When shopping for networks, when you compare costs to what you are paying today, look and see if your bill displays a line item for the Federal USF, the Universal Service Fund.  The surcharge changes quarterly and as of this writing, is 17.4%.  If  you are not paying it now, the carrier has the right to back bill you for this surcharge.  So you need to be aware when you compare costs, that your real cost will be higher than you think.

The Federal USF Surcharge should be applied to everything:

• Sale, lease, installation and servicing of equipment.
• Local loops
• Network backbone

This is not a tax.  The carrier passes through their charge to you as a surcharge.

Prior to 2008, MPLS networks were not subject to the USF.  But that has changed.  And some companies that have networks contracted before 2009, may never have seen this charge on their bill…though they are liable should they be back charged by the carrier.

Despite the millions, or tens of millions, of dollars that your enterprise may have paid out in USF surcharges over the years, the surcharge has never been federally mandated on you, as a business user. The government doesn’t make you liable for AT&T or Verizon’s quarterly USF expense, no matter how much the carriers leads you to think that’s the case.  But unless your contract specifically excludes your payment of this surcharge, you are obligated to pay it, even if years have passed without seeing it on your bill.

But does it really have to be that way? Here’s where things get a little interesting. Before the FCC’s latest move, carriers had been taking different approaches to including some form of USF charge on their MPLS invoices. Those carriers that had already imposed a USF pass-through for MPLS had not necessarily applied it to all of their MPLS services. For example, some imposed it on access charges but not port charges.  But all that has changed as of 2012 and the rules are exceedingly clear.

And the truth is, there have sometimes been “seams” into which customers have been able to place themselves in order to pay less in USF charges on data services. Right before MPLS became popular, many of the nation’s big companies were buying not pure frame relay, but a service called FRASI — Frame to ATM Service Interworking. AT&T in particular tended not to charge a USF pass-through on FRASI customers, more or less officially reasoning that since one end of each connection was ATM — which, unlike frame relay, was never ruled a “basic service” — it would not impose the charge. But the larger reason was that AT&T wanted to win big frame/ATM deals vs. rivals by charging less money, and that was a good thing.  But now AT&T must access the USF, because the rules are finally clear.

So surcharges can ultimately be part of your question. If you’re not in the process of doing a deal, and you are an MPLS customer, the (harsh) reality is that your costs are probably higher than you expected, because your carrier will be paying more and will want to get it back from you. The USF surcharge itself is also going up to 17.4% of applicable revenues on April 1 and the number changes every quarter.  As of Jan 2012, there is no escaping the USF Surcharge.  Do you want to learn more, visit this web site for all the details.

As enterprises around the world replace traditional SDH/SONET private line networks with newer generation point-to-point EoMPLS and port based VPLS services, carriers have rapidly expanded their Ethernet network footprints to keep up with the demand. As a consequence of growing EoMPLS service availability, competition is increasing and prices are declining around the world. However, data from TeleGeography’s Ethernet Pricing Service show that great price disparities persist, both among regions and service providers.

Ethernet services are most readily available, and prices are lowest, in Europe and North America. In Q1 2012, the median monthly price of a GigE EoMPLS pseudowire circuit between Frankfurt and London was $3,448, down 29 percent from Q1 2011, while the median price from New York to Los Angeles was $8,000 per month, 12 percent lower than in the previous year. By contrast, Ethernet circuits connecting to Asian cities remain far more expensive due to more limited service availability, less robust competition, and higher underlying transport costs. The median price of a GigE EoMPLS connection from Hong Kong to London in Q1 2012 was $40,138, down 17 percent from a year earlier, while a circuit from Hong Kong to Tokyo cost $31,096 per month, more than five times the price of a comparable link on the longer London-New York route.

Price differences among service providers can be just as dramatic as price differences among regional markets, with the variance between the high and low prices on some lower-cost routes just as great, or greater, than on more expensive routes. For example, in Q1 2012, the high price of a GigE circuit from Hong Kong to London was $52,500—approximately three times greater than the low price of $16,704. The range of prices on the much less costly London-New York route is far wider: the highest price quoted to TeleGeography, $17,250 per month, was 10 times more than the low price of $1,729.

“Prices of enterprise WAN services often vary far more widely than prices of wholesale carrier services,” said TeleGeography analyst Brianna Boudreau. “While prices will continue to decline as wide area Ethernet services become more standardized and availability grows, large price disparities are likely to persist for years to come.”

TeleGeography’s Ethernet Pricing Service benchmarks the price of international Ethernet service by provider, capacity, service, and route.

