Daily Archives: February 25, 2020

Archiving data in DNA may be possible

Archiving data in DNA: scientists seem to believe it more and more, and the approval of a new $25 million funding in the United States for a new project confirms it. The new project will be born in the context of the MIST (Molecular Information Storage) program and has been set up by the Intelligence Advanced Research Projects Activity (IARPA). The project should foster the development of new scalable molecular storage techniques based on DNA.

The project will be led by the Georgia Tech Research Institute (GTRI) and foresees the use of DNA for the storage of digital data, storage that can eventually scale in the exabyte regime without renouncing the reduced physical space requirements of this technology. Currently the technology to store data in DNA already exists but further progress is needed to make this practical, even at a commercial level, and competitive with other data storage techniques, including magnetic tape and optical discs.

“The goal is to significantly reduce the size, weight and power required for storage data storage,” specifies Alexa Harter, director of the CIPHER (Cybersecurity, Information Protection and Hardware Evaluation Research) laboratory. “What would require acres in a data farm today could be stored in a device the size of a tabletop. We want to significantly improve all types of metrics for long-term data storage.”

Data storage technology in DNA provides such a compact type of storage device that one million terabyte hard drives could be stored in a volume the size of a sugar cube, as Nicholas Guise, a researcher at GTRI, explains. Moreover, data, unlike most other data storage digitization technologies, can be stored for hundreds of years if not thousands of years once it is inserted into DNA.

It is thought, however, that this technology can, at least initially, only be used for very important data that must be stored indefinitely but is rarely accessed. This is because, at least for the time being, the time needed to read and decode the data is still a little too long and therefore not suitable for all those data that must be accessed often and quickly.