2010 Bandwidth Explosion Expected, Demand for fiber will skyrocket
As bandwidth consumption continues to grow, TelecomSupportDesk.com - a telecom master agent specializing in high bandwidth circuits and customized network designs - is seeing an increased demand for bigger bandwidth pipes. 2010 will be the year of a huge explosion in the demand for bigger bandwidth pipes. Growing trends in shared cloud computing & centralized private clouds as well as the increase in the consumer demand for bandwidth seem major driving factors.
As an authorized AboveNet agent, TelecomSupportDesk often relies on their fiber optic network. “When a business comes to us looking for high bandwidth from 100Mbps, 1Gpbs even to 10Gbps metro Ethernet solutions, chances are high that AboveNet is the right solution for them,” Mr. Wind (VP of Sales at TelecomSupportdesk.com) comments. “Especially the extreme low latency fiber optics (as low as 1 ms both ways) between two locations is why AboveNet often wins the deal. A low latency route can be extremely crucial for financial institutions as a change in milliseconds can often mean losses of millions in financial trades or transactions.”
In a recent interview with TelecomSupportDesk.com Michael Brown, AboveNet’s VP of Indirect and Carrier Sales, explains how AboveNet has been able to build such an extensive, national fiber network and what differentiates them from other fiber connectivity providers. He also discusses the importance of disaster recovery and business continuity, as well as the importance that low latency can play in a network.
“AboveNet was founded over 15 years ago and started out by an electrical contractor. They had to fight the City of New York to get a franchise agreement to put fiber optic cabling in the ground in order to compete with the ILECs. This is a very capital intensive business and required additional funding. The Metromedia Group decided to invest in the company and changed the name to Metromedia Fiber Network.”
Metromedia Fiber Network had quite an explosive growth in the late ‘90s which occurred as a lot of the telecom companies of the ‘90s were restructuring. They reorganized as AboveNet, Inc. in 2003 and today they are listed on the New York Stock Exchange (ABVT).
“By having gone through this reorganization we have developed a financial discipline and a success-based model,” Michael Brown continues. “We were one of the early companies that got into building metro fiber networks. We used to own datacenters, managed service providers within datacenters and carrier neutral peering which we sold years ago. And with that we went back to our roots, which is being a metro access connectivity provider enabled by dark fiber.”
For those unfamiliar with it, could you please explain the term Dark Fiber?
Dark fiber is a strand of optical fiber with no electronics or optical gear at either end, hence it is dark and not lit up with a laser.
What is the biggest differentiator between AboveNet and other carriers offering fiber routes?
AboveNet has spent considerable time and money building fiber routes in about 15 metros in the US and London. No company is going to spend that much today to build these ubiquitous fiber rich routes nor is anyone going to lend the money to do so. We may have a few competitors in a market here and there, but they are typically local players. Often these competitors are a utility company that has some utility right-of- way, and even though it is not their market, they figured they would put some fiber in the ground. But at the end of the day – whether it is a water utility or a power utility company – if there is a water break or if a power line goes down because of an ice storm, they are likely to be focused on restoring the electricity or water, and after that they are going to look at restoring the fiber. Remember, their main business and responsibility is being a utilities company. So we (AboveNet) may have some competitors in markets, but they are not in all our markets and typically it is not their primary business but just an additional revenue stream for the company.
Beyond that, what also differentiates us is the way we have constructed. Where we could, we have put in multiple conduits. These are 4 inch PVC ducts with three 1.25 inch innerducts where we could put large count fiber cables in, 432 or 864 counts of fiber strands. And when we exhaust the 432 or 864 fiber cable, we still have two more innerducts or another duct itself to pull our fiber through. By having constructed like this we have really made it easy and cost effective for AboveNet to expand. The big cost in building a fiber network is the construction, so if all you have to do to expand a route is pull more fiber through an innerduct, it is infinitely cheaper than having to dig up the ground. Most of our competitors never built with that design philosophy. They put one 96 count cable in the ground and that was it. They have no way of adding to it. Or because they were trying to squeeze it in near a utility line, they couldn’t put in a large conduit system and had to squeeze in a single fiber run with limited amounts of fiber.
So that would mean that AboveNet can expand more easily?
Correct. We are in more markets, we can expand more easily, and we have higher count fiber cables. But at least as important is that we have put in the latest generation fiber types in the ground. There is something called non-zero dispersion shifted fiber. This type of fiber has been designed for dense wave division multiplexing and works at the 1550 nanometer wavelength range. Most fiber that is in the ground is called SMF-28 or just standard single mode fiber and is optimized for 1310 nanometers, which is where SONET gear operates. AboveNet has composite cables that have a combination of both the standard single mode (SMF-28) and the non-zero dispersion shifted. Whether it is Lucent TrueWave or Corning LEAF, we have those fiber types in the ground. And they are optimized to support dense wave division multiplexing, which is what most carriers and enterprises are looking at to deploy. They want a dense wave network supporting all their applications, and not a SONET based TDM network.
