Why I avoid Telecom Equipment Providers

[qazulight]

This is a response to Rocko on the FreeforAll board. I was asked to cross post it here. I expect that you guys that can really dig into companies can tear me up, but hey, that’s what we are here for, to have our biases checked.

This is a pick that has been pitched to me over and over again. Before investing in any backbone equipment, I recommend you sit down with AT&T’s White Paper on Domain 2.0

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&a…

NFV aims to address these problems by evolving standard IT virtualization technology to consolidate many network equipment types onto industry standard high volume servers, switches and storage that
can be located in data centers, network PoPs or on customer premises. As shown in Figure 2, this involves the implementation of network functions in software, called VNFs, that can run on a range of
general purpose hardware, and that can be moved to, or instantiated in, various locations in the network as required, without the need for installation of new equipment.

Additionally, when it comes to backbone, there is a trade off between going faster, or just burying more fiber.

Finally, with the Network 2020, or Domain 2.0, AT&T is aiming at everybody. In other words, where INFN is going after operators of data centers as customers, AT&T is going after the operators of data centers for quick snack before dinner.

Inside the Domain 2.0 paper you will see that not only is AT&T going after the Data Centers, they are attempting to move from black boxes to white boxes. The difference is moving from an embedded system like a game console to a Dell Computer. I assure you, if you are selling high mark up equipment you are going to be in trouble. Lest you think this is an AT&T centric thing, nope; AT&T is just trying to run with the big dogs, Google, Amazon, and so on. They are all building Software Defined Networks. (SDN) and moving toward them. Additionally I am seeing intelligence still being pushed from the center out.

Let me explain. Today AT&T has rolled out Voice Over IP over LTE to 93 markets. (Verizon has more I think) The product is marketed as HD voice. From the customers perspective it is a game changer. Going back to the old voice is like watching TV on an HD screen then looking at a 20 inch CRT. Once you hear it, you don’t want to go back. But, it is even more important from the inside out. A voice over IP call can be handled, or switched from a virtual switch, and the computer that acts as the virtual switch can be a virtual computer. In other words, the computer can be spread among several boxes in a building, or in several boxes in several buildings. More than that, it can be instantiated. (Think of the Transporter on Star Trek) anywhere. If the server center in Houston is threatened with flooding, the computers themselves can decide to spin themselves down and spin up in Oakland. If Oakland has an earth quake, the simply reappear in Kansas City.

But even that is just the beginning. We will see by 2025 a move to stop back hauling so much data. We will be able to switch calls, and setup virtual paths and servers near where they are needed. So the demand for data transmission can be reduced. Moreover, we can spin up capacity on the fly. You got Sturgis, we got capacity, Sturgis goes back to sleepy town, the capacity goes away.

I don’t see it. I have AMBA, and Ivensense (Big Loser) and lots of AT&T, but no back haul machines.

Oh, I didn’t answer your question. Fujitsu has been deploying 3 terabyte systems, (per fiber) since 2010 with 5000 mile hops subsea. The last time I was involved in fiber, we were putting down 216 fiber cables, but that was 15 years ago. I am guessing 1000 is more common. I mean, if you dig the ditch, you might as well put something in it.

Cheers
Qazulight

[buynholdisdead]

This is a pick that has been pitched to me over and over again. Before investing in any backbone equipment, I recommend you sit down with AT&T’s White Paper on Domain 2.0

Hi Qazu,

I suspect you are talking about INFN?

Additionally, when it comes to backbone, there is a trade off between going faster, or just burying more fiber.

This is incorrect. They have the fiber in place and they are getting higher bandwidth across a single fiber strand.

Finally, with the Network 2020, or Domain 2.0, AT&T is aiming at everybody. In other words, where INFN is going after operators of data centers as customers, AT&T is going after the operators of data centers for quick snack before dinner.

This is something the Telco’s have dreamed of doing for years. I heard this same idea, only with different technology back in the 80’s. Back then At&T actually had a construction side with their own division that produced some fine equipment. They were called Western Electric and then spun off as Lucent and then sold to Alcatel. Back then they still couldn’t produce a piece of equipment that was Vendor Agnostic. I read the paper that you referenced and this is happening in the Data Centers at this time but it has more to do with Cisco and Anet. They are competing against the white boxes. But I think this trend has less to do with AT&T than it has to do with companies like FB and Google. I am not saying that eventually couldn’t happen to the transport side, because someone coming in with a better product is always a real threat, but I do not see that happening in the transport side for at least 10 years. (That is a guess on my part)

Inside the Domain 2.0 paper you will see that not only is AT&T going after the Data Centers, they are attempting to move from black boxes to white boxes. The difference is moving from an embedded system like a game console to a Dell Computer. I assure you, if you are selling high mark up equipment you are going to be in trouble. Lest you think this is an AT&T centric thing, nope; AT&T is just trying to run with the big dogs, Google, Amazon, and so on. They are all building Software Defined Networks. (SDN) and moving toward them. Additionally I am seeing intelligence still being pushed from the center out.

