Adjustment methods, systems, and power lateral isolation clusters
By acquiring the operating load information of power isolation nodes, the number and status of power isolation nodes can be dynamically adjusted, solving the adaptability problem of power horizontal isolation clusters when load fluctuates, and improving the flexibility of load adjustment and cluster stability.
Patent Information
- Application Number
- CN202611321024.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-28
- Publication Date
- 2026-11-17
AI Technical Summary
Existing horizontal isolation power clusters lack adaptability to load fluctuations, resulting in inflexible load adjustments that affect the stability of power services and resource utilization.
By acquiring the operating load information of each power isolation node, the number and load status of the power isolation nodes are dynamically adjusted, and a load balancing and status awareness mechanism is adopted to adapt to load fluctuations.
It improves the adaptability of power horizontal isolation clusters to load fluctuations, enhances the accuracy of scaling up and down decisions and cluster stability, and avoids false triggering caused by occasional anomalies.
Smart Images

Figure CN123123392A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of power engineering technology, and in particular to a method, system and power lateral isolation cluster adjustment method. Background Technology
[0002] With the digitalization and intelligentization of the power system, the amount of power business data has surged, and the randomness and suddenness of business requests have increased significantly, placing higher demands on the stability, real-time performance, and cluster resource utilization of power business isolation forwarding.
[0003] In current power horizontal isolation cluster deployments, users often manually adjust the number of power isolation nodes in the cluster according to their needs, which is not adaptable to load fluctuations. Summary of the Invention
[0004] Therefore, it is necessary to provide an adjustment method, system, and power lateral isolation cluster that can improve the adaptability of power lateral isolation clusters to load fluctuations, in order to address the above-mentioned technical problems.
[0005] Firstly, this application provides a method for adjusting a power horizontal isolation cluster, wherein the power horizontal isolation cluster includes at least two power isolation nodes; including:
[0006] Upon receiving a power isolation request, obtain the operating load information of each power isolation node;
[0007] Based on the operating load information, power isolation requests are assigned to at least one power isolation node, and the operating load information of the power isolation node to which the power isolation request is assigned is updated.
[0008] Based on the updated operational load information, determine the node load status information of each power isolation node, as well as the cluster load status information of the power horizontal isolation cluster.
[0009] The number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the cluster load status information.
[0010] In one embodiment, the number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the cluster load status information, including:
[0011] Based on the node load status information, determine the service request frequency and node load rate of each power isolation node;
[0012] If at least one service request frequency is greater than or equal to the first frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information.
[0013] When the cluster load rate is greater than or equal to the first overall load threshold, determine the number of nodes in each power isolation node whose node load rate is greater than or equal to the first node load threshold, the first duration of the node load rate being greater than or equal to the first node load threshold, and the first degree of exceedance of the node load rate being greater than or equal to the first node load threshold.
[0014] If the number of first nodes is greater than or equal to a first quantity threshold, the first duration is greater than or equal to a first duration threshold, and the first exceedance level is greater than or equal to a first degree threshold, the number of power isolation nodes in the power horizontal isolation cluster is increased.
[0015] In one embodiment, increasing the number of power isolation nodes in a power lateral isolation cluster includes:
[0016] New nodes are pre-configured in the silent buffer pool, and the configuration parameters of new nodes in the silent buffer pool are synchronized with those of the same type of power isolation nodes in the power horizontal isolation cluster.
[0017] Determine the load level of the power horizontal isolation cluster;
[0018] Based on the load level, newly added nodes after configuration parameter synchronization are activated sequentially.
[0019] In one embodiment, newly added nodes after configuration parameter synchronization are activated sequentially according to the load level, including:
[0020] Based on the load level, the task allocation amount is increased for newly added nodes after the configuration parameters are synchronized.
[0021] Determine the load difference between the node load rate of the newly added node and the load threshold of the first node;
[0022] If the load difference is less than or equal to the preset load difference, stop increasing the task allocation of new nodes.
[0023] In one embodiment, adjusting the number of power isolation nodes in the power horizontal isolation cluster based on node load status information and cluster load status information further includes:
[0024] If at least one service request frequency is less than or equal to the second frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information, wherein the second frequency threshold is less than the first frequency threshold.
