A new method of arranging advanced computer components—memristors— on a chip could enable them to be used in traditional computing, cutting energy consumption a hundredfold. This would serve to improve performance in low-power applications, such as smartphones or wearables, as well as making more efficient supercomputers.
“Historically, the semiconductor industry has improved performance by making devices faster. But although the processors and memories are very fast, they can’t be efficient because they have to wait for data to come in and out,” said Wei Lu, professor of electrical and computer engineering at the University of Michigan.
These memristors—a portmanteau of memory and resistor—may be the answer, as they can be programmed to have different resistance states. The elements enable memory and processing in the same device, which cuts out the data transfer bottleneck experienced by conventional computers.
Computers with these new devices, which the researchers call “memory-processing units,” could be particularly useful for implementing machine learning and artificial intelligence algorithms. They are also well suited to tasks that are based on matrix operations, such as simulations used for weather prediction.
To learn more about this project, click here.