Cloud platform node management method and device and nonvolatile storage medium

By introducing leader nodes into the cloud platform, detecting and dynamically adjusting the resource utilization rate at the node level, the balance between resource conservation and performance maintenance is achieved, the problem of low resource utilization efficiency in the existing technology is solved, and the cloud computing system is improved Overall efficiency and stability.

CN118963990APending Publication Date: 2024-11-15CHINA TELECOM CORP LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410925786.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing cloud platform load balancing technology cannot take into account resource conservation and performance maintenance, resulting in low resource utilization efficiency.

Method used

By introducing leader nodes into the cloud platform, detect resource utilization at each node level, dynamically migrate virtual machines to adjust node status, ensure that resource utilization is within a reasonable range, and achieve a balance between resource conservation and performance maintenance.

Benefits of technology

It improves resource utilization efficiency, reduces the energy consumption of cloud computing centers, and ensures the stability and reliability of system performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963990A_ABST
    Figure CN118963990A_ABST
Patent Text Reader

Abstract

The invention discloses a cloud platform node management method and device and a nonvolatile storage medium. The method comprises the steps that a leader node in each node hierarchy detects the resource utilization rate of other nodes in the corresponding node hierarchy, and the node hierarchy is obtained through division according to the resource utilization rate of each node in a cloud platform; a first node and a second node are determined in a node hierarchy according to the detection result, the first node is a node needing to be migrated out of the virtual machine, the second node is a node migrated in the virtual machine, and the first node and the second node are both in the same node hierarchy with the leader node; and after the virtual machine in the first node is migrated to the second node, the leader node adjusts the working state of each node according to the resource utilization rate of each node in the node hierarchy. The technical problem that the resource utilization efficiency is low due to the fact that an existing cloud platform load balancing technology cannot give consideration to resource saving and performance maintenance is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to a cloud platform node management method, device and non-volatile storage medium. Background Art

[0002] In the cloud computing environment, three levels of service resource pools have gradually been formed, including Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS). For many years, how to balance resource conservation and performance maintenance to manage load balancing between platform nodes, thereby improving resource utilization and reducing energy consumption in cloud computing centers has always been a hot topic in the field of cluster computing. A large part of the operating costs of cloud computing centers is caused by power consumption and cooling. In addition, most power generation technologies (such as nuclear power and coal power generation) are harmful to the environment. In past work, there are some cloud platform management technologies, however, these technologies cannot balance resource conservation and performance maintenance to manage cloud platform nodes.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a cloud platform node management method, device and non-volatile storage medium to at least solve the technical problem of low resource utilization efficiency caused by the inability of existing cloud platform load balancing technology to balance resource conservation and performance maintenance.

[0005] According to one aspect of an embodiment of the present application, a cloud platform node management method is provided, including: a leader node in each node level detects the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform; determining a first node and a second node in the node level based on the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine needs to be migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node level.

[0006] Optionally, the leader node in each node level detects the resource utilization of other nodes in the corresponding node level, including: the leader node periodically updates the logical links between it and other nodes in the corresponding node level, and detects the resource utilization of other nodes in the corresponding node level through the logical links; when the leader node detects that the resource utilization of an existing node exceeds a first preset value, the detection result of the node level is determined to be that the resource utilization exceeds the first preset value; when the leader node detects that the resource utilization of a non-existent node exceeds the first preset value, and the resource utilization of an existing node exceeds a second preset value, the detection result of the node level is determined to be that the resource utilization exceeds the second preset value, wherein the first preset value is greater than the second preset value; when the leader node detects that the resource utilization of a non-existent node exceeds the second preset value, the detection result of the node level is determined to be that the resource utilization is normal.

[0007] Optionally, determining the first node and the second node in the node level based on the detection result includes: determining a migration measure at the node level based on the detection result, and determining the first node and the second node in the node level based on the migration measure, wherein the migration measure includes a first migration measure and a second migration measure, and the resource utilization of the second node corresponding to the first migration measure is lower than the resource utilization of the first node, and the resource utilization of the second node corresponding to the second migration measure is higher than the resource utilization of the first node; when the detection result at the node level is that the resource utilization exceeds a first preset value, determining the migration measure at the node level to be the first migration measure, and determining the first node and the second node in the node level based on the first migration measure; when the detection result at the node level is that the resource utilization exceeds a second preset value, determining the migration measure at the node level to be the second migration measure, and determining the first node and the second node in the node level based on the second migration measure.