NEW YORK, April 23, 2012 /PRNewswire/ – Perseus Telecom, a global provider of connectivity, along with Reliance Globalcom, the Global telecommunications services arm of India’s largest integrated telecom operator Reliance Communications, today announced the launch of the world’s fastest available trans-Atlantic network connection, QuanTA, between major global financial exchanges. The launch of QuanTA represents a landmark development in the latency race to zero and creates new opportunities for trading firms on both sides of the Atlantic. (obviously latency will never be zero or close to zero.  But when it comes to trading, milliseconds matter)

The partnership between Perseus Telecom and Reliance Globalcom has resulted in the creation of an innovative high-speed, ultra low-latency network connection across the Atlantic by leveraging an existing system on the FLAG Atlantic-1 (FA-1) North cable, a trans-Atlantic six-fiber pair system between Long Island, New York and Lands End, United Kingdom. Designed with the latest advancements in optical technology, better dispersion compensation methodology, faster processing equipment and shorter cable paths, this ultra low-latency link represents a capital assured, cost-efficient solution for the sub-60ms RTD latency financial market participants require across the Atlantic. 

“Fast-paced trading environments demand even faster connectivity, particularly across the Atlantic where, traditionally, the patchwork grid of cable systems across the ocean had not allowed for a truly low-latency network,” comments Dr. Jock Percy, CEO of Perseus Telecom. “Our ability to create this network in partnership with Reliance Globalcom at relatively low cost and minimal time to market is a testament to our expertise in developing network solutions for the financial industry. We’re incredibly excited at the launch of QuanTA which we believe marks a game-changing development for the global trading community. As high-frequency trading strategies proliferate, ultra low-latency network connectivity becomes even more of a competitive differentiator.”

Commenting on this offering, Mr. Rory Cole, President and COO – Carrier Business, Reliance Globalcom, said, “Being one of the world’s leading carriers with connectivity across the globe, we are happy to announce that Reliance Globalcom now offers the fastest ultra low-latency route on FA-1. With this service, we are now connecting the U.S. and U.K. on the fastest link, addressing the business needs of our customers, especially in the financial sector, to help them gain a significant market advantage.”

The launch of QuanTA provides trading firms with a more secure, low-latency and efficient network connection across the Atlantic. Now, firms tapping into liquidity at exchanges in the U.S. and Europe can do so with guaranteed minimal latency on a highly reliable connection, minimizing potential risk in network outages or latency.

This is a very brief post, motivated by a consulting engagement that MPLS-Experts is working on right now. This client is building a global private network to service its offices, using eight or ten collocation facilities as the Points-of-Presence. Each collo will be connected with two diverse 1Gbps Layer-1 point-to-point circuits.

So the question came up, what optical circuit do you need for 1 Gig? Not something your average client uses.

OC is short for Optical Carrier, used to specify the speed of fiber optic networks conforming to the SONET standard.

This list shows the speeds for common OC levels.
OC = Speed
OC-1 = 51.85 Mbps
OC-3 = 155.52 Mbps
OC-12 = 622.08 Mbps
OC-24 = 1.244 Gbps
OC-48 = 2.488 Gbps
OC-192 = 9.952 Gbps
OC-255 = 13.21 Gbps

We’ll need a partial OC-24 to provide 1 Gbps on each circuit.

For anyone performing cellular back-haul, there’s a new specification for handling wireless data traffic from a combination of traditional TDM networks and packet-based transport technologies as wireless operators migrate from 2G/3G to 4G and LTE services.

The Broadband Forum has just issued its “Technical Specification for MPLS in Mobile Backhaul Networks,” also known as TR-221.

TR-221 focuses on the applications of MPLS technology in a range of services that may be used to transport wireless traffic in the access and aggregation networks, including IP, TDM, ATM and Ethernet.

It defines the global requirements of MPLS technology in these networks in respect of encapsulation, signalling and routing, QoS, OAM, resiliency, security, and synchronization. It also covers expected services over the back-haul network, including voice, multimedia services, data traffic and multicast traffic, such as multimedia broadcast and multicast services (MBMS).

Adherence to these requirements will create global standards for MPLS-oriented equipment, establishing more network interoperability, speeding deployments and lowering the overall costs of the backhaul network, the Broadband Forum said.

Defining a range of reference architectures for MPLS-based mobile backhaul networks, TR-221 includes specifications for the various transport scenarios applicable to all mobile networks (2G, 3G and LTE). It also specifies the equipment requirements for the control, user and management planes to provide unified and consistent end-to-end transport services for mobile backhaul.

Robin Mersh, CEO of the Broadband Forum, said: “TR-221 is a critical part of establishing multi-vendor interoperability in converged MPLS-based backhaul networks. As mobile operators look to preserve their investment in traditional TDM and ATM networks whilst developing their 4G/LTE architectures, TR-221 will enable them to integrate new packet-based MPLS technologies into their established networks. Operators will be able to evolve their networks to be faster and more efficient to meet the increasing multimedia needs of the mobile user, whilst preserving a lower cost per bit in the backhaul network.”

