A method and system for reducing edram refresh energy consumption in a neural network chip

Abstract

The present invention provides a method and system for reducing the energy consumption of eDRAM refreshing in a neural network chip. The method includes: training the original neural network model, determining the target neural network model with maximum fault tolerance and the target neural network model corresponding to the target neural network model. The data retention time of the eDRAM; scheduling each layer of the target neural network model, determining the computing mode of each layer and the data survival time of each layer under the lowest computing energy consumption; setting the target neural network model according to the The calculation mode of the layer is executed on the neural network chip. For the storage partition of each layer, if the storage partition does not store valid data or the data lifetime of this layer is less than the data retention time, the storage partition is not refreshed . The invention can remove unnecessary refreshing operations to the greatest extent, and greatly reduce energy consumption of eDRAM refreshing in neural network chips.

Description

technical field [0001] The invention belongs to the field of neural network chip acceleration, and in particular relates to a method and system for reducing energy consumption of eDRAM refreshing in a neural network chip. Background technique [0002] With the advent of the era of artificial intelligence, intelligent tasks such as image recognition, speech recognition, and natural language processing are ubiquitous in life. Neural networks have achieved world-leading results in such intelligent tasks and are widely used in the industry. For example, Baidu image search, Microsoft speech recognition and Google online translation, etc., are all implemented based on neural networks. Due to its regular calculation and large parallelism, the neural network needs and is especially suitable for using neural network chips to accelerate its calculation. However, due to the large amount of data and the limited internal memory capacity of the chip, a large number of external memory ac...

AI Technical Summary

Problems solved by technology

[0004] The invention is used to solve the defect that the eDRAM type neural network chip in the prior art has relatively large refresh energy consumption and weakens the energy consumption benefit brought by the reduction of external storage access

Images