[0008] Optionally, when the detection result at the node level is that the resource utilization exceeds a first preset value, determining the migration measure at the node level as the first migration measure, and determining the first node and the second node in the node level based on the first migration measure include: when the migration measure at the node level is the first migration measure, determining the node in the node level whose resource utilization exceeds the first preset value as the first node, and selecting the virtual machine with the lowest resource utilization in the first node as the virtual machine to be migrated; determining the second node in the node level as the node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated, including an idle node without a mounted virtual machine.

[0009] Optionally, after determining that the second node in the node hierarchy is a node whose resource utilization is still less than a first preset value after the migration of the virtual machine to be migrated, including an idle node without a mounted virtual machine, the method also includes: when the second node in the node hierarchy is an idle node, the leader node in the node hierarchy adjusts the working status of the second node to on.

[0010] Optionally, when the detection result at the node level is that the resource utilization exceeds a second preset value, determining the migration measure at the node level as the second migration measure, and determining the first node and the second node at the node level based on the second migration measure include: when the migration measure at the node level is the second migration measure, selecting two virtual machines with the smallest resource utilization in the node level as virtual machines to be migrated, and determining the node in the node level on which the virtual machines to be migrated are mounted as the first node; and determining the second node in the node level as the node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated.

[0011] Optionally, after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node hierarchy, including: determining that the node without the virtual machine mounted is an idle node; adjusting the working status of the idle node to node shutdown.

[0012] According to another aspect of an embodiment of the present application, a cloud platform node management system is also provided, including: a cloud platform includes multiple nodes, the multiple nodes are divided into multiple node levels according to the resource utilization of the nodes, and each of the multiple node levels contains a leader node, the leader node is the node with the highest resource utilization in the node level, wherein the leader node is used to detect the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing the resource utilization of each node in the cloud platform; determining a first node and a second node in the node level according to the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine is migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, adjusting the working status of each node according to the resource utilization of each node in the node level.

[0013] According to another aspect of an embodiment of the present application, a cloud platform node management device is also provided, which is applicable to the leader node in each node level, and includes: a detection module, which is used to detect the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform; a determination module, which is used to determine a first node and a second node in the node level according to the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine needs to be migrated in, and the first node and the second node are both in the same node level as the leader node; an adjustment module, which is used for the leader node to adjust the working status of each node according to the resource utilization of each node in the node level after migrating the virtual machine in the first node to the second node.

[0014] According to another aspect of an embodiment of the present application, a non-volatile storage medium is further provided, in which a program is stored, wherein when the program is running, a device where the non-volatile storage medium is located is controlled to execute a cloud platform node management method.

[0015] According to another aspect of an embodiment of the present application, there is also provided an electronic device, including: a memory and a processor, the processor being used to run a program stored in the memory, wherein the cloud platform node management method is executed when the program is running.

[0016] According to another aspect of an embodiment of the present application, a computer program product is also provided, including a computer program, which implements a cloud platform node management method when the computer program is executed by a processor.

[0017] In an embodiment of the present application, a leader node in each node level is used to detect the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing the resource utilization of each node in the cloud platform; a first node and a second node are determined in the node level based on the detection result, wherein the first node is a node from which the virtual machine needs to be migrated out, and the second node is a node to which the virtual machine is migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node level, and achieves the purpose of load balancing that takes into account both resource conservation and performance maintenance through dynamic cloud platform node migration management, thereby achieving the technical effect of improving resource utilization efficiency, and further solving the technical problem of low resource utilization efficiency caused by the inability of existing cloud platform load balancing technology to take into account both resource conservation and performance maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 It is a flowchart of a cloud platform node management method provided according to an embodiment of the present application;

[0020] Figure 2 is a flow chart of a first migration measure provided according to an embodiment of the present application;

[0021] Figure 3 is a flow chart of a second migration measure provided according to an embodiment of the present application;

[0022] Figure 4 It is a structural diagram of a cloud platform node management device provided according to an embodiment of the present invention;

[0023] Figure 5 It is a structural schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] In order to better understand the embodiments of the present application, the technical terms involved in the embodiments of the present application are explained as follows:

[0027] IaaS: Infrastructure as a Service is a cloud computing service model that provides virtualized computing resources, storage space, and network functions. Users can configure and manage these infrastructures according to their needs without having to purchase, maintain, and manage actual hardware devices. Through IaaS, users can quickly deploy and expand applications while reducing costs and complexity.

