Memory detection model training method and memory detection method and device

Abstract

The invention discloses a memory detection model training method. The method comprises the steps of obtaining a memory state historical data set, generating a real fault label set according to the memory state historical data set, and generating a to-be-trained feature set according to the memory state historical data set, training the to-be-trained memory detection model according to the to-be-trained feature set to obtain a prediction fault label set, and if the prediction fault label set and the real fault label set meet a model verification condition, training according to the to-be-trained memory detection model to obtain a memory detection model. The invention further discloses a memory detection method and device. According to the memory detection model provided by the invention, the memory fault condition can be predicted according to the granularity of the memory module level, and the health condition and the risk level of the memory are fully considered, so that the fault positioning accuracy of memory detection is improved.

Description

technical field [0001] The present application relates to the technical field of the Internet, and in particular to a memory detection model training method, a memory detection method and a device. Background technique [0002] With the development of science and technology, computers have entered thousands of households. The hardware system of a computer is composed of an arithmetic unit, a controller, a memory, an input device, and an output device. The memory in the computer is divided into internal memory and external memory. Memory is used to store programs and data that are currently in use, or that will be used at any time. Once there is an error or failure in the memory, it will cause the program to not work properly or crash. Therefore, it is of great significance to study the possible faults of memory. [0003] At present, the industry generally adopts the fault matching model to detect memory, that is, to extract correctable errors (Correctable Errors, CE), un...

AI Technical Summary

Problems solved by technology

[0004] However, the above fault matching model can only predict UE at the granularity of the system level, and the prediction method based on the system level is relatively granular, so it is difficult to locate specific memory modules, resulting in low memory detection accuracy

Images