[0025] When the cluster load rate is less than or equal to the second overall load threshold, determine the number of second nodes in each power isolation node whose node load rate is less than or equal to the second node load threshold, the second duration of the node load rate being less than or equal to the second node load threshold, and the second degree of exceedance of the node load rate being less than or equal to the second node load threshold, wherein the second overall load threshold is less than the first overall load threshold, and the second node load threshold is less than the first node load threshold.
[0026] If the number of second nodes is greater than or equal to the second number threshold, the second duration is greater than or equal to the second duration threshold, and the second exceedance value is greater than or equal to the second degree threshold, reduce the number of power isolation nodes in the power horizontal isolation cluster.
[0027] In one embodiment, reducing the number of power isolation nodes in a power lateral isolation cluster includes:
[0028] Mark redundant nodes in a power lateral isolation cluster;
[0029] Stop assigning tasks to the marked redundant nodes and perform task migration operations;
[0030] After the redundant nodes complete the task migration operation, the redundant nodes are taken offline one by one.
[0031] In one embodiment, performing a task migration operation includes:
[0032] Determine the task migration configuration, which includes the migration target scope of redundant nodes, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules.
[0033] Based on the scope of the migration target, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules, redundant nodes are migrated.
[0034] In one embodiment, power isolation requests are distributed to at least one power isolation node based on the operating load information, including:
[0035] If the data volume of a power isolation request is greater than or equal to a preset data volume threshold, the power isolation request is split into multiple fragmented tasks with a data volume less than the preset data volume threshold.
[0036] Based on the operating load information, determine the node load rate of each power isolation node;
[0037] Based on the load rate of each node, each sharded task is allocated to each power isolation node. The number of sharded tasks allocated to any power isolation node is negatively correlated with the node load rate of the power isolation node.
[0038] Secondly, this application also provides an adjustment system for a power horizontal isolation cluster, wherein the power horizontal isolation cluster includes at least two power isolation nodes; the system includes:
[0039] The information acquisition module is used to acquire the operating load information of each power isolation node when a power isolation request is received.
[0040] The task allocation module is used to allocate power isolation requests to at least one power isolation node based on the running load information, and update the running load information of the power isolation node to which the power isolation request is allocated.
[0041] The load determination module is used to determine the node load status information of each power isolation node and the cluster load status information of the power horizontal isolation cluster based on the updated running load information.
[0042] The cluster adjustment module is used to adjust the number of power isolation nodes in the power horizontal isolation cluster based on the load status information of each node and the cluster load status information.
[0043] Thirdly, this application also provides a power horizontal isolation cluster, which includes a load balancing node, a status-aware node, and at least two power isolation nodes.
[0044] The load balancing node controls the status sensing node and at least two power isolation nodes to implement the steps of the above method.
[0045] Fourthly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:
[0046] Upon receiving a power isolation request, obtain the operating load information of each power isolation node;
[0047] Based on the operating load information, power isolation requests are assigned to at least one power isolation node, and the operating load information of the power isolation node to which the power isolation request is assigned is updated.
[0048] Based on the updated operational load information, determine the node load status information of each power isolation node, as well as the cluster load status information of the power horizontal isolation cluster.
[0049] The number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the cluster load status information.
[0050] Fifthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:
[0051] Upon receiving a power isolation request, obtain the operating load information of each power isolation node;
[0052] Based on the operating load information, power isolation requests are assigned to at least one power isolation node, and the operating load information of the power isolation node to which the power isolation request is assigned is updated.
[0053] Based on the updated operational load information, determine the node load status information of each power isolation node, as well as the cluster load status information of the power horizontal isolation cluster.
[0054] The number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the cluster load status information.
[0055] Sixthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:
[0056] Upon receiving a power isolation request, obtain the operating load information of each power isolation node;
[0057] Based on the operating load information, power isolation requests are assigned to at least one power isolation node, and the operating load information of the power isolation node to which the power isolation request is assigned is updated.
[0058] Based on the updated operational load information, determine the node load status information of each power isolation node, as well as the cluster load status information of the power horizontal isolation cluster.
[0059] The number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the cluster load status information.
