Microsoft has developed a groundbreaking new approach to data storage through its cutting edge effort, Project Silica. This new project laser etches hundreds of terabytes of data into glass. It’s going to change the way we preserve our data for future generations. This technology allows you to pack huge volumes of data into a small package. It can potentially be for thousands of years!
>Through Project Silica, Microsoft encodes data on quartz using minuscule three-dimensional units called voxels. Each voxel is roughly 0.5 micrometers in each dimension, with about 6 micrometers between adjacent voxels. This very specific setup enables Microsoft to pack in a staggering 4.84 terabytes of storage. They do this in a small glass chip that’s only 12 square centimeters and 2 millimeters deep. The storage space could store the equivalent of about 2 million printed books or 5,000 ultra-high-definition 4K movies.
Microsoft is able to accomplish this incredible trick thanks to the use of femtosecond lasers. Each of these lasers produces extremely high-power pulses, lasting only several quadrillionths of a second. The extreme accuracy of the technology means data can be written quickly, up to 25.6 megabits per second with just one beam! Another hurdle to its scalability is the high cost of the femtosecond lasers themselves. In short, Microsoft will need to make some deep cost cuts on production costs while still maintaining the efficacy of the technology.
The extreme durability of glass plays a huge role in its future as a long-term data storage medium. Archive worthy Glass is naturally impervious to moisture, extreme temperatures, and electromagnetic interference making it an easy choice for potential archival uses. According to Richard Black, a key figure in the development of Project Silica, “It’s designed for data you want to write once and preserve for a very long time.” This new technology is one of amazing resilience. Data stored with it would still be readable for more than 10,000 years at 290 °C, and even longer at room temperature.
To support this project, Microsoft has created two types of voxels, phase voxels and birefringent voxels. Each type only needs one laser pulse to record data. Black goes on to underscore how this design increases productivity. It further reduces the amount of power needed from the laser in order to store data. He notes, “This significantly reduces the power required from the laser to store data, and it does not require the laser focus to alternate between staying in the same place to deliver multiple pulses and movement to the next location.”
The implications of Project Silica go much further than just data storage. Black emphasizes its potential in preserving critical information for centuries: “These are archival—anywhere data must survive for centuries such as national libraries, scientific data, or cultural records.” As every organization starts to look for more dependable ways to preserve their data for the long haul, Project Silica might become a game-changing alternative.
How broadly feasible implementation around the country would be is an open question for further discussion and study. The technology has great potential. Yet barriers such as the exorbitant costs of femtosecond lasers may prevent its translation into other sectors. Microsoft knows about these challenges and is taking action to address them by seeking ways to make this powerful storage solution more accessible.

