Memory-Processing Unit to Cut Energy Consumption by a Factor of 100


Arranging computer components on-chip can improve performance in low power environments such as smartphones and supercomputers

A new way of arranging advanced computer components called memristors on a chip could enable them to be used for general computing which can cut energy consumption by a factor of 100. To combat the issue of processors and low efficiency of memory, memristors may be the answer as they can be programmed to have different resistance states, means that they store information as resistance levels.

These circuit elements enable memory and processing in the same device, cutting out the data transfer bottleneck experienced by conventional computers in which the memory is separate from the processor.

Memristors to implement AI and ML algorithms

Computers with these new blocks also called memory-processing units can be 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. The simplest mathematical matrices, akin to tables with rows and columns of numbers can map directly onto the grid of memristors.

Once the memristors are set to represent the numbers then operations that multiply and sum the rows and columns can be taken care of simultaneously, with a set of voltage pulses along the rows. The current measured at the end of each column contains the answers. Hence, the addition and multiplication in a processor can be done in one step.

Unlike ordinary bits which are 1 or 0, memristors can have resistances that are on a continuum. Some applications, such as computing that mimics the brain (neuromorphic) take advantage of the analogue nature of memristors.

But they got around this problem by digitising the current outputs by defining the current ranges as specific bit values 0 or 1. The team was also able to map large mathematical problems into smaller blocks within the array thus improving the efficiency and flexibility of the system.

Memristor as one block of future system

The researchers choose to solve partial differential equations as a test for a 32×32 memristor array which they imagine as just one block of a future system. These equations including those behind weather forecasting, underpin many problems of science and engineering but are challenging to solve.

The difficulty comes from the complicated forms and multiple variables needed to model physical phenomena. So to solve partial differential equations, it may require supercomputers. These problems often involve large matrices of data, so the memory-processor communication is solved with a memristor array. The equations they used in their demonstration simulated a plasma reactor such as those used for integrated circuit fabrication.