[0060] The aforementioned adjustment method, system, and power horizontal isolation cluster, upon receiving a power isolation request, acquires the operational load information of each power isolation node, allocates the power isolation request to at least one power isolation node based on this information, updates the operational load information of the allocated power isolation node, determines the individual node load status information of each power isolation node and the cluster load status information of the power horizontal isolation cluster based on the updated operational load information, and adjusts the number of power isolation nodes in the power horizontal isolation cluster based on the node load status information and the cluster load status information. This application, after allocating the power isolation request to the power isolation node, determines the node load status information at the node level and the cluster load status information at the cluster level in the power horizontal isolation cluster. Based on these two different load dimensions, it adaptively adjusts the number of power isolation nodes in the power horizontal isolation cluster, improving the adaptability of the power horizontal isolation cluster to load fluctuations. Furthermore, it effectively prevents accidental node number adjustments caused by occasional anomalies such as single-node jitter, improving the accuracy of scaling decisions and the stability of the power horizontal isolation cluster. Attached Figure Description
[0061] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0062] Figure 1 This is a flowchart illustrating the adjustment method of a power horizontal isolation cluster in one embodiment;
[0063] Figure 2 This is a flowchart illustrating the adjustment method for a power lateral isolation cluster in another embodiment;
[0064] Figure 3 This is a flowchart illustrating the adjustment method for a power horizontal isolation cluster in yet another embodiment;
[0065] Figure 4 This is a schematic diagram illustrating the process of triggering an adjustment to the number of power isolation nodes in the power horizontal isolation cluster in one embodiment;
[0066] Figure 5 This is a cluster structure diagram of a power lateral isolation cluster in one embodiment;
[0067] Figure 6This is a flowchart illustrating the adjustment method of a power horizontal isolation cluster in one embodiment of this application;
[0068] Figure 7 This is a structural diagram of the adjustment system of a power lateral isolation cluster in one embodiment. Detailed Implementation
[0069] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0070] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.
[0071] In one exemplary embodiment, such as Figure 1 As shown, a method for adjusting a power horizontal isolation cluster is provided. The power horizontal isolation cluster includes at least two power isolation nodes, and includes the following steps S110 to S140. Wherein:
[0072] Step S110: Upon receiving a power isolation request, obtain the operating load information of each power isolation node.
[0073] The power horizontal isolation cluster comprises at least two power isolation nodes, and power isolation requests are communication requests requiring isolation. Power isolation nodes are dedicated horizontal unidirectional security isolation devices for the power system, located between different regions to achieve isolation strength approaching or reaching physical isolation. The power system network is divided into different regions according to security levels. For example: Production Control Region (Zones I / II): responsible for core production operations such as real-time monitoring and dispatching, with the highest security level. Management Information Region (Zones III / IV): responsible for non-production operations such as office work and management, connecting to external networks (such as the Internet), with a lower security level than the Production Control Region.
[0074] Among them, the operating load information is information used to describe the operating load status of the power isolation node, such as CPU utilization, memory usage, data forwarding rate, request frequency, etc.
[0075] In this embodiment, upon receiving a power isolation request, active data collection and / or passive data reception can be performed on each power isolation node to obtain the power isolation node.
[0076] Step S120: Based on the operating load information, allocate the power isolation request to at least one power isolation node, and update the operating load information of the power isolation node to which the power isolation request is allocated.
[0077] This embodiment can allocate power isolation requests to at least one power isolation node based on the operating load information, and update the operating load information of the power isolation node to which the power isolation request is allocated. For example, if the data volume of the power isolation request is greater than or equal to a preset data volume threshold, this embodiment can split the power isolation request into multiple shard tasks with a data volume less than the preset data volume threshold, and allocate each shard task to a power isolation node. This embodiment can also allocate the power isolation request to any power isolation node if the data volume of the power isolation request is less than the preset data volume threshold.
[0078] In some embodiments, allocating power isolation requests to at least one power isolation node based on each operating load information includes:
[0079] Step S121: If the data volume of the power isolation request is greater than or equal to a preset data volume threshold, the power isolation request is split to obtain multiple sharded tasks with a data volume less than the preset data volume threshold.
[0080] Step S122: Determine the node load rate of each power isolation node based on the operating load information.
[0081] Step S123: According to the load rate of each node, each sharded task is allocated to each power isolation node. The number of sharded tasks allocated to any power isolation node is negatively correlated with the node load rate of the power isolation node.
