch the network, embrace the cloud
This all still incremental and represents creepingly slow and small steps. A bolder path forward is ditching the voice network alltogether and outsourcing the voice business from the cloud. Fix the busienss case through success-based SaaS. Free the organization and capex to nail some high RoI projects. Be the broadband and mobility expert and still deliver voice services, just without buying and running a voice network (IP or TDM).

Some further thoughts on this:

Next Generation Voice Platform – Leaping From Network to Cloud

SaaS – Freedom from Product Release Tyranny


 

 
gleavieboy
gleavieboy
11/24/2013 | 8:08:46 PM
Re: Risk factor
I'm actually complimenting you on this thread, Seven:) [and on others, too!!]. I agree with your position that not every TDM box gets replaced with an IP equivalent.  I think that the "network" will absolutely collapse into multi-service elements; in fact the true "software telco" will use common IT platforms, hypervisors, DNS, load-balancers, the lot. We're some ways away from that, and telcos will take different paths to get there - but it's a software-centric future, whichever way you choose to build, or resource it. It's this array of options that has got me truly excited about the next 5 years; and the prospect of seeing you on a panel with Martin Taylor would be a highlight for sure.  Let me know if I can help make that happen.

G
brookseven
brookseven
11/24/2013 | 10:49:05 AM
Re: Risk factor
gleavie,

Next time I am on a panel with Martin Taylor, we can laugh about your commentary on me.

I am trying to point out something simple:  There is a presumption that Class 5 switches will be replaced by IP versions of the same thing.  Suppose that plan (in bulk) does not happen?

Does that mean that there is no market or nobody buys a new Class 5.  No.  What I am trying to do is push things to the logical extremes so that people see the flaws in their thinking.

Consider:  Consumer wireless dominates and they drive the consumer bits of the current class 5 into the ground.

Business goes to either Hosted IP PBX or Runs their own IP PBX.  What is really required of a Class 5 infrastructure?  If you can locate the IP address on the other end, then really why does a class 5 switch exist at all?

So, I think that is what people ought to think about in this IP transformation world.  I have an infrastructure that already routes IP Packets.  I have a device that can ring, do Class Services, etc.  

I am not kidding about the wireless ACD.  We used a SaaS ACD service that linked our customer service and inside sales guys.  Instead of a desk phone, we just gave them a headset for their cell phones.  They logged into the system via their phones and poof - cell phone based ACD.  What is missing is attendant function and 4 digit calling and a few other odds and ends.  Now since we paid for a unlimited calling wireless plan for folks (BYOD), you can imagine how we might cut all wireline phone service.  Given the price of IP desk phones, this is a huge capex savings.  

If you think about it for about 10 minutes, you will see how  to offer to wireless phones all the features of a desk phone linked by a SaaS service.  Once somebody does that, then imagine that you have a virtual communications service (heck throw in IM) that reaches everyone no matter where they are.

So not only be bolder....be even bolder than that.

seven

 
gleavieboy
gleavieboy
11/23/2013 | 9:35:57 PM
Re: Risk factor
I know that we all enjoy Seven's role as the pantomine villain on these boards :) but i tend to agree with a lot of his points here (and that's from a Metaswitch guy)!  But then if this was a telecom version of ESPN's "Around the Horn" I think I'd also be awarding points to pretty much everyone in this thread.  Yes, many of the existing switches will rot in the ground, never to be replaced.  Even if the big Telcos set about swapping them all out now for NGN or IMS, we'd all be in the ground ourselves long before it was complete.  And for sure, wireless is an inexorable trend for consumers at least; and the path that the biggest Telcos are hell bent on following. They'll be under threat too though, as a new breed of "post-carrier operator" enters the ring.

I'm sure we'll see viable wireless solutions for business too, but right now it's more the domain of wireline and a number of our customers are leveraging that opportunity to also reduce their reliance on traditional, government-subsidized voice services.  That helps kick start plateauing revenue streams for sure, but long-term, the cost basis for network infrastructure must also be reset. That will best be achieved by running core network functions on private or public clouds, and leveraging the web-scale, cost benefits of open-source code (see http://www.projectclearwater.org for example).  So yeah, there are valid business cases to be considered for going from Class 5 to NGN, but depending on strategy, the correct jump might yet be straight to NFV (network functions virtualization). Luckily for the airlines, there is no one size fits all solution - which means the vendor sales teams will be collecting airmiles and criss-crossing the planet for the foreseeable future.
brookseven
brookseven
11/22/2013 | 8:58:30 AM
Re: Risk factor
spc,

Yes - nobody in business ever uses a cell phone.  Thats right - keep thinking that.

And yes the largest carriers are the special case.  The ones with the large capex budgets.

And yes, you can make an ACD out of cell phones.  Just because you have not figured out how does not make it so.  There are SaaS providers that do this today.

seven

 

 
spc_isdnip
spc_isdnip
11/22/2013 | 8:46:09 AM
Re: Risk factor
Seven, it's not reasonable to argue strategy when your assumptions are so far off base.  Maybe you're happy talking on a small square battery-powered box whose principal design function is to watch cat videos and whose user interface, for telehone calls, is about as user-friendly as Joseph Kony.  Maybe you're happy with the craptastic sound of mobile networks, which keep pushing voice compression downwards towards unintelligibility.  Maybe you don't realize that a large number of telephone calls are answered by people whose jobs involve answering the phone all day, who wear headphones connected to specialized call distribution systems, whose own lines can't even be dialed directly by the public.  Cell phones don't do that.