Federal regulators have approved Inc.’s merger with Global Crossing, and the $3 billion deal is expected to close as soon as next week.

The U.S. Department of Justice approved the deal after reviewing it on antitrust grounds, and the Federal Communications Commission has authorized the transaction, too, Broomfield-based Level 3 announced Thursday afternoon.

“Overall, both companies are looking forward to completing the transaction and working toward creating what we believe will be a new kind of competitor that will offer customers the network reach, scale and reliability they are looking for, all from a team steeped in an entrepreneurial culture and singularly focused on the customer experience,” Jeff Storey, chief operating officer and president of Level 3, said in an email.

In April, Level 3 (Nasdaq: LVLT) reached agreement to buy Florham, N.J.-based Global Crossing (Nasdaq: GLBC) in a stock and debt-assumption deal.

Global Crossing shareholders will get 16 shares of Level 3 stock for each share of their stock. Level 3 will assume $1.1 billion in Global Crossing debt.

Both companies provide data-transport services and own large fiber-optic networks that carry Internet and corporate traffic (think MPLS networks).

The merged company will have fiber-optic networks in more than 50 countries across the Americas and Europe. The combined companies generated $6.26 billion in 2010 revenue.

Combining with Global Crossing will — after the companies wring about $300 million in near-term savings on operations — improve Level 3’s debt-to-cash-flow ratio, which should reduce its borrowing costs if it needs to refinance the combined company’s $7 billion in debt.

U.S. regulators have approved the merger between Level 3 Communications and Global Crossing, the companies announced Thursday.

Level 3 and Global Crossing said the U.S. Department of Justice and Federal Communications Commission have approved the deal.

Broomfield, Colo.-based Level 3 expects to close on the acquisition as early as next week.

The $3 billion merger will create a global communications company with ownership over networks in more than 50 countries and connections to more than 70 nations, and Level 3 has said the acquisition would allow it to better serve enterprises, content providers, carriers and governments throughout North America, Latin America and Europe.

If you haven’t experienced it once in your career, you are lucky.  Carriers and other service providers often include automatic renewal language in their contracts.  After a three-year term, if you do not give notice of termination at least 30 (or 60) days before the end of your term, the contract will automatically renew for the same original term, such as three years, if that was the original term.  This language has stood up in court.

Another carrier trick is to not mention auto-renewal in the paper contract that you sign.  Instead, the contract references “terms and conditions at www.terms.carriername.com shall be incorporated into this agreement”, with the auto-renewal language displayed at this website.  Many customers do not notice this. Or, since it is not in the written agreement, in the future, it is assumed that the agreement does not auto-renew.  Wrong!

Protect yourself! Insist on including this language in your carrier contracts under the Term and Renewal section:

Upon the expiration of the Initial Term, this Agreement shall be renewed
automatically for successive Renewal Terms equivalent in duration of one (1) month
unless terminated by either Party by providing one (1) month written notice of
its intention not to renew this Agreement prior to the end of the Initial Term
or any current Renewal Term.

This language will protect you and your company from unexpected surprises several years down the road.

There’s a price for everything in this world, and  Internet based IP VPNs are no exception. While  IP VPNs are a cheaper alternative to any MPLS network, it doesn’t necessarily mean they’re for everyone, as customer requirements always vary. In this posting, I will explain both the Internet IP VPN advantages and disadvantages.

Let’s take a look at a few IP VPN advantages over most MPLS circuits:

• Cheaper rates. Internet service providers provide a simple NxT1, Ethernet or Cable connection to the Internet, using the highest possible speed with. The price for internet connectivityis considerably cheaper than almost any WAN MPLS service, making it extremely attractive for companies seeking to cut telecom costs.
• Fully configurable. WAN engineers have total control over the VPN tunnel created between sites. They are able to perform on-the-fly configuration changes to compensate for any network problems or help rectify any problem that might arise. With full access to the VPN, terminating equipment like routers and firewalls, engineers have the ability to see the condition of the internet circuit and take any action(s) deemed necessary…provided they have the staff resources and skills.
• VPN backup included. For mission-critical sites, backup via another internet circuit is possible if your primary connection fails.  Time response for the backup line to come online is configurable by the network engineer, and there is no need to wait for the ISP to fix a line so your company can continue working.
• Two-in-one. When configuring the site-to-site VPN, engineers can also configure remote VPN access for users traveling around the country or world, a feature most companies would have to pay additional money for to receive from their service providers.
• Upgradable features. Perhaps one of the strongest advantages is the fact that your site-to-site VPN characteristics are strictly dependant on those that your VPN routers/firewall support. This means that as new features are introduced with the newer router operating systems (i.e., Cisco IOS), they will be available to your engineers to implement. For example, QoS pre-classification was a feature Cisco introduced in its IOS that fixed a number of QoS features for different services running over VPN tunnels. Dynamic Multiple VPN (DMVPN) was another great feature allowing scalable IPsec VPN tunnels between multiple sites. DMVPN allows every endpoint to dynamically build a VPN tunnel with any of its other peers, providing a low-cost mesh VPN solution.

