New Solutions for Local Loop Access

There is a hot product area that has attracted plenty of venture capital (VC) money. This hot area I am referring to is local loop access. VC companies are literally jumping into this market to exploit the opportunity created by the need for high-speed connections between enterprise facilities and service providers. And now the earliest of the VC-created companies are starting to roll out their products to an enthusiastic audience. T1 bonding is one of the solutions coming to market as a result of venture capital investments.

To understand the emergence of T1 bonding, one must look at the market dynamics of T1 lines. Contrary to early expectations, T1 is in a period of strong and powerful growth. T1 has benefited from the growth of the Internet and its associated incarnations -- intranets, extranets and VPNs. Clearly the Internet kicked new life into T1's lifecycle, at least in regard to the use of this technology for local loop access -- the communications space between the enterprise facility and the service provider.

Unfortunately, not all is well with T1. Rated at a speed of just over 1.5 Mbps, T1 is proving too slow for the most demanding of access needs. To see this, just imagine connecting an Ethernet LAN to a T1 port. Even the slowest variant of Ethernet --10 Mbps -- would swamp a T1 connection -- 1.54 Mbps.

Many think the solution to the problem lies with T1's big brother, T3. Rated at a brawny 44 Mbps -- the equivalent of about 30 T1 lines -- T3 can handle the most demanding local loop access requirement.

Unfortunately, T3 has its own set of problems. First, it is not universally available. Unlike T1, it is mostly a metropolitan area service. Large enterprises with dispersed facilities cannot expect to get it everywhere they need it. Second, T3 suffers from a price/performance liability. Priced at 10 to 15 times the cost of a T1, T3 can be cost justified only if there is a requirement for local loop access that falls in the range of 10 to 20 Mbps.

These complications, therefore, explain the interest in local loop technologies. Right now there is a genuine sweet spot in the market for products that can deliver local loop access anywhere with speeds greater than T1, but at prices less than T3.

There are many categories of products being targeted at this demand area. One of the more interesting might be called T1 bonding technology. Bonding refers to a technique by which many lower speed connections can be combined into one higher speed pipe. With this particular solution, multiple T1 lines can be combined to provide bandwidth that scales between T1 and T3. Most importantly, each of the component T1 lines is purchased at a low price. Users can add T1 lines to the bonding fabric until they reach the cross over point at which T3 become competitive. Or, if they reside in a place where T3 is not available, they can continue to add T1 lines until the aggregate bandwidth meets or exceeds T3's capacity.

Many of these products have dual personalities. On one side, they consist of bonding technologies that manage and optimize collective T1 lines. On the other side, they possess a complete router personality that includes support for routing table updates, QoS (quality of service) and a range of queuing mechanisms. This router personality guarantees, for example, that voice and video traffic will have their QoS requirements taken into account when data is stripped over the bonded T1 lines.

Note that this type of product is not being targeted at the end-user. Rather it is being sold to ISPs that will offer it to clients as a way of getting higher speed Internet access. With this offering, ISPs can be sure that T3 availability or pricing will not dampen a client's ability to obtain higher speed services. --Sam Alunni is vice president of networking at Sterling Research (Sterling, Mass.). Contact him at