Method and device for adaptively switching OpenStack computing node into control node
A technology for adaptive switching and node control, applied in program control devices, computing, program control design, etc., can solve the problems of slowing down the running speed of the cloud platform, difficult to meet the requirements of high timeliness, and low timeliness. Dependency conflict, achieve self-healing or expansion, and avoid performance degradation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0053] A method for adaptive switching of OpenStack computing nodes to control nodes, wherein the OpenStack includes a control node group and a computing node group, such as figure 1 , the method includes:
[0054] S1: Divide the computing node group into a switchable computing node group and a non-switchable computing node group, and elect a computing node to be switched from the switchable computing node group through an election algorithm;
[0055] S2: Trigger monitoring on a periodic basis. If a control node group is found to have a node failure or the total load is too high, live migrate the virtual machine on the computing node to be switched to other computing nodes in the group;
[0056] S3: Combine the container technology with the automatic management tool to switch the computing node to be switched to the control node and join the control node group described in step S2.
[0057] Before the hot migration of the virtual machine described in step S2, the computing re...
Embodiment 2
[0078] In this implementation, the calculation method of the total load of the control node group is calculated using a forecasting algorithm, and the others are the same as in Embodiment 1, and the forecasting algorithm is specifically:
[0079] Based on the historical monitoring data of the control node, a neural network linear regression model with multiple inputs and single outputs is used for prediction, and the neural network linear regression model is:
[0080] Z=WX+B
[0081] Where Z is the load forecast value of the control node, X={x 1 ,x 2 ,...,x N} is the input sample, which is the time or the number of tenants, W={w 1 ,w 2 ,...,w N} is the weight matrix, B={b 1} is the offset matrix, use the mean square error function as the cost function, calculate W and B through forward calculation and backward conduction, and obtain the total load of the control node group according to the obtained load prediction value Z, if the total load exceeds the preset threshold,...
Embodiment 3
[0083] An apparatus for adaptively switching an OpenStack computing node to a control node corresponding to Embodiment 1, the OpenStack includes a control node group and a computing node group, and the apparatus includes:
[0084] The monitoring agent module is used to send heartbeat packets to each control node in the control node group to determine whether a node failure occurs in the control node group; this module is also used to collect the load information of each control node in the control node group, and according to the collected load information to calculate the total load of the control node group, or predict the total load of the control node group based on historical load information, and judge whether the control node group is overloaded according to the set load threshold;
[0085] The node switching module is used to divide the computing node group into a switchable computing node group and a non-switchable computing node group. The computing node to be switche...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
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