IBM engineers have developed a CMOS integrated nano-photonics technology that may help in producing 100Gb/s optical transceivers. This latest development will enable datacenters to provide faster data rates and bandwidth for Big Data and cloud computing applications.
Instead of using electrical signals, silicon chips will use light pulses over wires to transfer information at longer distances and at high speeds in upcoming computing solutions.
Small optical components are used by silicon photonics to send pulses of light and transfer significant amount of data at extreme speed between supercomputers, large datacenters, and computer chips in servers.
This helps in resolving the issues related to conventional interconnects and congested data traffic. The innovative technology helps in integrating different types of optical parts side-by-side using electrical circuits on a silicon chip through sub-100nm semiconductor technology.
The silicon photonics chips developed by IBM utilize four colors of light that travels inside an optical fiber and not on the copper wiring to send information across a computing system. Within a single second, the new transceiver can digitally share six million images or 63 million tweets, or it can download a full HD digital movie in a couple of seconds.
Making silicon photonics technology ready for widespread commercial use will help the semiconductor industry keep pace with ever-growing demands in computing power driven by Big Data and cloud services. Just as fiber optics revolutionized the telecommunications industry by speeding up the flow of data -- bringing enormous benefits to consumers -- we’re excited about the potential of replacing electric signals with pulses of light. This technology is designed to make future computing systems faster and more energy efficient, while enabling customers to capture insights from Big Data in real time, stated Arvind Krishna, senior vice president and director of IBM Research.
As the technology sector enters a new eon of computing, cloud computing services and IT systems are needed to process and examine large volumes of Big Data rapidly, mainly between cloud computing services and within datacenters. To achieve this effect, data has to be moved quickly between system components without any major obstruction. Silicon photonics not only improves response times, but also delivers faster insights from Big Data by reducing data bottlenecks between computing components and within the systems.
The latest CMOS integrated nano-photonics technology integrates structures as well as critical electrical and optical components, thus providing a low-cost silicon photonics solution and allowing fiber packaging on a silicon chip. Silicon chip foundry can employ the standard fabrication processes and make this technology ready for large-scale commercialization.
IBM’s CMOS integrated nano-photonics technology makes use of the special characteristics of optical communications. These characteristics comprise high-speed data transmission across kilometer-scale distances and the capacity to superimpose different colors of light inside an optical fiber to increase the volume of data transferred and at the same time maintain low energy consumption. These properties integrate to facilitate fast data transfer between supercomputers, racks within servers, and computer chips to reduce the restrictions of congested data traffic introduced by modern interconnect technologies.
Optical interconnects move data through light pulses by way of optical fibers, forming an essential part of next generation datacenters and modern computing systems. Hardware parts of computer, irrespective of their size, can easily and effectively communicate with one another at rapid speeds through optical interconnects. This flexible design of datacenters will aid in reducing the cost of power and space and the same time would help in increasing the analysis and performance capabilities for users working in different fields.
At present, datacenters are using optical interconnect solutions that are based on vertical cavity surface emitting laser, also known as VCSEL, technology where the optical signals are sent by means of multimode optical fiber. The growing demands for high data rate and distance between ports are resulting in the development of affordable single-mode optical interconnect technologies that are capable of resolving the distance and bandwidth limitations associated with multimode VCSEL links.
The new CMOS Integrated Nano-Photonics Technology presents a low-cost solution to expand the data rates of optical links. It is possible to integrate the key components of an optical transceiver (optical an electrical) on a single silicon chip. These components are developed to work with traditional silicon chip manufacturing processes.
A reference design has been demonstrated by IBM Systems Unit and IBM engineers in Zurich and New York to target datacenter interconnects with up to 2km. This chip shows reception and transmission of high-speed data through four laser colors, where each laser color functions as a separate 25Gb/s optical channel. The four channels within a transceiver design can be wavelength multiplexed on-chip to give 100Gb/s bandwidth across a duplex single-mode fiber. This helps in reducing the cost of the fiber plant deployed in the datacenter.
IBM will present more information at the 2015 Conference on Lasers and Electro Optics, which is being held in San Jose, California from May 10 to 15, 2015. Details will be shared during the presentation titled “Demonstration of Error Free Operation Up To 32 Gb/s From a CMOS Integrated Monolithic Nano-Photonic Transmitter,” which will be delivered by Chi Xiong, Douglas M. Gill, Jessie C. Rosenberg, Jonathan E. Proesel, Marwan Khater, Jason Orcutt, Doris Viens, John Ellis-Monaghan, William M. J. Green, Wilfried Haensch, and Yurii Vlasov.
IBM Research has been at the forefront of silicon photonics for over 10 years. The company has achieved a range of technology milestones starting from 2006 onwards. Silicon photonics technology is part of IBM's $3 billion investment to develop next-generation chip technology to meet the ever-growing need for Big Data and cloud computing systems.