[0082] To avoid load imbalance among power isolation nodes, this embodiment can split power isolation requests into multiple sharded tasks with data volumes less than the preset threshold when the data volume of the power isolation request is greater than or equal to the preset data volume threshold. It is understood that each sharded task can be obtained by evenly distributing the power isolation request or by splitting the power isolation request according to a predetermined data volume. Furthermore, this embodiment can determine the node load rate of each power isolation node based on its operating load information. For example, this embodiment can select any parameter among the CPU utilization, memory usage, and data forwarding rate of the power isolation node as the node load rate. Alternatively, this embodiment can use a weighted calculation of CPU utilization, memory usage, and data forwarding rate to obtain the characteristic value as the node load rate. Then, according to the node load rate, each sharded task is allocated to each power isolation node. The number of sharded tasks allocated to any power isolation node is negatively correlated with the node load rate of the power isolation node. Therefore, this embodiment can allocate more sharded tasks to power isolation nodes with lower node load rates to maintain load balance among the power isolation nodes after allocating the sharded tasks.
[0083] Step S130: Based on the updated operating load information, determine the node load status information of each power isolation node and the cluster load status information of the power horizontal isolation cluster.
[0084] The node load status information describes the load status of each power-isolated node, such as the request frequency and node load rate. The cluster load status information describes the load status of the power-isolated horizontally isolated cluster, such as the cluster load rate.
[0085] For example, in this embodiment, any parameter among CPU utilization, memory occupancy, and data forwarding rate of the power isolation nodes in the updated operating load information can be selected as the node load rate. Alternatively, this embodiment can use a weighted calculation of CPU utilization, memory occupancy, and data forwarding rate to obtain the characteristic value as the node load rate. This embodiment can use the node load rate of each power isolation node as node load status information. Further, this embodiment determines the cluster load status information of the power horizontal isolation cluster based on the node load rate of each power isolation node. For example, this embodiment can calculate the average of the node load rates of each power isolation node in the power horizontal isolation cluster to obtain the cluster load rate of the power horizontal isolation cluster. Alternatively, this embodiment can calculate a weighted sum of the node load rates of each power isolation node in the power horizontal isolation cluster to obtain the cluster load rate of the power horizontal isolation cluster. This embodiment can use the cluster load rate of the power horizontal isolation cluster as cluster load status information.
[0086] Step S140: Adjust the number of power isolation nodes in the power horizontal isolation cluster based on the load status information of each node and the cluster load status information.
[0087] As an example, this embodiment can determine the service request frequency and node load rate of each power isolation node based on the node load status information. If at least one service request frequency is greater than or equal to a first frequency threshold and / or the node load rate is greater than or equal to a first node load threshold, the cluster load rate of the power horizontal isolation cluster can be determined based on the cluster load status information. If the cluster load rate is greater than or equal to a first overall load threshold, the number of power isolation nodes in the power horizontal isolation cluster can be increased.
[0088] As another example, this embodiment can determine the service request frequency and node load rate of each power isolation node based on the node load status information. If at least one service request frequency is greater than or equal to a first frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information. If the cluster load rate is greater than or equal to a first overall load threshold, the number of first nodes in each power isolation node whose node load rate is greater than or equal to the first node load threshold, the first duration of the node load rate being greater than or equal to the first node load threshold, and the first degree of exceedance of the node load rate being greater than or equal to the first node load threshold are determined. If the number of first nodes is greater than or equal to the first number threshold, the first duration is greater than or equal to the first duration threshold, and the first degree of exceedance is greater than or equal to the first degree threshold, the number of power isolation nodes in the power horizontal isolation cluster is increased.
[0089] As an example, this embodiment can determine the service request frequency and node load rate of each power isolation node based on the node load status information. If at least one service request frequency is less than or equal to a second frequency threshold and / or the node load rate is less than or equal to a second node load threshold, the cluster load rate of the power horizontal isolation cluster can be determined based on the cluster load status information. If the cluster load rate is less than or equal to a second overall load threshold, the number of power isolation nodes in the power horizontal isolation cluster can be reduced.
[0090] As another example, this embodiment can determine the cluster load rate of the power horizontal isolation cluster based on the cluster load status information when at least one service request frequency is less than or equal to a second frequency threshold, wherein the second frequency threshold is less than a first frequency threshold. When the cluster load rate is less than or equal to a second overall load threshold, the embodiment determines the number of second nodes in each power isolation node whose node load rate is less than or equal to the second node load threshold, the second duration of the node load rate being less than or equal to the second node load threshold, and the second degree of exceedance of the node load rate being less than or equal to the second node load threshold, wherein the second overall load threshold is less than the first overall load threshold, and the second node load threshold is less than the first node load threshold. When the number of second nodes is greater than or equal to the second number threshold, the second duration is greater than or equal to the second duration threshold, and the second degree of exceedance is greater than or equal to the second degree threshold, the embodiment reduces the number of power isolation nodes in the power horizontal isolation cluster.
