New Type of Computer Memory Can Increase Productivity and Reduce Energy Consumption
A new type of computer memory has been developed by a team of scientists from Cambridge University that can significantly increase productivity and reduce the energy consumption of communication technologies. By 2030, communication technologies are expected to consume almost a third of world electricity. The new memory processes data similarly to synapses in the human brain and can store and process information in one place. The results of the study have been published in the journal Science Advances.
The world’s demand for energy is growing every year due to the development of artificial intelligence, the internet, algorithms and data based on data. This is attributed to the disadvantages of existing computer memory technologies that require the movement of data between memory and processing, which consumes both energy and time.
The team of scientists proposed a possible solution in a memory resistance, capable of a continuous range of states rather than the traditional memory which has only two states: one or zero. With such memory, a device can accommodate 10 to 100 times more information on one device.
Scientists created a device prototype based on Gafnia oxide, material that is already used in the semiconductor industry. They discovered the addition of barium to thin films of Oxide Gafnia leads to the formation of unusual structures that allow electrons to pass and change the electrical resistance of the material. The structures resemble synapses in the brain that store and process information in one place, making it promising for the rapidly developing areas of artificial intelligence and machine learning.
The Cambridge enterprise, a commercial unit of the university, has filed a patent for the technology. The scientists are now focusing on collaboration with the industry to research further possibilities for the application of the materials. Gafnia oxide being a material already used in the semiconductor industry means it is easy to integrate with the existing production processes.