Chinese semiconductor industry

Status
Not open for further replies.

tokenanalyst

Brigadier
Registered Member

A computing-in-memory macro based on three-dimensional resistive random-access memory​


Non-volatile computing-in-memory macros that are based on two-dimensional arrays of memristors are of use in the development of artificial intelligence edge devices. Scaling such systems to three-dimensional arrays could provide higher parallelism, capacity and density for the necessary vector–matrix multiplication operations. However, scaling to three dimensions is challenging due to manufacturing and device variability issues. Here we report a two-kilobit non-volatile computing-in-memory macro that is based on a three-dimensional vertical resistive random-access memory fabricated using a 55 nm complementary metal–oxide–semiconductor process. Our macro can perform 3D vector–matrix multiplication operations with an energy efficiency of 8.32 tera-operations per second per watt when the input, weight and output data are 8, 9 and 22 bits, respectively, and the bit density is 58.2 bit µm–2. We show that the macro offers more accurate brain MRI edge detection and improved inference accuracy on the CIFAR-10 dataset than conventional methods.

1660780151260.png
Please, Log in or Register to view URLs content!

The Institute of Microelectronics of the Chinese Academy of Sciences has made important progress in the field of three-dimensional storage and computing integrated chips​


Please, Log in or Register to view URLs content!
 

tokenanalyst

Brigadier
Registered Member
how does this compare to current prevailing practices when it comes to memory?
From my understanding current memory is more general data storage for computational processes of other circuits like the cache of a CPU or a pen drive. While this more about artificial intelligence, neuromorphic computing to be more exact, so this devices not only store data but also storage calculation stages and do calculations on their own, kinda like the neurons in a brain.
 

latenlazy

Brigadier
From my understanding current memory is more general data storage for computational processes of other circuits like the cache of a CPU or a pen drive. While this more about artificial intelligence, neuromorphic computing to be more exact, so this devices not only store data but also storage calculation stages and do calculations on their own, kinda like the neurons in a brain.
So vectorized data as opposed to values?
 
Status
Not open for further replies.
Top