[0028] PaaS: Platform as a Service is a cloud computing service model that provides developers with a complete application development and deployment platform, including operating system, database, development tools and operating environment. Through PaaS, developers can develop, test and deploy applications more quickly without having to worry about the underlying infrastructure.

[0029] SaaS: Software as a Service is a cloud computing service model that provides software applications through the Internet. Users do not need to purchase, install or maintain software, but use the software services provided by the cloud through subscription. SaaS providers are responsible for the operation, maintenance, security and updates of the software, and users only need to access the software through the Internet to use it, reducing users' management and maintenance costs.

[0030] VM: Virtual Machine is a virtualization technology that can create multiple independent virtual computing environments on a single physical computer. Each virtual machine has its own operating system and applications and can be run and managed like an independent physical computer. Virtual machines can provide higher resource utilization and flexibility, allowing users to run multiple virtual machines simultaneously on a single physical device, thereby making more efficient use of computing resources.

[0031] In the related art, local technology focuses on reducing energy consumption by reducing the physical power consumption of a single node, such as reducing clock frequency, power supply voltage or saving interconnection components such as switches, network interface cards (NICs) and links. However, most of these technologies only focus on the management of a single node, and lack a global concept of maintaining the performance of the entire system. Cloud computing platforms still need to maintain the operation of all nodes when the load is low; on the contrary, the full cluster technology involves multiple nodes, and its basic idea is to summarize the system load, and then determine the minimum set of nodes that can handle the load, and save energy by shutting down redundant nodes. However, determining the minimum point set consumes a lot of resources, and the full cluster method can only be applied to specific applications, because its front end must provide a good load balancing algorithm to redirect the connection status of the back-end node, otherwise it may make the minimum node set resource utilization too high, affecting node performance and service stability. Therefore, there is a problem of low resource utilization efficiency caused by not being able to take into account resource conservation and performance maintenance. In order to solve this problem, a relevant solution is provided in the embodiment of the present application.

[0032] According to an embodiment of the present application, a method embodiment of a cloud platform node management method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] An embodiment of the present application provides a cloud platform node management system, including: a cloud platform includes multiple nodes, the multiple nodes are divided into multiple node levels according to the resource utilization of the nodes, and each node level in the multiple node levels includes a leader node, the leader node is the node with the highest resource utilization in the node level, wherein the leader node is used to detect the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing the resource utilization of each node in the cloud platform; determining a first node and a second node in the node level according to the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine needs to be migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, adjusting the working status of each node according to the resource utilization of each node in the node level.

[0034] Specifically, a cloud platform is a service model based on cloud computing technology. The platform can be equipped with multiple nodes, and multiple virtual machines can run tasks on the nodes. These virtual machines can provide a variety of cloud computing services including computing, storage, website servers, database systems, development environments, etc., and virtual machines can be migrated between nodes through migration commands in the cloud management platform.

[0035] In the above operating environment, the present application embodiment provides a cloud platform node management method, such as Figure 1 As shown, the method comprises the following steps:

[0036] Step S102, the leader node in each node level detects the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform;

[0037] In some embodiments of the present application, in order to achieve effective allocation and scheduling of resources, a hierarchical management method is also provided, which classifies different nodes according to the level of physical resources consumed by them, including resource utilization. This method can better understand and manage the resource utilization of nodes, thereby improving the efficiency and performance of the entire cloud platform. First, determine the second preset value as the basis for stratification. The second preset value should be determined according to the actual situation, and factors such as system performance indicators, historical data, and business needs can be considered. The reasonable selection of the preset value is crucial for subsequent resource management and scheduling, so careful analysis and evaluation are required. In actual operation, in order to ensure the stability of the business platform, a first preset value needs to be set as the maximum resource consumption when the node is running. This threshold can be determined based on business needs and the carrying capacity of the platform. Generally, a certain margin should be left to cope with emergencies and peak loads. Based on the above settings, the nodes can be divided into three different levels, and each layer is set with different first preset values ​​and second preset values. As the number of layers increases, the first preset value and second preset value of each layer node also increase. Each level represents a different resource consumption level and management strategy, and the resource consumption level is the resource utilization level:

[0038] Tier 1: The resource consumption of the nodes is between 0 and the second preset value of the first layer. These nodes usually consume less resources and are suitable for lightweight tasks and services, such as some low-traffic applications or development environments. Tier 2: The resource consumption of the nodes is between the second preset value of the first layer and the second preset value of the second layer. These nodes consume moderate resources and can undertake general business loads, such as websites with medium traffic or database services. Tier 3: The resource consumption of the nodes is between the second preset value of the second layer and the second preset value of the third layer. These nodes consume more resources and are usually used to undertake high-traffic and high-load tasks, such as busy e-commerce websites or large database systems.

[0039] By dividing nodes into different levels, resource management and scheduling can be better performed. For example, for nodes that consume less resources, a more flexible resource release strategy can be adopted to improve the resource utilization of the entire platform; for nodes that consume more resources, stricter resource restrictions and monitoring measures can be adopted to ensure the stability and performance of the platform. The hierarchical management method based on preset values ​​can better grasp and manage the resource utilization of the cloud computing platform, thereby improving the efficiency, stability and performance of the platform.

[0040] In some embodiments of the present application, since the memory and CPU requirements of the tasks run by the virtual machine may change dynamically and the execution time cannot be known in advance, it is necessary to select a leader node from each layer of nodes, which is responsible for maintaining the stable superposition of the energy consumption of the nodes in this layer. The leader is generally the node with the largest load in the same layer of servers. In each layer, the leader regularly updates its logical links with members. The leader regularly checks that the workload of its members is within the preset resource value of its layer.

[0041] Specifically, to implement this mechanism, you can follow the steps below:

[0042] Select leader node: The leader node of each layer should be the node with the highest load in the layer. This can be achieved by monitoring the resource utilization of the nodes. Usually, the leader will be the node that consumes the most resources because it has the ability to handle more tasks and jobs. Maintain logical links: The leader needs to regularly update its logical links with member nodes to ensure unimpeded communication with other nodes in the layer. This can be achieved by regularly sending heartbeat signals or other forms of communication. Check member workload: The leader regularly checks the workload of its member nodes, including resource utilization, whether it is greater than the first preset value of its layer and whether it is greater than the second preset value of its layer. This can be achieved by monitoring the resource utilization of the nodes and the length of the job queue. If it is found that the workload of a node exceeds the preset value, including the first preset value and the second preset value, the leader needs to take measures to adjust it to maintain the stable operation of the entire layer. Adjust the location of virtual machines: If it is found that the workload of a node exceeds the preset value, the leader needs to migrate the virtual machines on the nodes of this layer to balance the resource utilization of the entire platform.

[0043] Through the above mechanism, the leader of each layer can effectively maintain the stable superposition of the energy consumption of the servers in this layer and keep the resource utilization of the entire platform within a reasonable range. This dynamic adjustment mechanism can adapt to the changing needs of different operations and tasks, thereby improving the efficiency and performance of the entire platform.

[0044] Step S104, determining a first node and a second node in the node hierarchy according to the detection result, wherein the first node is a node from which the virtual machine needs to be migrated out, the second node is a node to which the virtual machine is migrated in, and the first node and the second node are both in the same node hierarchy as the leader node;

[0045] In some embodiments of the present application, if the detection result is that the resource utilization of the node is greater than the first preset value, it means that the resource utilization of this layer is too high, and the cloud computing platform is in an intensive computing state, that is, when the load is high, the leader node will take release measures, that is, the first migration measure: transfer the virtual machine on the node with high resource utilization to the node with low resource utilization to improve performance. The node with high resource utilization is the first node, and the node with low resource utilization is the second node. When the second node is an idle node in the closed state of this layer, the leader node adjusts its working state to on and migrates the virtual machine to run.

