Data Storage Format
In the view of the employees of the Backblaze, who had not had a long-term experiment with hard drives, some HDDs could store information without damage for about 10 years, not so much.
Unfortunately, modern memorial devices cannot serve forever. That is why researchers are trying to find a way to keep data as long as possible, ideally for several million years.
Permanent reservoir
Researchers from the Swiss High Technical School of Zurich believe that the answer may lie in DNA. According to Peter Schandballt (Peter Shandbolt), who had written material on the prospects for CNN data storage, the DNA structure was complex and at the same time compact, so that " one DNA gram is theoretically capable of supplying all data from large companies such as Google and Facebook, even with a reserve " .In the technical language, one gram of DNA can store up to 455 eqsabaytes of information where one exzabeit equals a billion gigabaytes.
These values are assigned to certain chemical components, monomers which, by chemical methods, are chained to form polymers. To read the recorded information, it's enough to use the mass spectrometer, a DNA sequence reader.
Knowing double DNA spiral
Figures
Information on the genome of living organisms can be stored for a long time in the occults: this has been encrypted by the genome of a polar bear who lived 110,000 years ago and the horse's genome, which is 700,000 years old.According to the Swiss High Technical School of Zurich (ETH Zurich), they want to " share DNA storage opportunities " volume of data with the stability of DNA found in fossil remains. "
" We have found unusual ways of working with DNA to make it stable enough, and it is reported by Robert Grass, Professor of Chemical and Applied Biological Sciences ETH Zurich, in an interview with CNN. - We wanted to link DNA, namely the high data storage capacity and its archaeological characteristics. "Thus, the information storage process will look like: DNA in a glass casing (inert and neutral) is stored at low temperature and dry (water and oxygen are negatively affected by DNA).
Under such circumstances, information in her thighs can remain intact for several thousand years.
" We have been able to make sure that the DNA we created with the information we stored is falling at the same speed as the DNA found in the fossils. If this goes on, we will soon be able to keep data for up to a million years.”Initial results
In order to verify how long the data can be stored in DNA, scientists have encrypted 83 kilobytes of data.The Swiss Federal Statute of 1291 and the Al-Arhamed Palampsest served as a material. The selection of these documents, according to scientists, shows not only the potential applicability of the method, but also its historical importance. ETH Zurich estimates that these data will remain constant from 10,000 to million years (if DNA is frozen).
With regard to the number of information, George Church (George Church), a professor of genetics from the Visa Institute (Garvard), set all previous records on the number of DNA data recorded on his knees, he and his colleagues managed to obtain a 5, 5 petabits (approximately 700 terabite) on the cubic millimetre of liquid, he created 70 billion copies of his genetic book. For modern electronics, such indicators are not yet available.
However, as always, there are difficulties in dealing with DNA repositories, such as the cost of encrypting data. However, researchers are confident that, as medical technologies evolve, they will decline.
" The cost of deciphering the human genome amounted to several million dollars. You can spend it for a few hundred dollars today, claiming Grass. New technologies for medical and genome analysis can soon be introduced into the world IT. "But the biggest obstacle that still prevents the use of DNA to store information in practice is time.
Even with modern encryption technologies, reading DNA molecules takes a lot of hours, a little more in order than reading a conventional computer file. That's why this type of storage is not suitable for frequently used data.