Storing massive amounts of data poses increasingly serious challenges, such as rising costs, increased complexity in data management, potential security breaches, and difficulties in quickly accessing or analyzing the collected data.
It also poses significant environmental concerns due to energy consumption, but startup Cerabyte has a solution to this problem by writing data onto ceramic nanolayers, a method the company likens to the ancient Egyptians’ practice of chiseling hieroglyphs into rocks.
Cerabyte’s prototype consists of one reading rack and several library racks. The storage medium takes the form of a thin, single-access square glass plate (also called a platter or carrier) with a dark nanoscale ceramic layer stored in cartridges in a robotic library. Data is engraved in QR code-like patterns on the ceramic using a femtosecond laser pulse.
Wait a minute
A new video of the system in action has surfaced on YouTube, which you can watch below. Previous reports suggested that the cartridges could store up to 10,000 TB each, but that is not the case.
A Cerabyte spokesperson told us: “This is a data density that is justified by experimentation and part of our roadmap as we move from laser beam to particle beam, but not within the next five years.”
They further explained: “For our prototype, we used only commercial off-the-shelf components to quickly demonstrate the feasibility of the technology. In general, we use depreciated semiconductor fabrication tools. This means the setup builds on today on maskless lithography Today’s particle beam technology (which costs more than 100 million euros) will be the basis for write/read processes in ten years and the costs will be well below $1/TB/. The main challenge of data storage today is the flattening of development curves, and therefore also of performance and costs.”
Although 10,000 TB cartridges may not be on our doorstep, the spokesperson did give us some good news. “We are currently improving data density by using a UV laser and have therefore increased data density three times compared to our initial setup. This was a major breakthrough in being able to remove the data layer.”