Photonic and Optical processor technologies for high performance and energy efficient computing

Rajesh Uppal January 1, 2023 AI & IT, Photonics Comments Off on Photonic and Optical processor technologies for high performance and energy efficient computing 460 Views

A Photonic and optical computer is a computer that performs its computation with photons as opposed to the more traditional electron-based computation. The Electronic computers are relatively slow, and the faster we make them the more power they consume,the future optical computers shall also be energy efficient. An electric current creates heat in computer systems and as the processing speed increases, so does the amount of electricity required; this extra heat is extremely damaging to the hardware. Photons, however, create substantially less amounts of heat than electrons, on a given size scale, thus the development of more powerful processing systems becomes possible.

Optical computers promise to be superfast since light travels at 186,000 miles per second. In a billionth of a second, one nanosecond, photons of light travel just a bit less than a foot, not considering resistance in air or of an optical fiber strand or thin film. Optical computing is also promising for Brain-like or neuromorphic computing, mimicking the human brain using electronic chips.

While the computers of today use transistors and semiconductors to control electricity, the Optical Computers of the future may utilize crystals and metamaterials to control light particles called photons.By applying some of the advantages of visible and/or IR networks at the device and component scale, a computer might someday be developed that can perform operations significantly faster than a conventional electronic computer.

The next generation computers and servers will replace electrical signals on copper lines with optical signals on waveguides. Optical lasers and photodiodes are used to generate and receive the data signal. The chips using transmitters and receivers can communicate with each other optically. Researchers are now trying to use same principle for the communication on a chip–between cores and transistors. In future optics shall not only enhance inter and intra chip communications but also processing by Optical computers. There are two different types of optical computers, Electro-Optical Hybrid computers and Pure Optical computers.

In the past, the realization of optical computers faced many challenges because of the packaging difficulties associated with free-space coupling and holographic interconnects, and the difficulty in shrinking optical devices.

The technology that is enabling optical processors and optical computers is silicon photonics. Silicon photonics refers to the application of photonic systems using silicon as an optical medium and yet have low manufacturing costs as a result of using conventional silicon-integrated-circuit processes. The silicon material used in such photonic systems is designed with sub micrometer precision and is deployed into the microphotonic components. Silicon photonics combines technologies such as complementary metal oxide semiconductor (CMOS), micro-electro-mechanical systems (MEMS) and 3D Stacking.

Computing Using Photopolymers and Light

A new type of computing, developed by researchers at McMaster University, uses a single-component, light-responsive system to perform computing operations without relying on external electrical power or processors.

The computing operation is performed by shining patterned bands of light and shadow through different facets of a photopolymer cube and reading the results that emerge. A researcher shines layered stripes of light through the top and sides of a glass case holding the polymer, which is roughly the size of a dice used in a board game.

The polymer begins as a liquid and transforms to a gel in response to the light. A neutral carrier beam passes through the back of the cube to a camera that reads the results as refracted by the material in the cube. This material forms into filaments that react to the patterns of light to produce a new pattern that expresses the computing results. Data input as binary (dark-bright) strings generate a unique distribution of filament geometries, which corresponds to the result of a specific operation.

The researchers said that the material in the cube reads and reacts intuitively to the light, in much the same way a plant would turn to the sun. The system responds to low-intensity, incandescent light, which, similar to ambient sunlight, comprises all visible wavelengths and is spatially and temporally incoherent.

The researchers were able to use their new process to perform simple addition and subtraction questions. “We’re very excited to be able to do addition and subtraction this way, and we are thinking of ways to do other computational functions,” said professor Kalaichelvi Saravanamuttu.

The form of computing is highly localized, needs no power source, and operates completely within the visible spectrum. The technology, part of the branch of chemistry called nonlinear dynamics, uses materials designed and manufactured to produce specific reactions to light. “These are autonomous materials that respond to stimuli and do intelligent operations,” Saravanamuttu said.

According to the researchers, the working principles of this photopolymer are transferable to other nonlinear systems and incoherent sources including LEDs. The work represents a new form of computing that someday could be used for more complex functions, possibly organized along the structures of neural networks. The research was published in Nature Communications (https://doi.org/10.1038/s41467-019-10166-4).

