For more than a decade, large carriers and ISPs that own twisted copper plants have looked at layer1 and layer2 based ADSL bonding technologies such as MLPPP (Multi Link Point-to-Point Protocol). MLPPP, or similar layer1/layer2/layer2.5 based legacy bonding protocols never became practical because of several inherent shortcomings of the technology. For starters, it required a forklift upgrade at the telco POP, i.e. the DSLAM with the new DSLAM card that supports MLPPP, which meant high upfront cost. This upgrade was needed for every small coverage area of a DSLAM that was planned for bonded ADSL service offering, i.e. a one-to-one upgrade on the server side. The second shortcoming of the legacy DSL bonding was the fact that because of the variations in the DSL line performances, the DSL bonding efficiency was usually very low, even in some instances forcing the bonding algorithms to choke to the level of the lowest common denominator of the DSL lines.
However, in recent years with the introduction of Broadband Bonding, the ADSL bonding approach has discovered a new life since both of the limitations were practically eliminated with the move towards bonding at layer3 and above. What this type of DSL bonding, namely Broadband Bonding, enables is that the peering unit that was required to be at the DSLAM with legacy methods, can now be moved anywhere in the Internet (since bonding is done at a routable layer now) and therefore the one-to-one requirement of the legacy DSL bonding is eliminated. In effect, the telco can install a single (or few) peering DSL bonding Broadband Bonding server units in a few of their POPs, say in their data centers, and can have a DSL rollout in very large geographical regions without the upfront cost. Similarly, since the algorithms are adaptive and agile in Broadband Bonding, the performance is also improved drastically. Add on top of this the intelligence of the higher layers providing additional QoS functionality and you end up with a practical DSL bonding technology, also known as Broadband Bonding.
Here is a case-study of an ISP that implemented Broadband Bonding for T1 and DSL bonding. As the case study explains, the CapEx for the rollout was minimal.
The beneficial implication of the higher layer Broadband Bonding solutions is that there is no telco / carrier coordination involved. Anyone can create their own bonded ADSL WAN connection without requiring installation by the carrier. Enterprises can deploy bonded DSL in their branch offices cost effectively by just using Broadband Bonding devices. They can even mix and match ISPs to add further diversity and reliability, but that topic deserves its own write-up…
Cahit Akin, CEO, Mushroom Networks, Inc.
Mushroom Networks is the provider of SD-WAN (Software Defined WAN) and NFV solutions capable of Broadband Bonding that enables self-healing WAN networks that route around network problems such as latency, jitter and packet loss.