The Terabit Ethernet Chase Begins
The University of California, Santa Barbara (UCSB) has already worked on areas such as optical switching and silicon photonics. So why not combine parts of those and launch a quest for Terabit Ethernet?
UCSB professor Dan Blumenthal says the goal of the Terabit Optical Ethernet Center (TOEC; Website still pending), announced last week, is to create Terabit Ethernet by 2015 -- and to follow it up with 100Tbit/s Ethernet by 2020. (See UCSB Pursues Terabit Ethernet.)
It's not pure research. Blumenthal wants TOEC, in conjunction with companies and other university researchers, to produce something the industry can use, not a one-time lab experiment that only works with duct tape and glue.
"We're not talking about lab hero experiments," says Blumenthal, whose name is familiar in optical circles. He and fellow professor John Bowers were founders of Calient Technologies Inc. , and their ongoing optical research has made them regulars at OFC/NFOEC.
The real-world focus of the work has helped attract Agilent Technologies Inc. (NYSE: A), Google (Nasdaq: GOOG), Intel Corp. (Nasdaq: INTC), Rockwell Collins Inc., and Verizon Communications Inc. (NYSE: VZ) to help with the research.
"'Centers' are not really well aligned with industrial finance," but a team working on a problem is, Blumenthal says.
TOEC is being funded by various grants that Blumenthal pooled together. No new funding is involved; it's more like pointing several bits of funding in the same direction.
That's not to say UCSB intends to develop Terabit Ethernet by itself. Rather, Blumenthal envisions his center becoming a focal point, the organizer for research, workshops, and standards development. Blumenthal is reaching out to other universities to join the cause.
TOEC could probably use the help, because Terabit Ethernet is looking like no simple task. Bob Metcalfe, Ethernet's creator and now a Polaris Venture Partners partner, speculated two years ago that a terabit standard might require a rethinking of everything, even the fiber itself. (See Bob Metcalfe on the Terabit Ethernet.)
Some help will come from inside the university. TOEC is part of UCSB's Institute for Energy Efficiency, and Blumenthal intends to draw on TOEC's sibling organizations, such as the silicon photonics group headed by Bowers, or the Greenscale Center for Energy-Efficient Computing (a data-center center, basically) run by Fred Chong.
So, what's the recipe for Terabit Ethernet? Keeping in mind that everything is still in the "maybe" phase, here are a few possibilities Blumenthal listed, most of them drawn from ongoing UCSB research.
- Photonic integrated circuits (PICs) are a must, Blumenthal says.
- So are coherent receivers, but at a scale well beyond what's being used for 100Gbit/s Ethernet. A likely candidate is 1,024-QAM: quadrature amplitude modulation (QAM) transmitting 10 bits per symbol, a scheme likely to require 100GHz electronics.
- To make that coherent receiver energy-efficient, TOEC is "trying to move a lot of what's in the digital signal processor into the optics," Blumenthal says.
- New materials for fiber-optics aren't out of the question. "We won't start out with that, but it'll move in that direction," Blumenthal says.
- Other items on the shopping list include optical phase-locked loops, new semiconductor optical amplifiers (SOAs), and methods for drastically lowering on-chip optical losses.
The questions go beyond the optical layer. For 10Gbit/s and 100Gbit/s Ethernet, some padding and frame delineation was added, to make operations more synchronous, Blumenthal says. "Do we keep doing that? Or do we go purely asynchronous? We don't know yet.
"Once you put the word 'Ethernet' in there, it's not about just transmission. It's about being backwards-compatible. That's the beauty of Ethernet. We can't lose that essence."
— Craig Matsumoto, West Coast Editor, Light Reading