So while a substantial share of the residential market has stopped using land lines (not all), the more profitable business phone market is still primarily wired.  Cell phones cal wireline phones very often.  That isn't going to change.  You can't extrapolate every trend to zero -- wireline will decline somewhat more, but not to zero. And the wireline plant will need to be upgraded too, as the copper is in dreadful shape after two decades of price cap disinvestment.  Thus cable is picking up more of the wireline phone market, even business lines, and they are not in the wireless business.  Business is getting fiber in fits and starts, often from MSOs or CLECs.

VZ and ATT, while large, are the special case. They have wireless coverage across their LEC footprints.  They might be able to make deals to run wireline SIP across their IMS clusters, though the mobile switches are in the wrong places and operate quite differently.  So they can beg regulators to let them out of their Carrier of Last Resort obligations -- the ones they accepted as part of the price cap deal -- and try to dump their "chronic" customers onto cellular.  It's not always a good substitute -- witness the pushback from Fire Island. And have fun paying your mobile bill if you decide to watch Netflix movies on your HDTV streamed across that wireless connection.
brookseven
brookseven
11/22/2013 | 1:30:36 AM
Re: Risk factor
spc,

Pretend you are AT&T.

You have Class 5 voice switches that have TDM and you have wireless TDM voice switches.  You know that about the time that you get rid of the switch you buy today ALL wireline phone lines will be gone.  Now before you protest, remember that is 30 years from now.  If you really believe there will be a significant number of land lines in 2050 for voice, then you are kidding yourself.

For the wireline network, you can extend the life of the switches by slowing retiring them.  This is the choice that AT&T and Verizon are both doing.

But wait, they have to replace their wireless voice switches.  Since they know that really in the long term that is ALL they will have they buy them big enough to take on that small load that will over the long term come from the wireline network.  This will be less than the continued increase in cell phones, so will be lost in the noise in the increase in wireless "line" count.  So they are ALREADY paying for all the RTUs anyway.  Additional cost = 0.

And what I would be asking a business in the long term is:  Why are you issuing folks 2 phones?  Why are you not finding a way to use their cell phone for everything?  So, yes I think the wireline business PBX market - in the long term - is dead.

I also don't care what anybody above the age of 18 says about their landline.  I have already provided that and the equipment exists.  There are NO net new lines for older folks so all I have to do is keep the ones I have in production.  I can do that for another 10 - 20 years without buying a replacement switch.

This all used to be called Wireline-Wireless convergence.  You are missing the point that the Tier 1s are BOTH kinds of carriers.  All they have to do is 30 years from now add your 1 phone line to their wireless switch and it will cost them exactly 0 extra dollars to do so.

seven
spc_isdnip
spc_isdnip
11/21/2013 | 6:25:53 PM
Re: Risk factor
Seven, by other infrastruture, I presume you mean the wireless switch, not to tell businesses that their PBX trunks should be replaced by VoiceLink cell phone boxes?

A problem with using one switch for both is that you're likely to have a higher RTU license cost on the more elaborate wireless switch.  The price of a switch nowadays is mostly RTUs, after all -- a Metaswitch is basically a couple of Dell servers with their software, plus an overpriced media gateway that you can substitute an Audiocodes for.  I don't think they really have a mobile product, though they have IMS, but their overall price for a wireline switch is a lot less than traditional vendors', and that probably applies to RTUs.

And remember there are mobile carriers who aren't the LEC in region, and LECs who aren't in the mobile business.  And there are different rules about rating and routng calls.

I'm not giving up my (cable) wireline phones just to talk on a mobile all the time.
brookseven
brookseven
11/21/2013 | 5:30:35 PM
Re: Risk factor
spc,

And you can then buy a wireless switch, which you are buying ANYWAY and use it for wireline....

Or you can buy the wireless switch and then buy a wireline switch....

 

And that IMS switch can also just do SIP.

Given that there are a lot more cell phones than wireline and wireline is declining...then why bother buy a wireline VoIP switch.  Keep the 5E running until it dies and then move everyone to your other infrastructure.

See the point now?

seven

 
spc_isdnip
spc_isdnip
11/21/2013 | 4:48:49 PM
Re: Risk factor
Seven, it is not true that a wireline and wireless voice switch are the same thing.  If you don't believe me, first price out a nice wireline switch, like say a Metaswitch or Taqua.  Then price it out fully decked out for wireless use, just for the terrestrial portion.  Guess what happens to the price.  Wireless adds all sorts of hair to the mix, from handoff to roaming to aGPS-based E911.  And the nitwits at GSMA decided that VoLTE should not be simple VoIP over an assured wireless channel, but IMS, the Rube Goldberg approach to switching.  Hence the slow delivery and high cost.  While some wireline providers do follow IMS (like PacketCable v2), it's not necessary.

So we end up with a wireline approach that's basically SIP with maybe interfaces to TDM gateways, and a wireless approach that's got SIP buried nside IMS, buried inside all of the other requirements that mobility entails.  You wouldn't want to waste costly wireless capacicty on wireline, which doesn't need it, nor could wireline gear suffice for the wireless standards.

I think VoLTE should have been simpler...
Page 1 / 2   >   >>
More Blogs from Rewired
Two years after first announcing my retirement plans, I'm really doing it. No, really.
There was surprising optimism in the sector at the MEF18 event, but will it be reflected in revenues going forward?
Many larger network operators are bringing software development in-house, developing their own orchestrators and other tools, and using more open source to get the results they want, without vendor dependence.
Here's a quick snapshot of my impressions from this week's NFV-Carrier SDN: Automation and Monetization event in Denver.
The FCC lies about getting hacked and net neutrality proponents cheer when Alex Jones is censored, but so what?
Featured Video
Upcoming Live Events
November 5, 2019, London, England
November 7, 2019, London, UK
November 14, 2019, Maritim Hotel, Berlin
December 3-5, 2019, Vienna, Austria
December 3, 2019, New York, New York
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events