[0091] In the aforementioned adjustment method for a horizontally isolated power supply cluster, upon receiving a power isolation request, the method obtains the operating load information of each power isolation node. Based on this information, the power isolation request is allocated to at least one power isolation node, and the operating load information of the allocated power isolation node is updated. Based on the updated operating load information, the individual node load status information of each power isolation node and the cluster load status information of the horizontally isolated power supply cluster are determined. Finally, the number of power isolation nodes in the horizontally isolated power supply cluster is adjusted based on the node load status information and the cluster load status information. This application, after allocating the power isolation request to the power isolation node, determines the node load status information at both the node and cluster dimensions within the horizontally isolated power supply cluster. Therefore, based on these two different load dimensions, the number of power isolation nodes in the horizontally isolated power supply cluster is adaptively adjusted, improving the cluster's adaptability to load fluctuations. Furthermore, it effectively prevents accidental node number adjustments caused by occasional single-node jitter or other anomalies, improving the accuracy of scaling decisions and the cluster stability.
[0092] In one exemplary embodiment, such as Figure 2 As shown, the number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the cluster load status information, including steps S210 to S240. Wherein:
[0093] Step S210: Determine the service request frequency and node load rate of each power isolation node based on the node load status information.
[0094] Step S220: If at least one service request frequency is greater than or equal to the first frequency threshold, determine the cluster load rate of the power horizontal isolation cluster based on the cluster load status information.
[0095] Step S230: When the cluster load rate is greater than or equal to the first overall load threshold, determine the number of first nodes in each power isolation node whose node load rate is greater than or equal to the first node load threshold, the first duration of the node load rate being greater than or equal to the first node load threshold, and the first degree of exceedance of the node load rate being greater than or equal to the first node load threshold.
[0096] Step S240: If the number of first nodes is greater than or equal to a first quantity threshold, the first duration is greater than or equal to a first duration threshold, and the first exceedance value is greater than or equal to a first degree threshold, increase the number of power isolation nodes in the power horizontal isolation cluster.
[0097] Among them, the service request frequency is the frequency at which the power isolation node processes power isolation requests; the first frequency threshold is a preset value for the service request frequency; the first overall load threshold is a preset cluster load rate; the first node load threshold is a preset node load rate; the first quantity threshold is a preset number of nodes; the first duration threshold is a preset duration; and the first degree threshold is a preset value for the degree of exceeding the limit. The first degree of exceeding the limit value is the difference between the node load rate and the first node load threshold, or the ratio of this difference to the first node load threshold, describing the degree to which the node load rate exceeds the first node load threshold.
[0098] For scenarios requiring expansion of a power horizontal isolation cluster (increasing the number of power isolation nodes), this embodiment can extract the service request frequency and node load rate of each power isolation node from the node load status information. Then, if at least one service request frequency is greater than or equal to a first frequency threshold, the cluster load rate of the power horizontal isolation cluster is extracted from the cluster load status information. If the cluster load rate is greater than or equal to a first overall load threshold, the number of nodes in each power isolation cluster with a load rate greater than or equal to the first node load threshold, the first duration of the node load rate being greater than or equal to the first node load threshold, and the first degree of exceedance of the node load rate being greater than or equal to the first node load threshold are determined. Therefore, if the number of nodes is greater than or equal to the first number threshold, the first duration is greater than or equal to the first duration threshold, and the first degree of exceedance is greater than or equal to the first degree threshold, it indicates that the load pressure on the power horizontal isolation cluster is too high, and the number of power isolation nodes in the power horizontal isolation cluster can be increased. For example, the first duration can be the sum of the durations of each power isolation node with a load rate greater than or equal to the first node load threshold, or the average of the durations. The first exceedance level value can be the sum of exceedance level values for all power isolation nodes whose node load rates are greater than or equal to the first node load threshold, or the average of the exceedance level values.
[0099] Furthermore, in this embodiment, if there is no service request frequency greater than or equal to the first frequency threshold, or cluster load rate less than the first overall load threshold, or the number of first nodes less than the first quantity threshold, or the first duration less than the first duration threshold, or the first exceedance level less than the first degree threshold, then the number of power isolation nodes in the power horizontal isolation cluster will not be increased.
