We are excited to bring Transform 2022 back to life on 19th July and virtually 20th July – 3rd August. Join AI and data leaders for sensible conversations and exciting networking opportunities. Learn more about Transform 2022
Cognifiber, a company that manufactures stand-alone Photonic Artificial Intelligence (AI) accelerators for IoT analytics, has set its sights on a new goal – to replace conventional silicon with glass. By minimizing its in-fiber photonics, the startup claims it can bring server-grade capabilities to the edge.
The startup focuses on photonic computing, which goes a step further than most current commercial applications that use photonics to communicate only in racks in the data center. Cognifiber has now announced the “first of its kind” glass-based photonic chip specifically designed for Edge computing.
The company claims that it is able to minimize its previous design to a “fraction” of size. The “rack-size” system has been reduced to a 4U server with a height of 18cm, making it possible to deploy it in offices, leading to lower cost.
Photonic glass chips as solution
Edge Computing aims to increase edge computing capabilities, which will prevent IoT devices from relying on the cloud without affecting their performance, while reducing response time (also known as latency) for real-time operation. . As some examples of what Cognifiber envisioned its solution, it cites the operation of a fleet of smart meters and home assistants, connected vehicles, automated trains and drones.
“Combining photonic glass chips promotes our Edge solution to bring fast AI and machine learning locally to Edge devices, limited in their capabilities and power allowances,” said Ziv Zalewski, co-founder and CTO of Cognifiber.
The company claims that it can increase its in-fiber processing performance 100 times and reduce AI / ML training costs by 80%. Cognifiber’s technology is based on the exploitation of crossstock – which, traditionally, is to be avoided – between fibers. This field allows for programmable and reconfigurable interactions such as programmable gate arrays (FPGAs). Supports both linear and nonlinear operations, making the technology suitable for AI.
The Israeli startup, formerly part of Intel’s Ingenuity Partner Program, announced શ્રેણી 6 million in Category A last month. It has demonstrated “successful” proof of concept during its seed phase and aims to bring the complete system to market by the end of 2023.
Venturebeat’s mission Transformative Enterprise is about to become a digital town square for technology decision makers to gain knowledge about technology and transactions. Learn more about membership.