The current industry position for next-gen standards

While optimism for a late 2009 finalization of the 10G EPON standard abounds, the actual deployment of the technology likely will take much longer, due to several considerations.

Alcatel-Lucent is a co-author of the IEEE standard for 10G EPON currently being developed, and it has a significant role in developing the 10G GPON standard too, according to Marcus Weldon, CTO for Alcatel-Lucent's wireline network division. Weldon said significant deployments of 10G PON technologies, both GPON and EPON, likely won't occur in live networks until 2011. He said that even once 10G EPON has been standardized and products have been developed around the standard, uptake could be limited by the success of current deployments of 2.5G GPON.

"We don't see any urgent need in the market for 10G technology," Weldon said. "Currently with 2.5G GPON, the upstream and downstream speeds at capacity are more than sufficient for customers' needs, and very few deployments actually have the take rate to be at full capacity, meaning speeds are usually greater for individual users anyway. The biggest play I see for 10G EPON currently is high-density areas where it allows for more nodes per PON through greater splitting capacity for multi-dwelling units."

Scott Wilkinson, VP of marketing for Hitachi, is also less than optimistic about the market opportunity for 10G EPON.

"There are problems with both flavors of IEEE standardized 10G EPON currently available," Wilkinson said. "First, the 10G downstream/1G upstream does not offer enough bandwidth upstream; and the 10G downstream/10G upstream is prohibitively expensive, because it requires a pricy core chip set."

He echoed Weldon's comments about successful GPON deployments keeping future demand for 10G low.

"Right now, GPON provides plenty of bandwidth, a lot more than people use or need," Wilkinson said. "Many carriers are deploying GPON right now, and they'll be able to upgrade to 10G EPON when it becomes available without changing outside points of deployment."

Wilkinson expects some limited 10G EPON deployments, mainly to enterprise customers, some time early next year, with commercial deployments beginning in late 2010. He was quick to add though that the commercial viability of 10G EPON has yet to be demonstrated.

"It will be two to three years before we see any significant residential 10G EPON deployments in the U.S.," Wilkinson said.

Julie Kunstler, VP business development for Teknovus, said her company is well underway with development of 10G EPON chip sets. Kunstler expects chips to ship in late 2009. Kunstler sees high demand for 10G EPON technology, but she said the driving factors are very different for developing and developed nations.

In Japan, for instance, Kunstler said the majority of demand for 10G EPON has been due to large amounts of bandwidth needed to run high-end video applications. She said numerous vendors and carriers are testing the standard for future deployment to meet their ever-growing bandwidth needs.

In China, on the other hand, carriers want 10G EPON for the density it would enable in high-occupancy buildings in China's rapidly growing urban areas. Since existing infrastructure is poor in most parts of the country, Kunstler said carriers are looking at newer technologies for network overhaul, rather than building on top of existing systems as carriers have done in developed markets like the U.S. and Japan. She also noted that Teknovus has EPON deployments in every Chinese province save one, and it will migrate some existing networks to the 10G standard once it is finalized.

While Kunstler said she sees demand rising in both sectors for 10G EPON, she said increased demand will really drive adoption, because it will make the technology more affordable through scale.