If the brief list of the above  of Internet IP VPN advantages seems overwhelming , you have read a few of its disadvantages.

Here is a list of a few disadvantages of Internet IP VPNs over almost all WAN MPLS circuits:

• Limited QoS. In order to have a fully functional QoS model, you need to have control of all equipment and paths that your VPN packets run through. In the Internet IP VPN model, QoS is effective in each site’s LAN, up until the L interface of the routers. From there on, packets enter the ISP’s network, and your ISP will clearly state that there is no QoS for such connections. Everything is based on a “best effort” delivery mechanism and you can’t argue about that. Any QoS parameters inserted in your WAN packets are, in most cases, ignored by the ISP.
• No Class of Service Prioritization. It’s the internet, sorry.
• Higher Packet Loss and Latency. If you use interactive applications, video, voice domestically or are connecting to locations more than 3,000 miles away, the MPLS network will outperform the IP VPN hands down.
• Undependable voice and video. If you use voice or video over your network, the MPLS network will outperform the IP VPN, hands down with dependable and consistent performace.
• Possible bottlenecks and low speeds. In an Internet IP VPN scenario, your company connects to the Internet, which has quite a variation of performance.  If there is heavy traffic on the Internet, chances are you might experience lower speeds during peak-hour times. Again, there is no guarantee of the performance.
• VPN and router/firewall security. You are exposed directly to the Internet. This means that the security of your VPN and terminating equipment (routers and/or firewalls) are your responsibility. If your engineers do not take the necessary measures to secure the equipment correctly, this can lead to the exposure of your company to the Internet. This is not a topic to be taken lightly, as the damage can be devastating. It is extremely important to understand the risk involved and to have the required technical expertise to ensure the job is performed correctly. Under ideal circumstances, where the equipment is correctly configured, there is no need to worry—you’re safe.
• Denial of service attacks. With a direct Internet connection, you are exposed to any denial of service (DoS) attack. All attempts can be successfully repelled; however, keep in mind that the traffic will have to reach your router/firewall first. This means that the heaviest damage that can be produced by a DoS attack—for a correctly configured endpoint—is to create a bottleneck on your connection and greatly reduce speeds for the duration of the attack.

If you want a rock-solid WAN with almost no packet loss and the lowest possible latency and quality, consider an MPLS network.

Enterprise customers around the world are replacing legacy private line, Frame Relay, and ATM wide area networks (WAN) with Ethernet VPLS and MPLS IP VPN services. Companies’ choice of wide-area network type is shaped by a number of factors, including the applications they need to accommodate, the number of locations to be connected, the level of control the customers want to maintain over their networks, their capacity requirements, and the cost of the solution. However, data from TeleGeography’s Global Enterprise Networks Research Service suggest that the most important factor shaping an organization’s international network choice may simply be availability.

Ethernet VPN services are generally more cost effective than MPLS IP VPN services for capacity requirements above 50 Mbps, and are most appropriate for linking high-capacity headquarter sites and data centers. MPLS IP VPN
services tend to be better suited for linking large numbers of sites with more modest capacity requirements. However, dependence on Ethernet local access and the relatively slow rollout of Ethernet across MPLS PoPs means that Ethernet VPN solutions are not yet available in as many cities.

Ethernet deployments lag far behind MPLS VPN deployments, both by service provider and by geographic market. Over half of the 63 international service providers researched by TeleGeography offer MPLS VPN service in 10 or more countries, compared with less than one-third of Ethernet VPLS service providers.

The availability of IP VPN and Ethernet VPN services also differs by region. TeleGeography identified 39 IP VPN providers in Europe, 34 in Asia, 31 in the U.S. & Canada, and 19 in Africa and Latin American. Ethernet VPN services are less widely available in all of these regions, but the difference is particularly great in emerging markets. While 32 service providers offer layer 2 Ethernet VPN services in Europe, only 9 offer VPLS service in Latin America
and only 6 in Africa. While 22 service providers offer VPLS service in London, only 5 offer VPLS service in Mumbai.

This report should not limit your interest in Ethernet VPN services, but rather shape expectations on its availability.  To determine Ethernet VPN network availability for your company, contact us.

The above content provided from TeleGeography, the world’s leading independent reference source for global network infrastructure data.