AT&T as far as I know isn’t doing anything but requesting vendors to build products that will fit their needs. AT&T doesn’t have any way to produce a product that I know of. FB and Google are something to watch because I believe they have the knowledge and money to pull this off, but FB is one of INFN’s customers. Maybe they are using their equipment in order to find a way to compete. That is a possibility.

Let me explain. Today AT&T has rolled out Voice Over IP over LTE to 93 markets. (Verizon has more I think) The product is marketed as HD voice. From the customers perspective it is a game changer. Going back to the old voice is like watching TV on an HD screen then looking at a 20 inch CRT. Once you hear it, you don’t want to go back. But, it is even more important from the inside out. A voice over IP call can be handled, or switched from a virtual switch, and the computer that acts as the virtual switch can be a virtual computer. In other words, the computer can be spread among several boxes in a building, or in several boxes in several buildings. More than that, it can be instantiated. (Think of the Transporter on Star Trek) anywhere. If the server center in Houston is threatened with flooding, the computers themselves can decide to spin themselves down and spin up in Oakland. If Oakland has an earth quake, the simply reappear in Kansas City.

Yea they are kind of behind the market on this.

http://www.top10bestvoipproviders.com/?kw=voice%20over%20ip%…

Many of these companies have been doing that for years. That is one of the great things about Voip. It can be done anywhere in the country. In fact on this board at one time people where investing in a Voip company. I never invested any money into it because voice is a losing proposition. The lec’s and clec’s are bundling it into packages for free. It’s hard to compete with free.

But even that is just the beginning. We will see by 2025 a move to stop back hauling so much data. We will be able to switch calls, and setup virtual paths and servers near where they are needed. So the demand for data transmission can be reduced. Moreover, we can spin up capacity on the fly. You got Sturgis, we got capacity, Sturgis goes back to sleepy town, the capacity goes away.

That has been going on for years. They have placed boxes closest to the customers to keep data down. For Instance Netflix will try to have boxes closest to their customers to stream data. I know the Telco’s have boxes that store the most visited web pages closest to their customers but they continually have to update the boxes with the newest web pages or the data becomes stagnant. The data is not going to go away. You can see this by all the people using all the wireless devices. The data is just going to be getting larger and larger. I see it every day in how much data is being moved through our network and how much money we are putting into our network to upgrade it.

Oh, I didn’t answer your question. Fujitsu has been deploying 3 terabyte systems, (per fiber) since 2010 with 5000 mile hops subsea. The last time I was involved in fiber, we were putting down 216 fiber cables, but that was 15 years ago. I am guessing 1000 is more common. I mean, if you dig the ditch, you might as well put something in it.

Infinera has the first 8 terabyte system. They also have the first 500 gig line side card and soon to be the first 1.2 terabyte line side card.
While your white paper could come to happen, agnostic equipment that will be software based, I do not think it will happen for at least 10 years. But in this business you are always running with one eye over your shoulder. Also the transport companies are working on a SDN solution and Infn, of course, is too. Nobody is standing still.

Thanks for white paper. It had some great information.

Andy

[qazulight]

Thank you Andy,

I Lucent and Nortel were destroyed by phone companies. I have not seen any vendors prosper selling to the phone companies and after owning Invensense, I am having my doubts about Apple and Google.

Also, I believe your doubts about AT&T pulling off Domain 2.0 are justified. If there is a weakness at AT&T it is its software. But, AT&T has a lot of raw talent. I think that AT&T can afford to use the General Grant method of warfare.

Cheers
Qazulight

[buynholdisdead]

Hi Qazu,
I think you have to be worried about this investment, but what investment do you not have to worry about? If its technology there is always something just around the corner but as long as Revenue and Earnings are growing all we can do as watch.

I Lucent and Nortel were destroyed by phone companies. I have not seen any vendors prosper selling to the phone companies and after owning Invensense, I am having my doubts about Apple and Google.

I can’t answer for Lucent because I never worked for them but they seemed to be a good company, Nortel was a great company to work for but a poor investment. They spent anything and everything on their people until towards the end. I would say that spending was out of control. They never thought about tomorrow when it came to money but their technology was top notch.

Also, I believe your doubts about AT&T pulling off Domain 2.0 are justified. If there is a weakness at AT&T it is its software. But, AT&T has a lot of raw talent. I think that AT&T can afford to use the General Grant method of warfare.

AT&T really isn’t the company it used to be. I was invested in AT&T. Lost a good amount of money on that investment and then they were bought out by SBC. SBC then took AT$T’s name.

Qazu,
Infn isn’t an investment that I will be holding forever. There will come a time where this investment will go down but right now this is not the time. But there are very few investments that you can just let run. Oh and my bias is Chinese companies. I will not invest in them because I have lost to much on them. So I do understand not investing in certain companies, countries and sectors. There is a reason for our biases.

Andy