Method and device for predicting storage device performance based on svm machine learning model

Abstract

The invention discloses a method for predicting the performance of a storage device based on an SVM machine learning model. The performance data establishes the feature space vector; constructs the hyperplane decision function with coefficients to be solved; trains the hyperplane decision function according to several feature space vectors to solve the coefficients to be solved and brings the coefficients to be solved into the hyperplane decision function; according to the obtained The hyperplane decision function obtains the high and low performance prediction conclusions of storage devices corresponding to the input configuration information in different storage device environments. The invention also discloses a computer device. The method and device for predicting the performance of a storage device based on the SVM machine learning model proposed by the present invention have strong practicability, can improve the performance utilization rate of the storage device conveniently, and improve the performance test efficiency of the storage device.

Description

Method and device for predicting storage device performance based on SVM machine learning model technical field The present invention relates to the field of computer storage devices, more specifically, particularly to a machine learning model based on SVM A method and apparatus for predicting the performance level of a storage device. Background technique With the rapid development of scientific computing and various network applications, the amount of information generated by human beings is increasing, which makes The storage of data has been paid more and more attention by people, which makes the position of storage components in the whole computer system more and more. More importantly, storage has shifted from a single disk, tape to disk array, and then developed to the current popular storage network. Big The demand for large-scale data applications continues to emerge, and massive data and its applications have become a new development direction. It has ha...

AI Technical Summary

Problems solved by technology

The performance value of the storage device is under the condition that the hardware remains unchanged, and the number of links selected, the RAID (Redundant Arrays of Independent Drives, disk array) level, the number of disks included in the RAID, and the created LUN (Logical Unit Number, logical unit number ) quantity, concurrent number and other parameter configurations. In the process of storage device performance tuning, each parameter adjustment will affect the performance level. Generally, it is necessary to repeatedly debug and combine experience to select the optimal storage performance configuration. There is no The direct theoretical basis and scheme can refer to

Images