[0046] Specifically, the release measures are intended to ensure the continuous and stable operation of the business platform and optimize the utilization of host resources to deal with the situation where the resource utilization of the host is too high and reaches the first preset value. The host is the node of each layer. When the resource consumption of the host is close to or exceeds the first preset value, the system will automatically trigger the release measures to migrate the virtual machines in the host with higher resource consumption to the idle nodes with lower resource consumption to balance the load of the entire system. This process involves the following detailed steps: Monitor host resource consumption: The system continuously monitors the resource consumption of the host, including indicators such as CPU utilization, memory usage, and disk IO. Trigger alarm: When the resource consumption of the host reaches the first preset value, the system initiates the release measures. Identify resource overloaded nodes: The system identifies nodes whose resource utilization exceeds the first preset value. Find target nodes: The system searches for idle nodes and evaluates their resource consumption to determine the migration target. Migrate virtual machines: The resource-overloaded virtual machines are migrated to the selected target nodes, which involves saving and transmitting the state of the virtual machines to the target nodes to ensure business continuity and data integrity. Update resource status: After completing the migration of the virtual machine, the system updates the resource consumption status and re-evaluates the resource utilization of the host machine. Through these steps, the system can dynamically adjust resource allocation to ensure the stability and performance of the business platform, maximize resource utilization, and improve system availability and efficiency. Figure 2 A schematic diagram of the process of the first migration measure is shown. The original resource utilization of a virtual machine is 26%. For some reason, the resource utilization suddenly changes to 66%, causing the resource consumption of the node where it is located to reach 86%. The leader node detects that its resource utilization exceeds the first preset value, and starts the first migration measure to migrate the virtual machine with 20% resource utilization to the node with low utilization. It should be noted that the specific data in this embodiment is only used for example description. In specific application scenarios, the preset value can adopt other custom parameters according to actual needs and environment.

[0047] In some embodiments of the present application, if the detection result is that there is no node whose resource utilization exceeds the first preset value, and there is a node whose resource utilization exceeds the second preset value, it means that when the cloud computing platform is in a non-intensive computing state, that is, the load is low, the leader node will take squeeze measures to dynamically squeeze the existing running jobs widely distributed on lightweight nodes to the nodes with heavier loads, that is, the second migration measure: migrate the virtual machines of the low resource utilization nodes to the high resource utilization nodes to reduce redundant nodes, effectively utilize resources, and ensure the load balance of the nodes. The low resource utilization node is the first node, and the high resource utilization node is the second node.

[0048] Specifically, the migration process of the second migration measure is similar to the process of building an optimal tree, which aims to achieve the maximum energy saving effect in the cloud platform. Figure 3The flowchart of the second migration measure shown in the figure shows that when the second preset value is 66%, there are 6 virtual machines (occupying 4%, 6%, 10%, 14, 18 and 26% of resources respectively). During the squeezing process, the two virtual machines with the lightest load will be merged together at each step to ensure that energy consumption is minimized at the minimum migration cost. The steps of this process can be described as follows: Determine the threshold: First, determine a second preset value based on system requirements and performance indicators. In this example, the second preset value is set to 66%. Identify virtual machines with lighter loads: Monitor the virtual machines running on each physical host and determine the two virtual machines with the lightest loads. Merge virtual machines: Migrate the two virtual machines with the lightest loads to the same node to release some resources, thereby ensuring that resource utilization in the entire cloud environment is more balanced. Repeat: Repeat the above steps regularly until the loads of all physical hosts are below the second preset value and the resource utilization in the entire cloud environment reaches the optimal state. Through this virtual machine squeezing process, the leader node can reduce the number of servers as much as possible while ensuring performance, thereby reducing energy consumption and achieving the goal of managing cloud platform nodes while taking into account resource conservation and performance maintenance. It should be noted that the specific data in this embodiment is only used for illustrative purposes. In specific application scenarios, the preset values ​​may adopt other customized parameters according to actual needs and environment.

[0049] Step S106, after migrating the virtual machine in the first node to the second node, the leader node adjusts the working state of each node according to the resource utilization of each node in the node hierarchy.

[0050] In some embodiments of the present application, after performing migration measures, including performing the second migration measures, the leader node again detects the resource utilization of the nodes at the same level. When it is detected that a node no longer mounts a virtual machine after the virtual machine migration, it is determined to be an idle node and its working state is adjusted to shut down, so as to save energy without affecting performance.

