Data Storage in a Petri Dish RRS feed

  • Question

  • Researchers are experimenting with different ways of storing data on incredibly small scales. A recent project at Keiko University for Advanced Biosciences showed the ability to encode data in bacteria and have it passed on through generations in an organism's DNA. The idea behind the Department of Energy project is to allow data to survive a nuclear disaster. Reading the data proved more difficult than encoding it and researchers believe it could be decades before an effective data-retrieval technique is developed. Another technique, storage of a single bit on an atom, was achieved in 2002 at the University of Wisconsin, when a small amount of gold on the surface of silicon was made to spur the self-assembly of tracks on the nanometer level. In 1959, Richard Feynman imagined such atomic memory with five atoms separating each bit, and the 2002 results showed that the minimum empty space around each bit is four atoms. Reading data encoded to atoms is relatively easy, requiring only a simple line scan with a scanning tunneling electron microscope. However, writing the data is more difficult and time consuming, since much less energy can be extracted from such small bits during readout. Where the DNA storage technique requires 32 atoms to store a single bit, this silicon atom memory requires 20 atoms per bit. Finally, University of Arizona researchers are working with microelectronic arms to read and write data in clusters of molecules on nanotech organic film. The system uses cantilevers that write data by injecting a current in the film that changes the electric resistance at the point of contact. The team believes 1 million of these cantilevers could be make to run in parallel, as they are simpler devices than the transistors that run parallel by the millions in today's processors
    Saturday, April 7, 2007 5:21 AM