Reconfigurable Silicon Photonic Circuits Provide Control of Light Patterns

Traditional spatial light modulators are based on liquid crystals or micromirrors and provide many independently controllable pixels. This technology has revolutionised optics in recent years, with many applications in imaging and holography, adaptive optics and wavefront shaping of light through opaque media. Researchers at the University of Southampton and the Institut d’Optique in Bordeaux, France, multimode interference (MMI) devices, which form a versatile class of integrated optical elements routinely used for splitting and recombining different signals on a chip. The geometry of the MMI predefines its characteristics at the fabrication stage.

They demonstrated that light could be routed between the ports of a multimode interference (MMI) power splitter with more than 97 percent total efficiency and negligible losses. The intricate interplay between many modes traveling through the MMI was dynamically controlled. A pattern of local perturbations, induced by femtosecond laser, was used to shape the transmitted light, demonstrating that all-optical wavefront shaping in integrated silicon-on-insulator photonics devices is possible.

By employing UV pulsed laser excitation to modify the spatial refractive index profile, the research team was able to maintain control of the optical transfer of telecommunication-wavelength light traveling through the device, thus allowing the functionality of the light to be redefined.

Photonics chip functionality is typically hardwired. Reconfigurable optical elements that would provide the ability to freely route light in a static silicon element offer an important building block for field-programmable photonics, the researchers said. “We have demonstrated a very general approach to beam shaping on a chip that provides a wide range of useful functionalities to integrated circuits,” said research fellow Roman Bruck. “The integrated spatial light modulator turns conventional silicon photonics components into versatile reconfigurable elements”.

Squeezing light into tiny channel promising for optical computing

The challenges in all optical processing is that Photons have tendency not to interact with each other. It has been found that passing light through optical crystals to cause certain nonlinear effects. Special nonlinear optical materials can make photons interact, but the effect is usually very weak. This means a long span of the material is needed to gradually accumulate an effect and make it useful.

Using nonlinear optics, Imperial College scientists were able to decrease the distance the light needed to travel by 10,000 times. So what would’ve needed centimeters of material now only requires micrometers of it. Note that one micrometer equals one millionth of a meter. This is the exact scale needed to allow optical computers to become viable.

They squeezed the light into a very small passageway, only about 25 nanometers wide. By doing so, the light became more intense as the photons within it were forced to merge over the short distance. The team achieved the effect by using a metal channel to focus the light inside a polymer previously investigated for use in solar panels. Metals are more efficient at focusing light than conventional transparent materials, and are also used to direct electrical signals. Imperial concludes that the new technology is therefore not only more efficient, but it could be integrated with current electronics.

Dr Michael Nielsen, from the Department of Physics at Imperial, said, “This research has ticked one of the boxes needed for optical computing. Because light does not easily interact with itself, information sent using light must be converted into an electronic signal, and then back into light. Our technology allows such processing to be achieved purely with light.”

As well as providing an important step towards optical computing, the team’s achievement potentially solves a longstanding problem in nonlinear optics. Since interacting light beams with different colours pass through a nonlinear optical material at different speeds, they can become out of step and the desired effect can be lost. He added, “In the new device, because the light travels such a short distance, it does not have time to become out of step. This eliminates the problem, and allows nonlinear optical devices to be more versatile in the type of optical processing that can be achieved.”

In 2019, Electro-optical device Developed by Advanced Nanoscale Engineering research group at the University of Oxford provides solution to faster computing

Scientists in Harish Bhaskaran’s Advanced Nanoscale Engineering research group at the University of Oxford, in collaboration with researchers at the universities of Münster and Exeter, the scientists have created an electro-optical device which, they say, “bridges the fields of optical and electronic computing; t his provides an elegant solution to achieving faster and more energy efficient memories and processors.”

The group, whose latest work has been published in Science Advances, comment, “Computing at the speed of light has been an enticing but elusive prospect, but with this development it’s now in tangible proximity. Using light to encode as well as transfer information enables these processes to occur at the ultimate speed limit — that of light.”

