When it comes to Gfast, the discussion I’ve had with multiple customers centers around the question of whether to utilize VDSL2 fallback functionality or to dedicate the whole MDU available spectrum to Gfast.
Of course, each has its merits, depending on your business drivers and network architecture.
However, when it comes to Gfast within MDUs, dedicating all available possible bandwidth to Gfast makes the most sense.
MDUs really are a battle between DOCSIS providers and some fiber-to-the-apartment providers for the last number of years. If you are deploying Gfast as the fastest time to market solution to beat the competition, then can you really afford to limit the speed offering?
Whether you are deploying Gfast chipsets that support VDSL2 fallback or utilizing existing DSLAMs and co-existing VDSL2 and Gfast across the same copper binders, you are paying a performance penalty. A general rule of thumb has been that it is around a 20 percent penalty, but what does that really mean for a marketable service?
Within an MDU environment, where you need to provide services to apartments within 50-100 meters, that can be the difference between marketing a DOCSIS-beating 500 Mbps service or not. Possibly, and more importantly, it can be the difference between a true residential or business gigabit when utilizing a Calix Gfast bonding solution with the flexibility of Dynamic Timing Allocation.
So when it comes to MDUs, ask yourself what is your business driver, better speeds for a few or competition beating services for all?