[0051] Through the above steps, a node management method can be implemented to reduce the energy consumption of the cloud platform without affecting the performance of the cluster. This method can dynamically adjust the state of the node according to the real-time load conditions, and comprehensively consider the dual goals of energy conservation and performance maintenance. By dynamically adjusting the load and state of the node, it can flexibly adapt under different workloads, thereby maximizing the energy utilization efficiency of the cluster while ensuring the stability and reliability of the system performance. It ensures that the system can effectively utilize resources under different load conditions, solving the technical problem of low resource utilization efficiency caused by the inability of existing cloud platform load balancing technology to take into account both resource conservation and performance maintenance.

[0052] The present application embodiment provides a cloud platform node management device, Figure 4 is a schematic diagram of the structure of the device, such as Figure 4 As shown, the device includes: a detection module 40, which is used to detect the resource utilization of other nodes in the corresponding node hierarchy, wherein the leader node is the node with the largest resource utilization in the node hierarchy, and the node hierarchy is a hierarchy obtained by dividing the resource utilization of each node in the cloud platform; a determination module 42, which is used to determine the first node and the second node in the node hierarchy according to the detection result, wherein the first node is the node from which the virtual machine needs to be migrated out, and the second node is the node to which the virtual machine needs to be migrated in, and the first node and the second node are both in the same node hierarchy as the leader node; an adjustment module 44, which is used for the leader node to adjust the working status of each node according to the resource utilization of each node in the node hierarchy after migrating the virtual machine in the first node to the second node.

[0053] In some embodiments of the present application, the leader node in each node level detects the resource utilization of other nodes in the corresponding node level, including: the leader node periodically updates the logical links between the leader node and other nodes in the corresponding node level, and detects the resource utilization of other nodes in the corresponding node level through the logical links; when the leader node detects that the resource utilization of an existing node exceeds a first preset value, the detection result of the node level is determined to be that the resource utilization exceeds the first preset value; when the leader node detects that the resource utilization of a non-existent node exceeds the first preset value, and the resource utilization of an existing node exceeds a second preset value, the detection result of the node level is determined to be that the resource utilization exceeds the second preset value, wherein the first preset value is greater than the second preset value; when the leader node detects that the resource utilization of a non-existent node exceeds the second preset value, the detection result of the node level is determined to be that the resource utilization is normal.

[0054] In some embodiments of the present application, the determination module 42 determines the first node and the second node in the node level according to the detection result, including: determining the migration measures of the node level according to the detection result, and determining the first node and the second node in the node level according to the migration measures, wherein the migration measures include the first migration measures and the second migration measures, the resource utilization of the second node corresponding to the first migration measure is lower than the resource utilization of the first node, and the resource utilization of the second node corresponding to the second migration measure is higher than the resource utilization of the first node; when the detection result of the node level is that the resource utilization exceeds the first preset value, the migration measure of the node level is determined to be the first migration measure, and the first node and the second node are determined in the node level according to the first migration measure; when the detection result of the node level is that the resource utilization exceeds the second preset value, the migration measure of the node level is determined to be the second migration measure, and the first node and the second node are determined in the node level according to the second migration measure.

[0055] In some embodiments of the present application, when the detection result at the node level is that the resource utilization exceeds the first preset value, the determination module 42 determines that the migration measure at the node level is the first migration measure, and determines the first node and the second node in the node level based on the first migration measure, including: when the migration measure at the node level is the first migration measure, determining that the node in the node level whose resource utilization exceeds the first preset value is the first node, and selecting the virtual machine with the lowest resource utilization in the first node as the virtual machine to be migrated; determining that the second node in the node level is the node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated, including an idle node without a mounted virtual machine.

[0056] In some embodiments of the present application, after the determination module 42 determines that the second node in the node hierarchy is a node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated, including an idle node without a mounted virtual machine, the method also includes: when the second node in the node hierarchy is an idle node, the leader node in the node hierarchy adjusts the working status of the second node to on.

[0057] In some embodiments of the present application, when the detection result at the node level is that the resource utilization exceeds the second preset value, the determination module 42 determines that the migration measure at the node level is the second migration measure, and determines the first node and the second node at the node level based on the second migration measure, including: when the migration measure at the node level is the second migration measure, selecting two virtual machines with the smallest resource utilization in the node level as virtual machines to be migrated, and determining the node on which the virtual machines to be migrated are mounted in the node level as the first node; and determining the second node in the node level as the node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated.

