Experts Enhance Light-Carrying Chips for Computers to Enable Highly Parallel Data Processing

Researchers were able to enhance light-based chipsets for computers! Right now, people are demanding more powerful computing abilities for applications in artificial intelligence.

These include speech and pattern recognition, or self-driving vehicles. The demands often exceed the capacities of conventional computer processors.

Light-Carrying Chipsets Advance Machine Learning! How Effective Can They Enhance Computers and Laptops Today?
A "Mistral" supercomputer, installed in 2016, at the German Climate Computing Center (DKRZ, or Deutsches Klimarechenzentrum) on June 7, 2017 in Hamburg, Germany. The DKRZ provides HPC (high performance computing) and associated services for climate research institutes in Germany. Its high performance computer and storage systems have been specifically selected with respect to climate and Earth system modeling. With a total of 100,000 processor cores, Mistral has a peak performance of 3.6 PetaFLOPS. With a capacity of 54 PBytes, its parallel file system is currently one of the largest in the world. Photo by Morris MacMatzen/Getty Images

And now, the researchers at the University of Munster are creating new methods and process architectures. They can cope with these tasks more efficiently. These experts revealed what they call photonic processors.

These chips process by means of light, and can process information much more rapidly and in parallel. Electronic chips nowadays are not capable of doing this kind of task.

Combining photonic structures with PCMs

According to Science Codex's latest report, the scientists and researchers combined the phase-change materials or PCMs with the photonic structures. They used PCMs as energy-efficient storage elements.

These materials are commonly integrated into BluRay discs or DVDs in optical data storage. However, the new chips are now unable to store and preserve matrix elements without the need for an energy supply.

To solve this issue, the scientists used a chip-based frequency comb as a light source to carry out matrix multiplications on multiple data sets in parallel.

Light-Carrying Chipsets Advance Machine Learning! How Effective Can They Enhance Computers and Laptops Today?
LEIPZIG, GERMANY - FEBRUARY 15: A participant sits at a computer monitor to play a video game at the 2019 DreamHack video gaming festival on February 15, 2019 in Leipzig, Germany. The three-day event brings together gaming enthusiasts mainly from German-speaking countries for events including eSports tournaments, cosplay contests and a LAN party with 1,500 participants. Photo by Jens Schlueter/Getty Images

Munter physicists explained that a frequency comb provides different optical wavelengths. These are processed independently of one another in the same photonic chip.

The innovation allowed the new chipsets to enable highly parallel data processing. They can now do this by calculating on all wavelengths continuously.

Top chip manufacturers today

As of the moment, almost all giant chipset manufacturers are still developing and focusing on electronic chips. Although this technology has certain limitations, they are still able to integrate great features that can really improve the experience of many users. ElectronicsAndYou's are listed as the top manufacturers right now.

  • Samsung Electronics
  • SK Hynix
  • Intel Corporation
  • Micron Technology
  • TSMC
  • Avago Technologies
  • Nvidia
  • Toshiba
  • Qualcomm
  • Texas Instruments

For more news updates about new chipsets, always keep your tabs open here at TechTimes.

This article is owned by TechTimes.

Written by: Giuliano de Leon.

ⓒ 2024 TECHTIMES.com All rights reserved. Do not reproduce without permission.
Join the Discussion
Real Time Analytics