[0100] This embodiment judges the frequency of business requests for power isolation requests from the perspective of a single power isolation node, and from the perspective of the cluster, it judges the overall load rate, the number of nodes exceeding the limit, the duration of exceeding the limit, the degree of exceeding the limit, etc., so as to avoid erroneously triggering the operation of increasing the number of power isolation nodes in the power horizontal isolation cluster.
[0101] In some embodiments, increasing the number of power isolation nodes in a power horizontal isolation cluster includes:
[0102] Step S310: Pre-configure the newly added node to the silent buffer pool, and synchronize the configuration parameters of the newly added node in the silent buffer pool with the power isolation node of the same type in the power horizontal isolation cluster.
[0103] Step S320: Determine the load level of the power horizontal isolation cluster;
[0104] Step S330: Activate the newly added nodes sequentially after synchronizing the configuration parameters according to the load level.
[0105] The newly added nodes are the power isolation nodes that are expected to be added to the power horizontal isolation cluster. The silent buffer pool is a data pool that does not receive external real business traffic; that is, newly added nodes in this silent buffer pool remain silent and do not process external real business traffic. Configuration parameters can include rules for handling power isolation requests, such as configuration rules, forwarding rules, and scheduling rules.
[0106] This embodiment can pre-configure new nodes in a silent buffer pool, and synchronize the configuration parameters of new nodes in the silent buffer pool with those of the same type of power isolation nodes in the power horizontal isolation cluster. For example, the power horizontal isolation cluster can include master nodes and slave nodes, and the types of power isolation nodes can include both master nodes and slave nodes. Therefore, new nodes do not process power isolation requests immediately, but instead synchronize their configuration parameters first. This avoids the situation where new nodes directly access real service traffic, which could lead to unsuitable new nodes interfering with the scheduling of the power horizontal isolation cluster.
[0107] Therefore, this embodiment can determine the load level of the power horizontal isolation cluster. The load level can be a preset value or determined based on the cluster load rate of the power horizontal isolation cluster. The load level is positively correlated with the cluster load rate of the power horizontal isolation cluster. Then, according to the load level, newly added nodes after configuration parameter synchronization can be activated sequentially. It can be understood that activating a new node means assigning tasks to the new node. For example, according to the load level, the task allocation amount of the newly added nodes after configuration parameter synchronization is increased incrementally, and the load difference between the node load rate of the new node and the first node load threshold is determined. If the load difference is less than or equal to the preset load difference, the increase of the task allocation amount of the new node is stopped. Thus, this embodiment can also increase the task allocation amount of the newly added nodes after configuration parameter synchronization according to the task increase rate positively correlated with the load level. In addition, this embodiment can also activate the newly added nodes after configuration parameter synchronization sequentially according to the number of nodes activated positively correlated with the load level. For example, if the load level is high, multiple new nodes are activated at once, and if the load level is low, one new node is activated at once.
[0108] Therefore, this embodiment pre-configures new nodes in a silent buffer pool and synchronizes the configuration parameters of new nodes in the silent buffer pool with those of the same type of power isolation nodes in the power horizontal isolation cluster, thereby preventing unsuitable new nodes from disrupting cluster scheduling. Furthermore, it determines the load level of the power horizontal isolation cluster and, based on the load level, sequentially activates the new nodes after configuration parameter synchronization, thus maintaining a gradual access of new nodes and avoiding a large-scale impact on the cluster caused by a single influx of new nodes.
[0109] In some embodiments, newly added nodes after configuration parameter synchronization are activated sequentially according to the load level, including:
[0110] Step S331: According to the load level, increase the task allocation amount for newly added nodes after the configuration parameters are synchronized.
[0111] Step S332: Determine the load difference between the node load rate of the newly added node and the load threshold of the first node;
[0112] Step S333: If the load difference is less than or equal to the preset load difference, stop increasing the task allocation amount of the newly added node.
[0113] This embodiment can incrementally increase the task allocation for newly added nodes after configuration parameter synchronization according to load levels. The task allocation is positively correlated with the load level, thus increasing the task allocation for newly added nodes faster as the load on the power horizontal isolation cluster increases, leading to faster load reduction. Furthermore, to prevent excessive load on newly added nodes, this embodiment can determine the load difference between the node load rate of the newly added node and a first node load threshold. If the load difference is less than or equal to a preset load difference, the incremental increase of the task allocation for the newly added node is stopped to avoid exceeding the first node load threshold.
