Telecom Investment Bets for Black Friday shoppin

The article below was posted to Light Reading today. In it, Infinera is discussed as the next potential disrupter with their ability to replace the traditional seven-layer stack with their new 2-layer approach. This talk extends past the prior MKM partner article where Infinera was discussed as one of two public plays for the next transition. The other companies seen as catalysts for the disruption are privately held by venture capitalists.…

What Are Telecom’s Best Investment Bets?

Carol Wilson

If picking investment winners and losers in telecom is part of your Black Friday shopping agenda, then you might want to take heed of what two veteran financial industry analysts were saying at the Gen15 event last week. A word to the wise: Their thoughts might prompt Thanksgiving indigestion.

Michael Genovese, managing director and communications equipment analyst at MKM Partners , admitted he sees most incumbent publicly traded companies at risk, due to the disruption that lies ahead in the move to the New IP. The real investment opportunities now are for venture capitalists looking for private companies to fund and for corporate development teams, looking for acquisitions, he said.

That said, there is a “real opportunity to become the next-gen OSS or BSS” in virtualized telecom, but it’s not clear yet who that company is, Genovese said.

[Mike Genovese of MKM Partners shares his thoughts on investing winners and losers, as Stan Hubbard and Catherine Trebnick look on.]

Both Genovese and Catherine Trebnick, VP-senior research analyst for data and communications infrastructure at Dougherty and Co., agree that security is hot for investment. Trebnick specifically called out security for the virtualized telecom infrastructure.

Optical transport companies are actually in relatively good shape, Genovese noted, because physical transport can’t be virtualized. He said recent consolidation has narrowed down the number of players, at least for backbone networks, to the benefit of all. That hasn’t yet happened in metro/data center interconnect, where the analyst is expecting more shakeout to come.

In network management/orchestration – what the MEF now calls Lifecycle Service Orchestration or LSO – there is a lot of opportunity but not a ton of agreement, Genovese noted. In fact, there is a shifting landscape of network topologies.

He and Stan Hubbard, the MEF program chair for Gen15 and moderator of the analyst panel, batted about the notion of new emerging topologies to replace the traditional OSI seven-layer stack, starting with the two-layer topology mentioned a day earlier by Infinera Corp. (Nasdaq: INFN)'s Stu Elby.

Elby laid out a basic network configuration that included Layer C, for cloud services, and Layer T for intelligent transport at the bottom. In between is the software-defined network layer that ties the two together. In his view, routing control moves up into the services layer and the data plane becomes transport.

[The World According to Stu]

Genovese called that approach “a simplified abstraction,” but added that it’s clear the existing seven-layer model is changing. Just walking around Gen15’s exhibit hall revealed multiple different vendor approaches to the new taxonomy, he said.

And that brings the discussion back to the operations and network management side of the house, where Genovese and Trebnick clearly expect major action.

Trebnick speculated about how existing players such as Amdocs Ltd. (NYSE: DOX) and Ericsson AB (Nasdaq: ERIC) will play and whether efforts such as Ciena Corp. (NYSE: CIEN) and its acquired Blue Planet orchestration system will gain traction. Her conclusion is that there is a lot of uncertainty still.

The likely winners among the newer players emerging may never become investment targets, Genovese noted, because it’s likely the carriers will tip off their established vendors to what’s hot and encourage a friendly acquisition. That would be one way that public companies will benefit from the current mass telecom transition.

Gen15’s addition of a day devoted to enterprise technology revealed how little difference there is now between the technologies the two groups are discussing and buying, Genovese noted. Like telecom, enterprises are trying to figure out how little they can get by with doing themselves, in terms of buying and operating equipment, and how much they can push into the cloud to reap the benefits of hyperscale.

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



Kevin, Could you translate what they are saying in that article briefly into English. I, for one, would really appreciate it.




1 Like

After reading the article and referencing some Wikipedia articles for various acronyms, I will share what it sounds like to me. I wouldn’t put too much weight on this.

