SAN JOSE, Calif. -- Kotura is weeks away from announcing a partnership with “a large CMOS foundry” that will make its silicon photonics chips. The news is one more sign companies including Cisco, Intel and Luxtera aim to ship the technology in 2014 as a disruptive, lower cost alternative to today’s 100 Gbits/second optics.
Kotura (Monterey Park, Calif.) is “in the early stages of the R&D process” with the unnamed foundry, said Arlon Martin, vice president of marketing, contracts and government affairs for Kotura. “There’s been a lot of work in the background, but we are not ready until late 2013 to sample product from them,” he said.
The privately-held company employs about 80 people and shipped its first silicon photonics chip in 2006. Since then sales have grown to about 100,000 units a year made in its own modest facility.
Kotura hopes to be in production in 2014 with a 4x5 mm module that pumps 100 Gbits/s over one optical fibre. It will consume about 3.5W and, thanks to its use of single-mode fibre, it can theoretically deliver distances of 20 meters to 2 kilometers.
The product uses four external lasers—its only optical components—modulating four 25 Gbit/s signals on to a single fibre with wavelength division multiplexing (WDM). The module also contains a third-party CMOS driver. Kotura’s CMOS chip, made in a 200nm process, uses some germanium for its light detector function.
Martin said the product could start selling for less than $1,000 in volumes and eventually decline to about $100. That compares with some estimates of as much as $100,000 per port for 100G Ethernet today and about $250 for 40G.
“Our fab is fine for volumes we are doing in telecom and great for us to get started, but a commercial fab lets us ramp” Martin said.
on the chip level, some think Si photonics optoelectronic chips are treated unfairly as all the standards are written in accordance to III-V materials. Some of these standards may not be absolutely necessary for use in practice. It's why most Si photonics products are used in AOC which only have end-to-end electrical specs. Hopefully big players like Intel, IBM can change it.
I think the battle of standards is in form factor. Both Cisco and Intel rolled out their own connectors to compete with existing MPO connector for highly parallel optics. Intel and IBM&Avago even wanted to completely eliminate all current panel mount transceivers module such as QSFP+, CFP, etc. which is a big threat to many component&module suppliers. We'll see how it goes.
First of all, it is not to be confused with 100G used in telecom which used advanced modulation to achieve serial 100Gb/s. It's for datacom 100G which can be 10x10Gb/s or 4x25Gb/s, the former was adopted earlier but slower and seems being replaced by the latter as 25G electronics are getting ready. Most early advocates of 10x10G shifted to 4x25G now. No matter for telecom 100G or for datacomm 100GBASE, no 100Gb/s electronics is actually used or needed.
Wonderful article! What do you refer to by " there are no standards"? 100G standard is more mature than 40G, e.g. PM-QPSK modulation is the only one adopted in 100G,while there is no standard modulation method in 40G.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.