A Data Operation Method Based on Nor Flash Module

Abstract

The invention discloses a data control method based on a NOR Flash module. The NOR Flash module includes at least one storage unit; each storage unit includes a gate terminal, a drain terminal and a source terminal; step S100 obtains initial parameters of the storage unit; step S200 according to The initial parameter of the storage unit is to obtain the threshold voltage of the storage unit; step S300 sends input data parameters to the data receiving end of the storage unit; the data sending end includes a gate terminal and a drain terminal; step S400 according to the threshold voltage of the storage unit and input data parameters, Obtain the output data parameters of the storage unit by using a preset algorithm; step S500 compare the preset output result parameters with the output data parameters, and process the storage unit according to the comparison result. The invention realizes the multiplication and addition operation based on the NOR Flash storage unit, improves the calculation speed, and further reduces the power consumption.

Description

technical field [0001] The invention relates to the field of semiconductor integrated circuits, in particular to a data operation method based on a NOR Flash module. Background technique [0002] As a non-volatile storage technology, flash memory is widely based, and the usual implementation is to store charges in field effect transistors (Field Effect Transistor, FET) based on floating gate (Floating Gate) or charge trap structure (Charge Trap), forming a storage Units, storage cells form an array for storing large amounts of data. According to the difference in array structure, flash memory can be divided into two types: NAND type and NOR type. The storage cells of NOR flash memory (NOR Flash) are connected in parallel. Existing flash memory It is used for storage rather than computation. [0003] The currently commonly used computing systems based on CPU, GPU, etc. are all based on the von Neumann structure, and its computing and storage components are separated; when pe...

AI Technical Summary

Problems solved by technology

This structure leads to the need to transfer a large amount of data between the storage unit and the computing unit when performing calculations, especially data-intensive calculations

However, on the one hand, this will cause the final calculation speed to be limited by the data transmission bandwidth, and on the other hand, it will cause an increase in power consumption.

Higher bandwidth, higher energy consumption, and better heat dissipation systems will all lead to higher costs, and even lead to physical limitations in performance improvement.

[0004] In addition, CPUs, GPUs, etc. all use digital circuits to implement computing functions, so their processing units require many transistors to implement, that is, they require a lot of resources, so the total number of processing units they can have is less, which limits their processing units. The degree of computational parallelism that can be achieved, which in turn limits its computational speed

Images