Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Kubernetes dispatching optimization method based on neural network

An optimization method and neural network technology, applied in the field of Kubernetes, can solve the problems of reduced system stability and high memory resource consumption

Inactive Publication Date: 2018-11-23
GUILIN UNIV OF ELECTRONIC TECH
View PDF1 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the prior art, the present invention provides an optimization method and system for intelligent scheduling of Kubernetes memory resources, so as to solve the problem of reduced system stability due to excessive consumption of memory resources in Kubernetes container applications

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Kubernetes dispatching optimization method based on neural network
  • Kubernetes dispatching optimization method based on neural network
  • Kubernetes dispatching optimization method based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0012] Neural network model design includes the following steps:

[0013] (1) Data preprocessing is mainly responsible for standardizing multivariate heterogeneous data with different standards and scales. In the Kubernetes system, the collected resource usage comes from various data of different nodes, and there may be abnormal resource fluctuations in the data. If there is a large difference in the data input point, it will cause the neuron weight to fluctuate or abnormally shift during the training process, which will affect the training effect of the model. To this end, we need to standardize the original data and map data of different categories and sizes into the same distribution. The present invention uses a normalization method to preprocess the data, and maps the input data to the [0,1] interval, the formula is:

[0014]

[0015] Where X is the mapping relationship of the input data, and X is the real data in the data set.

[0016] (2) The input layer is mainly...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a Kubernetes dispatching optimization method based on a neural network. The method comprises building of a prediction model and a resource allocation algorithm; the memory consumption of a Node in Kubernetes is predicted through the recurrent neural network; the change of the memory consumption in a future period of time is predicted; memory consumption data is input into the resource allocation algorithm to calculate out the number of instances needed to be added; and a Kubernetes system performs dynamic expansion according to the obtained number of the instances, so that a memory resource-based dynamic expanding and contracting function of the Kubernetes is completed. According to the method, the problem that the stability of the system is reduced due to excessively high memory resource consumption in container application of the Kubernetes can be solved.

Description

technical field [0001] The invention relates to the Kubernetes technology in cloud computing, in particular to a neural network-based Kubernetes scheduling optimization method. Background technique [0002] As a lightweight open source container orchestration system, Kubernetes splits applications into small modular services that can execute specific processes, and meets the needs of complex applications by specifying the association rules of these services, making complex applications more flexible and stable , which is easier to release applications and update modules. At the same time, the Kubernetes system integrates functions such as elastic scaling and automatic load balancing by default, making container operations more efficient and service scheduling more convenient. The consumption of various resources in the production environment puts high demands on the elastic scaling of the platform. Elastic scaling adjusts the resource load changes of the platform according...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/50G06F8/60G06F8/20G06N3/08
CPCG06F8/20G06F8/60G06F9/505G06N3/08
Inventor 强保华赵兴朝谢武陶林宁毅莫烨卢永全
Owner GUILIN UNIV OF ELECTRONIC TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com