In traditional computers, the moving of data back and forth between RAM and CPU makes the process slower and consumes more energy. Tech giant IBM has announced that it has created an unsupervised machine-learning algorithm that runs on one million phase change memory devices (PCM). PCM is a type of computer RAM that stores data by changing the state of the matter.
IBM’s algorithm was demonstrated running on one million PCM devices. Compared to our classical machines, this innovation is expected to bring 200 times improvements in both speed and energy efficiency.
As a result, this technology could turn out to be suitable for “enabling ultra-dense, low-power, and massively-parallel computing systems for applications in AI.”
Here, the PCM devices being talked about had been made using a germanium antimony telluride alloy, stacked between two electrodes. When a tiny electric current is applied to the material, due to heating, its state changes from amorphous to crystalline.
“The result of the computation is also stored in the memory devices, and in this sense the concept is loosely inspired by how the brain computes,” said Dr. Abu Sebastian, a scientist, and IBM Research.
The further details on IBM’s current efforts in-memory computing can be found in a research paper published in Nature Communications.