While as of recently, using light for certain processes has been experimentally demonstrated, a compact device to interface with the electronic architecture of traditional computers has been lacking. The incompatibility of electrical and light-based computing fundamentally stems from the different interaction volumes that electrons and photons operate in. Electrical chips need to be small to operate efficiently, whereas optical chips need to be large, as the wavelength of light is larger than that of electrons.

To overcome this problem, the scientists came up with a solution to confine light into nanoscopic dimensions. They created a design that allows them to compress light into a nano-sized volume through what is known as surface plasmon polariton. The dramatic size reduction in conjunction with the significantly increased energy density is what has allowed them to bridge the apparent incompatibility of photons and electrons for data storage and computation.

“We have demonstrated a nonvolatile nanoscale electro-optic device that enables both electrical and optical programming and readout using the synergetic combination of PCMs and nanoplasmonics. This is an unprecedented demonstration of an integrated, reversible, and nonvolatile phase-change memory cell that fully bridges the gap between electro-optic mixed-mode operations.

“This was enabled by using a plasmonic design that simultaneously reduces the footprint of the device, enhances light-matter interaction, and reduces the separation between electrical contacts, creating a compact and highly sensitive device. Our approach also enables a direct comparison of both optical and electrical read and write operations in a single device, demonstrating the relative merits and limitations of both.

More specifically, it was shown that by sending either electrical or optical signals, the state of a photo- and electro-sensitive material was transformed between two different states of molecular order. Furthermore, the state of this phase-transforming material was read out by either light or electronics thereby making the device the first electro-optical nanoscale memory cell with non-volatile characteristics. “This is a very promising path forward in computation and especially in fields where high processing efficiency is needed,” said Nikolaos Farmakidis, graduate student and co-first author.

“The nonvolatile nature of our platform provides an exciting outlook in the development of switchable and reconfigurable metadevices by means of optical or electrical stimuli, enabling novel approaches to switchable metamaterial-based optical components.”

The developers concluded, “We anticipate that a plethora of novel devices and platforms should arise in the coming years, which will capitalize on the bridge between the electrical and photonic domains [that we have demonstrated]. These devices potentially herald true device-level integration of hybrid optoelectronic computing platforms with in-memory computing and multilevel data storage, which is readily applicable to this work.”

Co-author Nathan Youngblood cemented, “This naturally includes artificial intelligence applications where in many occasions the needs for high-performance, low-power computing far exceeds our current capabilities. It is believed that interfacing light-based photonic computing with its electrical counterpart is the key to the next chapter in CMOS technologies.”‘A plethora of novel devices and platforms’

Ultrafast Photonic Computing Processor Uses Polarization, reported in June 2023

Researchers from the University of Oxford have developed a method using the polarization of light to maximize information storage density and computing performance using nanowires. Similar to how different wavelengths of light do not interact with one another — allowing fiber optic cables to carry multiple parallel streams of data — different light polarizations do not interact.

Each polarization can be used as an independent information channel, enabling more information to be stored in multiple channels, greatly enhancing information density. A photonic chip developed by the Universities of Oxford and Exeter exploits polarization properties to boost performance. Courtesy of June Sang Lee, University of Oxford.
A photonic chip developed by the universities of Oxford and Exeter exploits polarization properties to boost performance. Courtesy of June Sang Lee/University of Oxford.

“We all know that the advantage of photonics over electronics is that light is faster and more functional over large bandwidths,” said June Sang Lee, first author of the paper. “So, our aim was to fully harness such advantages of photonics combining with tunable materials to realize faster and denser information processing.”

In collaboration with C. David Wright, a professor of electronic engineering at the University of Exeter, the team developed a hybridized-active-dielectric nanowire, using a hybrid glass material that shows switchable material properties upon the illumination of optical pulses. Each nanowire demonstrates selective responses to a specific polarization direction, so information can be simultaneously processed using multiple polarizations in different directions.

The researchers reported that in using this concept they developed what they claim is the first photonic computing processor to use polarizations of light. The nanowires are modulated by nanosecond optical pulses that enable the chip to conduct computing operations over multiple polarization channels, leading to an enhancement in computing density by several orders compared to that of conventional electronic chips.