by Pierre Herve and Shlomo Ovadia, Intel Communications Group, Intel Corp.

Abstract
Optical networking technologies have been over the last two decades reshaping the entire telecom infrastructure networks around the world. As network bandwidth requirements increase, optical communication and networking technologies have been moving from their telecom origin into the enterprise. For example, today in data centers, all storage area networking is based on fiber interconnects with speeds ranging form 1 Gb/s to 10 Gb/s. As the transmission bandwidth requirements increase and the costs of the emerging optical technologies become more economical, the adoption and acceptance of these optical interconnects within enterprise networks will increase. This paper, which provides the framework for the different optical interconnect technologies in this special optical issue of the Intel Technology Journal, is organized as follows. First, a brief overview of the fiber optics interconnects technology evolution and its current application within the enterprise, is presented. Second, various interconnect evolution paths, such as, board-to-board, chip-to-chip, and on-chip interconnects, are discussed.

Introduction
The birth of optical communications occurred in the 1970s with two key technology breakthroughs. The first was the invention of the semiconductor laser in 1962 [1]. The laser generates a tightly focused beam of light at a single pure wavelength, a spot small enough to be connected to fiber optics. The second breakthrough happened in September 1970, when a glass fiber with an attenuation of less than 20 dB/km was developed [2, 3]. In the 1960’s, glass-clad fibers had an attenuation of about 1 dB/m, which was sufficient for medical imaging applications, but was too high for telecommunications. With the development of optical fibers with an attenuation of 20 dB/km, the threshold to make fiber optics a viable technology for telecommunications was crossed. In 1977, AT&T installed the first optical fiber cables in Chicago [3]. The first field deployments of fiber communication systems used Multimode Fibers (MMFs) with lasers operating in the 850 nm wavelength band.

These systems could transmit several kilometers with optical losses in the range of 2 to 3 dB/km. A second generation of lasers operating at 1310 nm enabled transmission in the “second window� of the optical fiber where the optical loss is about 0.5 dB/km in a Single-Mode-Fiber (SMF). In the 1980’s, the telecom carriers started replacing all their MMFs operating at 850 nm. Another wavelength window around 1550 nm was developed where a standard SMF has its minimum optical loss of about 0.22 dB/km. The development of fiber-based telecommunication systems in the 1990’s focused on increasing their transmission capacity. This was done first by increasing the signal modulation speed from 155 Mb/s to 622 Mb/s, to 2.5 Gps, and finally to 10 Gb/s, today’s modulation speed. The total available bandwidth of standard optical fibers is enormous; it is about 20 THz. Since it is impossible for a single-wavelength laser to utilize this enormous bandwidth, multiple single-wavelength laser transmitters are typically multiplexed and transmitted on a single fiber. This scheme, which was developed in the mid-1990s, is called Wavelength-Division-Multiplexing (WDM) [4]. Dense WDM (DWDM) optical communication systems with more than 60 wavelengths, where each wavelength carries 40 Gb/s data, have been demonstrated [5]. Thus, the demonstrated total transmission capacity of an SMF is more than 2.5 Tb/s.

» next |  download | 927.52 KB | Intel Software Dispatch resource center