Beijing time on April 12th, according to the website Computerworld reports, Microsoft and University of Washington researchers have demonstrated the use of synthetic DNA as a data storage media technology. The researchers said that if the technology is ripe for mainstream applications, equipped with modern storage density with the highest storage equipment, Wal Mart supermarket like the size of the data center, it may "downsizing" to the size of a lump of sugar. "We believe that consider DNA as a storage medium, to explore the related system design of the time is ripe".
DNA storage density amazing
The research team successfully coded the data of 4 picture files into the nucleotide sequence of the synthetic DNA fragment. More importantly, they can reverse the process of retrieving the correct nucleotide sequence from the larger DNA pool, rebuilding the image, and without losing 1 bytes of information. Another trial involved coding and reading of video files.
DNA storage technology
Rapid progress in the research of DNA storage technology
In a paper published in nature, the researchers said that the coding method makes about a cup of DNA capable of storing at least 100 million hours of HD video.
Data stored in the DNA chain can be stored for several thousand years, according to British researchers.
In the DNA storage technology, the DNA is relatively simple to read, the main obstacle is to write DNA. DNA storage technology there are two difficulties: first, the current method can only synthesize short chain DNA; second, write and read DNA are prone to error.
Microsoft and the University of Washington researchers said they have developed "a novel method, the data in the" 1 "and" 0 "string is converted into DNA sequences of four bases: adenine (a), guanine (g), cytosine (c) and thymine (T).
In order to access the stored data, the researchers coded the address information in the DNA sequence, which is equivalent to the postal code and street address. Polymerase chain reaction (PCR) technique to help researchers more easily identify the "zip code" they need to find".
The researchers then used DNA sequencing technology, read data, and through the use of street address to collate data, the data is restored to the original video, images or documents.
Researchers said, DNA is an attractive potential data storage media, in theory, its storage density is 8 orders of magnitude higher than the tape, one can hold in the hands of tape storage capacity of up to 185TB.
Microsoft and University of Washington researchers also confirmed the artificial synthesis of DNA's "longevity", saying that even in harsh environments its half-life is also more than 500 years, the tape storage time for 10-30 years, the hard drive for 3-5 years.
Not perfect enough
Researchers in the United States have highlighted the urgency of increasing storage density. Market research firm IDC and storage equipment manufacturers of EMC in the digital universe research papers said by 2020, including in the global computer, historical archives, movies, photos, enterprise systems and mobile devices in the amount of data will reach 44 trillion GB, "10 times in 2013. Although not all the information needs to be saved, the world is growing faster than the storage capacity ".
To enter the commercial phase, DNA storage systems have some problems to be solved. First, DNA synthesis and sequencing is still far from perfect, a key part of the DNA storage system is to develop an appropriate coding technology to improve fault tolerance by increasing redundancy.
Secondly, the random access data in DNA storage system is still a problem, and the read delay is much longer than the write delay. The current technology can only batch read data, even if only from the storage system to access a byte of information, the system must also be for the entire DNA pool sequencing and decoding.
The data is encoded as a sequence of DNA
Researchers have put forward a method to improve the random data access, that is, the use of PCR only to amplify the hope to read the data, and the corresponding DNA sequence sequencing. This method not only improves the speed of data reading, but also does not need to sequence the entire DNA pool.
The said, "this is an example for us to learn from nature in the data store."