Microsoft has announced the completion of research on Project Silica, which enables data archiving on glass plates for up to 10,000 years. The technology uses femtosecond lasers to write information in three dimensions, creating a storage medium resistant to water, high temperatures, and shocks. However, the main challenge remains the extremely long write time, which precludes commercial use in the near future.

Millennia of Data Durability

The glass media from Project Silica are designed to survive intact for 10,000 years, being resistant to factors that destroy traditional disks.

Femtosecond Laser Writing

Data is etched into the glass structure using ultrashort laser pulses, creating a three-dimensional grid of microscopic indentations.

High Capacity and Slow Write Speed

A single plate can hold 4.8 terabytes of data, but fully writing it currently takes over 18 days.

Conclusion of Research Phase

Microsoft concluded the project as basic research, with no announcement of a quick commercial launch for the technology.

Microsoft has formally concluded its multi-year research project named Silica, which aimed to create a method for archiving digital data for thousands of years. The technology is based on writing information into rectangular plates of borosilicate glass about 2 millimeters thick. Using a femtosecond laser, data is etched into the material's structure, creating a three-dimensional grid of microscopic structures called voxels. Reading is performed by illuminating the medium with polarized light and analyzing the image with machine learning algorithms, while the medium itself remains stationary.

Attempts to create data storage media capable of surviving centuries or millennia date back to ancient times, from clay tablets to papyrus. In the digital age, the challenge is particularly pressing because traditional media – hard drives, magnetic tapes, CDs – degrade within just a few decades, threatening the loss of vast cultural and knowledge resources.

A key advantage of glass media is their exceptional durability. According to a publication in the journal "Nature", they are resistant to water, magnetic fields, extreme temperatures, and physical shocks. These properties allow them to be stored in passive, non-climatized archives, radically reducing the costs of long-term storage compared to maintaining server rooms. The project's lead researcher, Antje Weisheit, emphasized the significance of this technology for preserving civilization's heritage.

„This is an exciting step towards solving one of the biggest challenges of our time: how to preserve our digital heritage for future generations.” — Antje Weisheit

Despite promising durability parameters, the technology faces serious limitations that block its commercial implementation. The fundamental barrier is the extremely slow write process. To fill the entire available capacity of the medium, which is 4,8 TB — the capacity of a glass plate, currently requires over 18 days of continuous laser operation. For comparison, modern disk arrays can write a similar amount of data in a few hours. Furthermore, the write and read devices are large, complex, and very expensive, making them impractical outside specialized research centers.

Long-Term Storage Media Comparison: Predicted durability: 30-50 years (LTO tape) → 10,000 years (glass); Time to write 5 TB: ~3 hours (SSD drives) → ~19 days (glass); Environmental resistance: Requires air conditioning → Resistant to water, magnetism, shocks

Microsoft's announcement did not include plans for further intensive development work or a product launch timeline. The company presented Project Silica as the successful conclusion of a basic research phase that expanded knowledge about data storage possibilities, but without a direct path to commercialization. The communication emphasized that the technology will remain in the research sphere and may form the basis for future, more practical solutions, perhaps in collaboration with national archives or cultural institutions.

Mentioned People

  • Antje Weisheit — Lead researcher of the Silica project at Microsoft Research