Scientists at Microsoft recently made a revolutionary advancement in data storage with the help of femtosecond lasers. This is a huge step forward for the company’s Project Silica. This project will potentially create a new, permanent data storage medium that uses glass to encode data for survival across millennia. So in addition to onboarding media inside warehouses, it can transform the way we think about archiving content for many generations.
Microsoft’s approach involved using femtosecond lasers to write 4.84 terabytes worth of data into tiny strands of glass. This chip is 12 square centimeters and 2 millimeters deep, made possible by the high-power laser pulses that last only quadrillionths of a second. This incredible capacity holds the equivalent of roughly 2 million printed books or 5,000 ultra-high-definition 4K movies.
This glass chip’s strength is especially impressive. It has the unique ability to store data for more than 10,000 years at 290 °C. At hotter ambient temps, it probably doesn’t last nearly that long. Richard Black, a principal scientist in the project, told the BBC about the benefits of storing data in glass. He stated, “At room temperature, glass is effectively a solid and does not flow on any meaningful timescale.” This characteristic imbues glass with remarkable durability against extreme humidity, thermal cycling, and electromagnetic interference. In that case, it’s the perfect fit for archival storage.
Microsoft’s Project Silica employs a sophisticated method of encoding data into the glass using two types of voxels: phase voxels and birefringent voxels. Each voxel is around 0.5 micrometers in each dimension and on the order of 6 micrometers apart from neighboring voxels. This intricate architecture allows the CI system to deliver remarkable writing speeds. With a single laser beam, it can achieve speeds of 25.6 megabits per second.
The technology is full of potential, but it is failing to overcome its shortcomings. The prohibitive expense of femtosecond lasers may pose serious threats to the feasibility and scalability of Project Silica. The International Data Corporation sees the world creating some 394 zettabytes of data by 2028. This dynamic increase underscores the complexity and immediacy of the need for durable, stable, reliable and long-term data storage platforms.
Richard Black focused on the more exciting potential applications of this technology. He noted that it’s particularly ideal for archival use, making sure that data can last hundreds of years. He remarked, “It’s designed for data you want to write once and preserve for a very long time,” highlighting its relevance for national libraries, scientific data, or cultural records.