Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for adaptive switching of openstack control node to computing node

An adaptive switching and computing node technology, applied in the transmission system, electrical components, etc., can solve the problems of complex deployment, low timeliness, and slow service speed, and achieve the effect of high deployment efficiency and high availability

Active Publication Date: 2022-07-12
上海仪电(集团)有限公司中央研究院
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Each server in the high-availability technology virtualization cluster is configured with an agent. The agent continuously detects the status of other servers, and the agent sends heartbeat signals to other servers at regular intervals. Suppose a server fails to respond to the heartbeat signal three times in a row. When a server fails and reports the corresponding failure, the agent will restart all the virtual machines on the failed server on other servers in the virtualization cluster to restore the virtual machine business, thereby ensuring business continuity, or adding a new server or replacing the original server. For some faulty servers, each computing service and network proxy service of the cloud platform is directly deployed on the operating system of the newly added server through the RPM package. However, this method has low timeliness due to the cumbersome process of server removal and deployment. , Recovery is slow, it is difficult to meet the requirements of high timeliness in the production environment
[0004] The prior art also proposes an effective solution to the above problems. Chinese patent CN108089911A proposes a control method and device for computing nodes in an OpenStack environment. The method includes: periodically monitoring the status data of the computing nodes in the control node, Determine whether the corresponding computing node is available, and if not, evacuate the virtual machine running on the corresponding computing node to the computing node that is determined to be available in the environment according to the monitored state data. The method determines whether the computing node is Available, and timely evacuation of virtual machines running on unavailable computing nodes can minimize RTO and RPO, restore virtual machine services in the shortest time, and maintain high availability, but the load of computing nodes receiving virtual machines will also increase It will cause the running speed of the service on this node to slow down. At the same time, the virtual machines of the unavailable computing nodes will be distributed to other available computing nodes, which will increase the operating burden of other computing nodes. This method cannot effectively solve the problem of excessive load on computing nodes. Affects the quality of service of tenant virtual machines

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
  • Method and device for adaptive switching of openstack control node to computing node
  • Method and device for adaptive switching of openstack control node to computing node
  • Method and device for adaptive switching of openstack control node to computing node

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] A method for adaptively switching an OpenStack control node to a computing node. OpenStack is a topology structure composed of a control node group and a computing node group, such as figure 1 , the method includes:

[0044] S1: Divide the control node group into a switchable control node group and a non-switchable control node group, and elect a to-be-switched control node from the switchable control node group through an election algorithm;

[0045] S2: Trigger monitoring on a periodic basis. If it is found that there is a node failure in the computing node group or the total load is too high, the adaptive upgrade process is triggered, otherwise the process ends;

[0046] The self-adaptive upgrade process is specifically: switching the control node to be switched to the computing node and joining the computing node group described in step S2 through an automated management tool in combination with the container technology.

[0047] The election algorithm selects a no...

Embodiment 2

[0064] In this implementation, the total load of the computing node group is calculated and obtained through the prediction algorithm. Others are the same as those in the first embodiment, and the prediction algorithm is specifically:

[0065] Based on the historical monitoring data of the computing nodes, the prediction is made through a multi-input single-output neural network linear regression model. The neural network linear regression model is:

[0066] Z=WX+B

[0067] Where Z is the predicted value of the computing node load, X={x 1 ,x 2 ,…,x N } is the input sample, the sample includes time, number of virtual machines and number of tenants, W={w 1 ,w 2 ,…,w N } is the weight matrix, B={b 1 } is the offset matrix, the mean square error function is used as the cost function, W and B are calculated through forward calculation and backward conduction, and the total load of the calculation node group is obtained according to the load prediction value Z of each calculat...

Embodiment 3

[0069] A device for adaptively switching an OpenStack control node to a computing node corresponding to the first embodiment. OpenStack includes a control node group and a computing node group, and the device includes:

[0070] A fault monitoring module, configured to detect whether a node failure occurs in the computing node group by sending a heartbeat packet to each computing node in the computing node group;

[0071] The load detection module is used to collect the load information elements of each computing node in the computing node group, calculate the total load of the computing node group according to the collected load information, or predict the total load of the computing node group according to the historical load information, and set the load threshold according to the load threshold. Determine whether the computing node group is overloaded;

[0072] The node processing module is used to divide the control node group into a switchable control node group and a non...

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 present invention relates to a method and device for adaptively switching an OpenStack control node to a computing node. The OpenStack includes several groups of control node groups and computing node groups. The method includes: S1: dividing several groups of control node groups into switchable control node groups The node group and the non-switchable control node group are selected from the switchable control node group through the election algorithm to select the control node to be switched; S2: Trigger the monitoring periodically. Adaptive upgrade process, otherwise the process ends; wherein, the adaptive upgrade process is specifically: switching the control node to be switched to a computing node through an automated management tool combined with the container technology and joining the computing node group in step S2. Compared with the prior art, the present invention has the advantages of high efficiency.

Description

technical field [0001] The invention relates to the technical field of OpenStack cloud platforms, in particular to a method and device for adaptively switching an OpenStack computing node to a control node. Background technique [0002] OpenStack is an open source cloud computing management platform that can manage a large number of distributed computing resources, storage resources and network resources, and provide a unified management panel. OpenStack is not only a community, but also a project and an open source software , which supports almost all types of cloud environments, providing a solution or toolset for deploying the cloud. Its purpose is: to help organizations run clouds for virtual computing or storage services, and to provide scalable and flexible cloud computing for public and private clouds. After years of production practice, OpenStack has developed very mature. [0003] In small and medium-scale cloud platforms, the general deployment architecture is a m...

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 Patents(China)
IPC IPC(8): H04L41/0668H04L41/147H04L43/12H04L43/0817H04L41/14
CPCH04L41/0668H04L41/30H04L41/147H04L43/12H04L43/0817H04L41/145
Inventor 刘梦可刘超
Owner 上海仪电(集团)有限公司中央研究院