Computer performance data determination method, device, computer equipment and storage medium

By pre-training and classifying the computer historical performance data of cloud computing platform service manufacturers, a performance prediction model is built, and the problems of low performance prediction accuracy and high labor investment cost in the existing technology are solved, achieving more accurate and efficient performance prediction.

CN111679959BActive Publication Date: 2025-04-01PING AN TECH (SHENZHEN) CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202010328656.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-04-23
Publication Date
2025-04-01
Estimated Expiration
2040-04-23

AI Technical Summary

Technical Problem

The prior art has low accuracy in performance prediction among cloud computing platform service manufacturers, especially in long-term performance prediction and large-scale performance indicator prediction, which leads to high labor investment costs and possible overfitting problems.

Method used

By acquiring multiple historical performance data sequences of the computer, pre-training is used using long and short-term memory networks, and these data sequences are classified to obtain different sequence types and corresponding subsets. Then, a performance prediction model is constructed using the pre-trained long and short-term memory network and the full connection layer, each sequence type is trained, and the performance data of the computer is finally predicted through the corresponding trained model.

Benefits of technology

It improves the accuracy of computer performance prediction, reduces the cost of labor investment, and reduces the risk of overfitting, especially in long-term and large-scale performance indicator prediction scenarios.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention relates to blockchain technology, and provides a method for determining computer performance data and related equipment, wherein the method obtains multiple historical performance data sequences of a computer; uses multiple historical performance data sequences to pre-train a long short-term memory network; classifies multiple historical performance data sequences to obtain N sequence types and N historical performance data sequence subsets corresponding to the N sequence types; uses each historical performance data sequence subset to train a performance prediction model consisting of a pre-trained long short-term memory network and a fully connected layer located after the pre-trained long short-term memory network, to obtain N trained performance prediction models corresponding to the N sequence types; and predicts the performance data of the computer through the trained performance prediction model corresponding to the sequence type of the performance data sequence to be predicted of the computer. In addition, the present application also relates to blockchain technology, and the performance data can be stored in the blockchain.
Need to check novelty before this filing date? Find Prior Art