[0058] In some embodiments of the present application, after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node hierarchy, including: determining that the node without the virtual machine mounted is an idle node; adjusting the working status of the idle node to node shutdown.

[0059] It should be noted that the various modules in the above-mentioned cloud platform node management device can be program modules (for example, a set of program instructions that implement a certain specific function) or hardware modules. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0060] The method embodiments provided in the embodiments of the present application may be executed in a mobile terminal, a computer terminal or similar electronic devices. Figure 5The hardware structure block diagram of an electronic device for implementing a cloud platform node management method is shown. Figure 5 As shown, the electronic device 50 may include one or more (502a, 502b, ..., 502n are used to illustrate) processors 502 (the processor 502 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 504 for storing data, and a transmission module 506 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 5 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 5 More or fewer components as shown, or with Figure 5 Different configurations shown.

[0061] It should be noted that the one or more processors 502 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the electronic device 50. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0062] The memory 504 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the cloud platform node management method in the embodiment of the present application. The processor 502 executes various functional applications and data processing by running the software programs and modules stored in the memory 504, that is, the cloud platform node management method described above is realized. The memory 504 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 504 may further include a memory remotely arranged relative to the processor 502, and these remote memories may be connected to the electronic device 50 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0063] The transmission device 506 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the electronic device 50. In one example, the transmission device 506 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 506 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0064] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the electronic device 50 .

[0065] An embodiment of the present application provides a non-volatile storage medium, in which a program is stored, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the following cloud platform node management method: a leader node in each node level detects the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform; a first node and a second node are determined in the node level according to the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine needs to be migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node level.

[0066] An embodiment of the present application provides an electronic device, comprising: a memory and a processor, the processor being used to run a program stored in the memory, wherein the following cloud platform node management method is executed when the program is running: a leader node in each node level detects resource utilization of other nodes in the corresponding node level, wherein the leader node is a node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform; a first node and a second node are determined in the node level according to the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine needs to be migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node level.

[0067] An embodiment of the present application provides a computer program product, including a computer program, which implements the following cloud platform node management method when executed by a processor: a leader node in each node level detects the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform; a first node and a second node are determined in the node level based on the detection result, wherein the first node is a node from which a virtual machine needs to be migrated out, and the second node is a node to which a virtual machine needs to be migrated in, and the first node and the second node are both in the same node level as the leader node; after migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node level.

[0068] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0069] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0070] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0071] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0072] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the relevant technology or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc. Various media that can store program codes.

[0073] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A cloud platform node management method, characterized in that: include: The leader node in each node level detects the resource utilization of other nodes in the corresponding node level, wherein the leader node is the node with the highest resource utilization in the node level, and the node level is a level obtained by dividing according to the resource utilization of each node in the cloud platform; Determine a first node and a second node in the node hierarchy according to the detection result, wherein the first node is a node from which the virtual machine needs to be migrated, the second node is a node to which the virtual machine is migrated, and the first node and the second node are both in the same node hierarchy as the leader node; After migrating the virtual machine in the first node to the second node, the leader node adjusts the working status of each node according to the resource utilization of each node in the node hierarchy.

2. The cloud platform node management method according to claim 1, characterized in that: The leader node in each node level detects the resource utilization of other nodes in the corresponding node level, including: The leader node periodically updates the logical links between the leader node and other nodes in the corresponding node hierarchy, and detects resource utilization of other nodes in the corresponding node hierarchy through the logical links; When the leader node detects that the resource utilization of a node exceeds a first preset value, determining the detection result of the node level as the resource utilization exceeding the first preset value; When the leader node detects that the resource utilization of the non-existing node exceeds a first preset value, and the resource utilization of the existing node exceeds a second preset value, determining the detection result of the node level as the resource utilization exceeding the second preset value, wherein the first preset value is greater than the second preset value; When the leader node detects that there is no node whose resource utilization exceeds the second preset value, it determines that the detection result of the node level is that the resource utilization is normal.

