During the early months of 2022, a small startup based in Israel named CogniFiber launched their first round of seed funding. The startup raised 6 million dollars with the ambitious goal of reimagining how modern computing is processed. Could this be a dramatic shift in the computing industry? As rack computer specialists, we delve into this discussion explaining the potential performance benefits involved with photonic computing.
Who Are CogniFiber?
Little is known about the new start-up CogniFiber. The company states that photonic computing can provide the proper infrastructure and implementation for artificial intelligence, as AI infrastructure today is either too expensive, too slow or consumes too much power. CogniFiber is planning to achieve it’s promise through the use of DeepLight technology.
What Is DeepLight Technology?
DeepLight technology uses light within silicon chips, these photonic chips harness the tremendous capabilities of optical communication platforms together with stable, robust and resilient fibre optics. This generates phase-independent computing modules that are scalable, fault-proof, easily manufactured, extremely fast and easily integrated into existing communication platforms. These advantages would make DeepLight technology a superior solution within the global information economy industry. The first full system prototype is expected in April 2022. The fibre-optic cables are said to be capable of performing complex algorithms within the fibres themselves before the signal hits the terminal. This technology will primarily be used in artificial intelligence-based industries and applications.
What Is Photonic Computing?
As the name suggests, photonic computing involves a computer system using light as opposed to electrical transistors. Although photonic computing has been around since the 70s, it could have huge potential to revolutionise computing and machine learning in the near future. Photons are effectively weightless and incredibly fast, generated using diodes or lasers. Current computing systems are reliant on electrical currents, and will soon reach their physical limit in terms of speed and efficiency. To meet the demands of the next generations of computer systems for; artificial intelligence and autonomous vehicles. New technological innovations such as photonic computing could be the perfect solution.
CogniFiber Vs Nvidia?
Nvidia’s most powerful AI system came with the launch of the Nvidia DGX A100. CogniFiber’s CEO stated that the DeepLight chips will be 100x more powerful than Nvidia’s current strongest system, being able to reach 500 million tasks per second. Scaling performance could also improve dramatically on the CogniFiber chips by using multiple wavelengths, the best part about these upcoming chips is their efficiency. Photonic computing only consuming a mere fraction of the power needed to run more traditional systems.
What Does This All Mean For G2?
Nothing for now; But we at G2 always have an eye on technological innovation. In the next 10-20 Years we could see our computers move from electron to photon-based. This has the potential to drastically impact heat generation and therefore noise and power requirements. Allowing for more powerful systems per U of rack space.
Bespoke Computer Rack – Get In Touch
It can be difficult to find the perfect computing solution for your business needs. That’s why our experts build your bespoke computer systems equipped with a suitable computer rack for wall mounting if needed. Get in touch with the experts here, or give us a call on 01252 711 114. We will be more than happy to help you.
