A “revolution”. Microsoft has announced to invest in a team of researchers to design tomorrow’s fiber, through its subsidiary Lumenisity, acquired in 2022. Connected with the English University of Southampton, it aims to offer much higher flows than the current fiber, while limiting the losses despite the distances that it can travel.
This fiber “with hollow nucleus” is opposed, in a way, with fiber “with a conventional solid nucleus. The latter uses light through a glass cable, which allows you to approach a maximum theoretical speed. Hollow nucleus fiber uses ultra thin glass membranes to maintain the signal in the most difficult conditions through air, reducing energy losses, explaining to researchers, while increasing flows up to 45%.
A longer and more efficient fiber
His advances, published in Nature magazine, in any case are conclusive enough for Microsoft to invest money in the project, especially for its Azure servers.
The company took the opportunity to announce to the registration that the signal loss was minimal, the first in the case of this technology made to be deployed in hundreds, even thousands, of kilometers. For example, traditional fiber can reach up to 0.14 Loss of decibel per kilometer. The finding of researchers on hollow nucleus loses 0.091 decibels per kilometer. Which means that the signal suffers less loss.
We better understand why this investment, while Microsoft “has an urgent need for fiber” and the company seeks to boost its cloud solutions, including a version of Windows does not require a conventional computer.
Until now, the hollow nucleus fiber suffered a loss of 1DB per kilometer, with the consequence an inevitable decrease in flow rates and, therefore, a bad potential connection passed the first ten kilometers.
If this materializes, the researchers explain that they could even become “a revolution in optical communications.” Especially because in the key, it is the possibility of being able to connect white or very remote areas that allow operators to cover small inhabited places.
The fact is that such “revolution” should not take place immediately. In the comments dating from the financing of Microsoft, Franciesco Poletti, Lumenisity, did not see that such technology arrived at the data centers before “at least five years.” He also highlighted the difficulty of creating “a global standard.”
However, the race is launched: China also works on the subject and has produced similar advances.
Source: BFM TV