3. The cloud platform node management method according to claim 1, characterized in that: Determining the first node and the second node in the node hierarchy according to the detection result includes: Determine a migration measure of the node level according to the detection result, and determine a first node and a second node in the node level according to the migration measure, wherein the migration measure includes a first migration measure and a second migration measure, a resource utilization rate of the second node corresponding to the first migration measure is lower than a resource utilization rate of the first node, and a resource utilization rate of the second node corresponding to the second migration measure is higher than a resource utilization rate of the first node; When the detection result of the node level is that the resource utilization rate exceeds a first preset value, determining the migration measure of the node level to be a first migration measure, and determining the first node and the second node in the node level according to the first migration measure; When the detection result of the node level is that the resource utilization exceeds a second preset value, the migration measure of the node level is determined to be a second migration measure, and the first node and the second node are determined in the node level according to the second migration measure.

4. The cloud platform node management method according to claim 3, characterized in that: When the detection result at the node level is that the resource utilization rate exceeds a first preset value, determining the migration measure at the node level to be a first migration measure, and determining the first node and the second node at the node level according to the first migration measure includes: When the migration measure of the node level is the first migration measure, the node whose resource utilization exceeds the first preset value in the node level is determined to be the first node, and the virtual machine with the smallest resource utilization in the first node is selected as the virtual machine to be migrated; the second node in the node level is determined to be the node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated, including an idle node without a mounted virtual machine.

5. The cloud platform node management method according to claim 4, characterized in that: After determining that the second node in the node hierarchy is a node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated, including an idle node without a virtual machine mounted thereon, the method further includes: In a case where the second node in the node hierarchy is an idle node, the leader node in the node hierarchy adjusts the working state of the second node to on.

6. The cloud platform node management method according to claim 3, characterized in that: When the detection result at the node level is that the resource utilization rate exceeds a second preset value, determining the migration measure at the node level to be a second migration measure, and determining the first node and the second node at the node level according to the second migration measure includes: When the migration measure of the node level is the second migration measure, the two virtual machines with the lowest resource utilization in the node level are selected as the virtual machines to be migrated, and the node in the node level on which the virtual machines to be migrated are mounted is determined to be the first node; the second node in the node level is determined to be the node whose resource utilization is still less than the first preset value after the migration of the virtual machine to be migrated.

7. The cloud platform node management method according to claim 1, characterized in that: After migrating the virtual machine in the first node to the second node, the leader node adjusts the working state of each node according to the resource utilization rate of each node in the node hierarchy, including: Determine that the nodes without mounted virtual machines are idle nodes; The working state of the idle node is adjusted to node shutdown.

8. A cloud platform node management system, characterized in that: The cloud platform includes a plurality of nodes, the plurality of nodes are divided into a plurality of node levels according to the resource utilization of the nodes, and each of the plurality of node levels includes a leader node, the leader node is a node with the largest resource utilization in the node level, wherein: The leader node is used to detect the resource utilization of other nodes in the corresponding node hierarchy, wherein the leader node is the node with the highest resource utilization in the node hierarchy, and the node hierarchy is a hierarchy obtained by dividing the resource utilization of each node in the cloud platform; a first node and a second node are determined in the node hierarchy based on the detection result, wherein the first node is a node from which the virtual machine needs to be migrated, and the second node is a node to which the virtual machine is migrated, and the first node and the second node are both in the same node hierarchy as the leader node; after migrating the virtual machine in the first node to the second node, the working status of each node is adjusted according to the resource utilization of each node in the node hierarchy.

9. A cloud platform node management device, characterized in that: Applicable to leader nodes in all node tiers, including: A detection module, used to detect resource utilization of other nodes in a corresponding node hierarchy, wherein the leader node is a node with the highest resource utilization in the node hierarchy, and the node hierarchy is a hierarchy obtained by dividing the resource utilization of each node in the cloud platform; a determination module, configured to determine a first node and a second node in the node hierarchy according to the detection result, wherein the first node is a node from which a virtual machine needs to be migrated, the second node is a node to which the virtual machine is migrated, and the first node and the second node are both in the same node hierarchy as the leader node; The adjustment module is used to adjust the working status of each node in the node hierarchy according to the resource utilization rate of each node after the leader node migrates the virtual machine in the first node to the second node.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the cloud platform node management method described in any one of claims 1 to 7.

11. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the program executes the cloud platform node management method described in any one of claims 1 to 7 when running.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the cloud platform node management method described in any one of claims 1 to 7 is implemented.