Solutions to help your business Sign up for our newsletters Join our Community
  • Share

For small telcos: Ethernet or GPON?

Home-run fiber is an increasingly popular choice for both technologies.

More on this Topic

Industry News

Blogs

Briefing Room

Many rural carriers are pursuing fiber-to-the-premises deployments — but I’ve been hearing some widely divergent views about whether it’s better to use a gigabit passive optical network or active Ethernet approach to FTTP.

The carrier making the largest-scale FTTP deployment — Verizon — has opted for GPON, and initially most small telcos also opted for a PON approach. But active Ethernet seems to be on the rise.

Carriers in the active Ethernet camp say it’s more future-proof and doesn’t cost much more to deploy.
But carriers that favor GPON say it costs less on an ongoing basis because it consumes less power. So which is better?

Two companies I thought should have a good handle on this topic are Calix and Occam Networks. Both companies sell a lot of FTTP equipment to rural telcos — and both offer equipment that can support either GPON or active Ethernet. I recently checked in with both of them to get their takes on the topic of active Ethernet vs. GPON — and heard quite a few common themes.

Both Geoff Burke, senior director of corporate marketing for Calix, and Juan Vela, director of solutions marketing and strategy for Occam, said both GPON and active Ethernet are popular with rural carriers. But both are seeing some shift toward active Ethernet.

When telcos first started deploying FTTP, the common thinking was that a PON approach would be best because it would eliminate the need to maintain active electronics in the outside plant. But there’s been an important change since then: The cost of deploying fiber has decreased. That means it’s now practical for the active electronics that support active Ethernet to be installed inside the central office (CO), supported by a dedicated fiber connection to each subscriber. That approach should make active Ethernet virtually future-proof because speeds can be easily upgraded by simply upgrading the electronics in the CO.

For that reason, rural carriers that face competition from cable companies or others are most likely to choose active Ethernet, Vela said. Competition, he added, “forces them to look at capacity as a weapon of choice.”

It’s difficult to directly compare the speeds that active Ethernet and PON can support, but in general, active Ethernet already has a speed advantage. Active Ethernet systems available today support either 100 Mb/s or 1 Gb/s symmetrical bandwidth per customer — although most carriers are not yet offering the full throughput that the system can support.

GPON, on the other hand, delivers 2.4 Gb/s downstream to a neighborhood node, where the signal for each customer is split off onto its own fiber. Each node can support anywhere from 16 to 128 customers, who share the total bandwidth. Upstream bandwidth of 1.2 Gb/s is also shared. Here, too, carriers typically don’t offer the full throughput the system can support, but in a typical system with 32 customers per node, average maximum speed works out to about 75 Mb/s downstream and about half that speed on the upstream side.

Perhaps that’s not a huge difference in comparison with a 100 Mb/s active Ethernet system. But future bandwidth upgrades are more complicated with PON because they require changes at each neighborhood node — unless the carrier has opted to use the same fiber architecture that active Ethernet has moved to, with a dedicated fiber path between each customer and the CO.

Both Burke and Vela told me that’s a move that more and more rural carriers are making at the recommendation of their engineering companies, even when they are deploying GPON technology. The engineering companies, Burke said, “have told a lot of IOCs to home-run fibers from a future-investment-protection perspective.”

A network architecture that’s based on home-run fiber not only simplifies network upgrades, it also creates the possibility of moving from GPON to active Ethernet at some point in the future.

That’s a choice that appeals to a lot of small carriers that have some compelling reasons to use GPON — at least for now. GPON remains less costly, particularly for carriers like Verizon, which have a lot of customers per square mile. Even for smaller carriers, Vela said GPON can cost between 10% and 25% less than active Ethernet. GPON also doesn’t require as much power as active Ethernet.

A final consideration is the carrier’s video strategy. GPON’s shared downstream path is well suited to an analog broadcast approach using a dedicated wavelength — an option commonly called an “RF overlay.” The RF overlay approach is a natural fit for a lot of smaller telcos because, as Vela said, “a lot of them own cable systems and have sunk costs in an analog headend that they want to leverage.” Active Ethernet, on the other hand, can only support IPTV, where each customer receives a customized digital stream.

Clearly there are no easy answers here, but small telcos seem to be approaching their FTTP network choices pragmatically and in a forward-looking manner. Making smart choices is “enormously strategic,” particularly for small IOCs, Burke said.

But the right choice could reap big rewards in a competitive marketplace. “If you’re the first guy with fiber, it may be costly and the economics may be a challenge,” Burke said. “But the economics for the second provider will be untenable."

Want to use this article? Click here for options!
© 2014 Penton Media Inc.

Learning Library

Webcasts

White Papers

Featured Content

The Latest

News

From the Blog

Briefingroom

Join the Discussion

Resources

Get more out of Connected Planet by visiting our related resources below:

Connected Planet highlights the next generation of service providers, as well as how their customers use services in new ways.

Subscribe Now

Back to Top