The telecom world used to be manual. The move to computerization was long and complex, and the end results reflect the legacy of how it developed at least as much as how it needed to work. To make any sense of it at all it was abstracted into layers with defined tasks at each layer and defined intereactions with adjacent layers. The way this has been improved is to implement much of it as virtual layers. The result is what the Brits might call a bit of a dog’s breakfast, or perhaps a virtual dogs’ breakfasts. (Or am I confusing the Brits with the Aussies? Apologies to both, either way.)

The current arrangement is ripe for replacement by something better. That will be a huge deal. Vast parts of the telecom world will be replaced. The replacement would be simpler and more coherent, and aligned to current technology. One example of something that could be part of that is the bit in the middle of that post about replacing the traditional OSI seven-layer stack with a two-layer topology presented by someone from Infinera. Or stated another way, it won’t be the established telecom giants who devise the answers, it will come from one or more upstarts, with INFN as one horse in the race.

Who wins in the end is unknown, but (interpreting a quote in the article) we will find out when the telecom giants give the right nudge to their friends and the winners are swallowed up ahead of reaping their just rewards.

But what do I know about telecom?



I bElieve he is quite right.

AT&T is making a big transformative push. It has three prongs. All IP, Software Defined, and Mobile First

For those that do not understand, IP is short hand, or slang even for Packet Switched telecommunications. We started out with Circuit Switched telecommunicaions. A metaphor would be the difference between taking trains and riding in cars. Trains are much more efficient, but when one considers all of the factors involved in ones life, cars win in over all economy.

We find the same situation in telecommunications. The overhead costs of circuit switched transport are much lower that packet switching. But, packet switching is so much more versatile that it is the future. AT&T and Verizon are on track to completely remove circuit switched communication by 2020. With the roll out of Vioce Over LTE we will see the final demise of the land line.

This eliminates all of the hard switching equipment, and allows the telecoms to build a system based on routers. However, there is more. With the rise of server centers, we have found that a great deal of information passes from machine to machine. We are able to virtualization not only the machines, but the routers between the machines. So that one big physical building would have nothing but “white boxes” with virtual machined creating a virtual network. The allows the telecoms to build the next step, a Software Defined Network.

A software defined network has several advantages over a hardware defined network. The ability to change, even to change at the command of the customer. The days of calling a telecom and waiting 6 months to get a service are coming to a close. The goal of Software Defined Networking is to allow a company to ramp up computing power and telecom transport power automatically and on the fly.

The final big change is fiber. We now have the technolgy to deploy fiber to every farm house and chicken coop in the U.S. This is big deal on a couple of levels. First, speed, there is no speed like light speed. With fiber we get to recreate the entire light spectrum over and over again. (We only get one over the air spectrum and that is why mobile data is expensive and why big companies pay big bucks for spectrum licenses.) The second is not so obvious, fiber is stone reliable. It doesn’t respond to lighting, wet, or cold. And the third is even less obvious, that is the compression of the 7 layer transport model. Once we go to an all fiber, all IP network, you do drop to a 2 or 3 layer transport model.

So, what does this have to do with investing? One, anybody planning on black box switching is in trouble. Cisco comes to mind. Second, the cost of telecom is falling and it will continue to fall. Things that are not economical now, will be and opportunities in these marginal places will become money makers.

One final thought. Wireless, or cellular is a big deal, and with a complete move away from circuit switched systems (2G) due in a year, 56 weeks actually, the resources currently being eating up by the legacy networks will become available for the LTE networks. This is not a one for one thing. LTE is much more efficient at sending bits of data that 2G, or Edge. I recall, from a recent training course, that LTE has an efficiency of 15 bits of data per hertz of bandwidth. (From my training a long time ago in a galaxy far far away. This is completely impossible.) The legacy 2G network has an efficiency of about 1.

Currently, we bring all data from the cells to a central location, but even at the speed of light, this is getting to be slow. My expectation is that we will eventually see fiber run directly between cell towers to make a system that looks more like a fishing net that a spider web.