When we built the routes we really tried to honeycomb the central business districts, and we built quite extensive backbones in the cities we are in. And we would build these backbones along right-of-ways that our competitors don’t have and where we could we would bury our fiber. For example, our network in Chicago, which is a HUGE network that goes all the way from Lake Michigan to Hoffman Estates in the northwest suburbs of Chicago, and to Lisle in the southwest suburbs of Chicago, it is nearly all under ground. Some of it is built along the Illinois DOT system and in many segments no one else is along many of those right-of-ways. So we have built an extensive system, we built it underground where we could, and we built it along right-of-ways that are unique to AboveNet.
So to sum up what differentiates AboveNet is we:
* Are in more markets
* Can easily add more fiber
* Have a better type of fiber
* Have an excellent presence in central business districts out to the far suburbs
* Are in 15 metro markets
* Are focused primarily on fiber optical networks
* Have unique rights of way
* Have buried the vast majority of our fiber for safety
No one else can really talk to that; it’s a unique value proposition. That’s really why we think we have a huge differentiation between our fiber and other carriers offering fiber.
Since we are not a retail provider we’re not offering dial tone, we don’t offer frame relay. Our fiber doesn’t have to go to every CO (Central Office). In fact, our fiber even avoids COs. The ILEC goes to EVERY CO.
What about CLECS?
Well guess what? They have got to pick up their UNEs (unbundled network elements), their copper loops at a specific CO. We avoid that. So we will typically have the shortest path between point A and point B. That means we can offer low latency and that is extremely important in so many industries today.
Yes, didn’t AboveNet recently win a big deal from the NYSE because of the low latency network? MB: Yes we did. We’re there – http://abovenet.com/newsandevents/pressreleases/pr090721.php They’r one of many financial customers and exchanges that use AboveNet because our fiber optic network can support the highest capacity and the lowest latency.
Aside from financial institutions are there other industries that have such high requirements?
Oh yes, there really are many markets that have such requirements. Media and entertainment where production, post-production, and editing collaborate and send huge digital files back and forth. But also social networking, we count almost every social network as our customer. I don’t know if you have any children but mine are on the bigger social networking sites and they definitely notice when there is any lag in the network. So they do not just need a network that can transport a lot of data; they need a network that can support that traffic with a low latency. That also goes for online gaming. My children are on Xbox live and heaven forbids if the gaming network is lagging, they will notice instantly. So many different industries require low latency; it’s really not just financial services anymore.
One of the most important applications for low latency, which really is applicable to every single industry, is disaster recovery and business continuity. If you are doing disc replication, especially synchronous disc replication (which is also known as disc mirroring or disc shadowing), where you are writing simultaneously to two different disc drives at two different locations, you HAVE to have a low latency network. It won’t work above particular latency or attenuation numbers. A low latency network is the number one determining factor that people are going to look at when deciding how they are going to deploy their disaster recovery and business continuity strategy.
How important would you say a disaster recovery and business continuity plan is?
Oh extremely important! There are many studies that have shown that most companies that do not have such a business continuity and disaster recovery strategy and network in place are in big trouble when they lose critical customer data. In fact, studies have shown that many firms do not survive the following year after a disaster if they did not have a recovery plan in place. It can cripple their business.
We at TelecomSupportDesk.com found many clients with a great interest in your DS3-Killer (aka Terminator) promotion. Why do you think that companies should drop their DS3’s for AboveNet’s fiber metro Ethernet?
The first reason is that a DS3 can’t provide you with more than 45Mbps. We can start you on a 100 megs and take you up to a 10 gig port. Most companies I know require more and more scalable bandwidth, well above DS3. Or they want to be able to burst.
Additionally, with a DS3 you’re probably still using copper technology, likely from the LEC or what the CLEC is providing. For disaster recovery, anyone selling a DS3 circuit probably comes to the building on the same pathway. That’s not much diversity. AboveNet typically comes in with our own infrastructure, as diverse as possible from the ILEC.
Therefore not only can we offer expandability above DS3 speeds, we can do it on our own infrastructure, diverse from the ILEC and we can provide you with native Ethernet. So you won’t be going on DS3 technology, you won’t be going on a SONET ring. Instead of packets over SONET, we offer native Ethernet which is what most companies like. They understand Ethernet, have it in their Local Area Network (LAN), and now they want it in their Wide Area Network (WAN). With our technology, we can converge your Ethernet and IP traffic.
Could you give an example of that?
Let’s say you want to go from your location to a datacenter and you also want to get IP at that datacenter. Perhaps you want to do server virtualization, disc mirroring, or you have signed up for cloud computing for enhanced CPU cycles and disc storing done remotely. In such a scenario you would want to get a low latency fiber tail from us, with whatever port speed and burst ability you need.
No comments:
Post a Comment