/Memory Allocation Reducing Data Copy Operations In A Neural Processing Unit
Abstract

A method for controlling a data allocatable neural processing unit is provided. The method may comprise: determining a dependency determination between data blocks of a specific layer of at least one neural network model; allocating memory banks to store the data blocks based on the dependency determination; and storing each data block in the allocated memory bank.

Full Text

What is claimed is:

A method for controlling a data allocatable neural processing unit is provided. The method may comprise: determining a dependency determination between data blocks of a specific layer of at least one neural network model; allocating memory banks to store the data blocks based on the dependency determination; and storing each data block in the allocated memory bank.
Timeline
Filed
06/17/2026
Published
10/01/2026
Granted
Not Available
IPC Codes(1)
G06F 12/02:Addressing or allocation; Relocation (program address sequencing G06F 9/00; arrangements for selecting an address in a digital store G11C 8/00)