Such speeds -- roughly 40 times faster than the most sophisticated corporate data networks -- now require expensive materials, a factor that helps push the cost of existing 40-gigabit modulators into the thousands of dollars. Intel, which boasts the biggest revenue among companies that make silicon chips, wants to use that material to create much less-expensive communication components, an effort it calls "silicon photonics."
Exactly when the company may offer such components, and how much they will cost, hasn't been determined. But Mr. Paniccia says the company is committed to commercializing silicon photonics technology in some fashion by the end of the decade. He adds that such laser components need to cost in the neighborhood of $5 each to be commercially viable.
Luxtera Inc., a start-up in Carlsbad, Calif., is planning to enter the market in the fourth quarter of 2007 with chips that include the equivalent of four lasers, each of them able to send 10 gigabits of data a second. Alex Dickinson, Luxtera's chief executive, said Intel's development is interesting from a scientific point of view. But he argues that Luxtera's approach can bring practical benefits sooner, for applications such as connecting together servers to create a supercomputer.
Luxtera's goal is to bring to market a complete optical link -- two transceivers and the fiber -- for the price of a copper interconnect ($200).
Today's optical links built with discrete components costs around $700 per 10 Gbps, versus $200 for the copper version. Copper is actually fine at 10 Gbps for up to 2 meters. But once you get beyond 10 Gbps or a 2 meter reach, you need more electronic componentry and more copper to maintain the communications. Copper interconnects tend to max out at around 50 meters or so. By contrast, a single optical fiber strand can carry data up to two kilometers. The other big downside to copper is its manageability. As communication distances increase, thicker copper cabling is required, increasing its weight and bending radius. Also, power consumption becomes an issue as data rates rise and communication distances increase. The 4 by 8 millimeter chip will be housed in a standard optical transceiver form factor so that it can be plugged into existing communication hardware. Their CMOS chip is being fabbed by Freescale Semiconductor using a 130nm SOI process.
It seems likely the Luxtera chip will be successful and move to smaller lithography processes for lower price and faster speed.