[0114] This embodiment determines the load difference between the load rate of a new node and the load threshold of a first node by incrementally increasing the task allocation amount of the newly added node according to the load level and after the configuration parameters are synchronized. If the load difference is less than or equal to a preset load difference, the increment of the task allocation amount of the new node is stopped. Thus, while ensuring that the load of the new node does not exceed the load threshold of the first node, the speed at which the task allocation amount of the new node is increased is adjusted according to the load level corresponding to the power horizontal isolation cluster to adapt to the load situation of the power horizontal isolation cluster.
[0115] In one exemplary embodiment, such as Figure 3 As shown, the number of power isolation nodes in the power horizontal isolation cluster is adjusted based on node load status information and cluster load status information, and also includes:
[0116] Step S410: If at least one service request frequency is less than or equal to the second frequency threshold, determine the cluster load rate of the power horizontal isolation cluster based on the cluster load status information, wherein the second frequency threshold is less than the first frequency threshold.
[0117] Step S420: When the cluster load rate is less than or equal to the second overall load threshold, determine the number of second nodes in each power isolation node whose node load rate is less than or equal to the second node load threshold, the second duration of the node load rate being less than or equal to the second node load threshold, and the second degree of exceedance of the node load rate being less than or equal to the second node load threshold, wherein the second overall load threshold is less than the first overall load threshold, and the second node load threshold is less than the first node load threshold.
[0118] Step S430: If the number of second nodes is greater than or equal to the second quantity threshold, the second duration is greater than or equal to the second duration threshold, and the second exceedance value is greater than or equal to the second degree threshold, reduce the number of power isolation nodes in the power horizontal isolation cluster.
[0119] Specifically, the second frequency threshold is lower than the first frequency threshold, the second overall load threshold is lower than the first overall load threshold, and the second node load threshold is lower than the first node load threshold. The second quantity threshold may or may not be equal to the first quantity threshold. The second duration threshold may or may not be equal to the first duration threshold. The second severity threshold may or may not be equal to the first severity threshold. The second exceedance severity value may or may not be equal to the first exceedance severity value.
[0120] For scenarios requiring scaling down (reducing the number of power isolation nodes) of a power horizontal isolation cluster, this embodiment can extract the service request frequency and node load rate of each power isolation node from the node load status information. Then, when at least one service request frequency is less than or equal to a second frequency threshold, the cluster load rate of the power horizontal isolation cluster is extracted from the cluster load status information. When the cluster load rate is less than or equal to a second overall load threshold, the number of second nodes with a load rate less than or equal to the second node load threshold, the second duration of the load rate being less than or equal to the second node load threshold, and the second degree of exceedance of the load rate being less than or equal to the second node load threshold are determined. Furthermore, if the number of second nodes is greater than or equal to a second number threshold, the second duration is greater than or equal to a second duration threshold, and the second degree of exceedance is greater than or equal to a second degree threshold, it indicates that the load pressure of the power horizontal isolation cluster is too low, and the number of power isolation nodes in the power horizontal isolation cluster can be reduced. For example, the second duration can be the sum of the durations of all power isolation nodes with a load rate less than or equal to the second node load threshold, or the average of the durations. The first exceedance value can be the sum of exceedance values for all power isolation nodes whose respective node load rates are less than or equal to the second node load threshold, or the average of the exceedance values.
[0121] Furthermore, this embodiment can maintain the same number of power isolation nodes in the power horizontal isolation cluster if there is no service request frequency less than or equal to the second frequency threshold, cluster load rate greater than the second overall load threshold, second number of nodes greater than the second number threshold, second duration greater than the second duration threshold, or second exceedance level less than the second degree threshold.
[0122] like Figure 4 As shown, this embodiment uses a three-level judgment to determine whether a power horizontal isolation cluster should be expanded or contracted. The first level is to judge the frequency of service requests for each power isolation node. The second level is to judge the overall load rate of the power horizontal isolation cluster. The third level is to evaluate the situation of exceeding the standard by considering the number of nodes whose load rate is less than or equal to the second node load threshold, the second duration of the node load rate being less than or equal to the second node load threshold, and the second degree of exceeding the standard of the node load rate being less than or equal to the second node load threshold. If all three judgments pass, it is determined that the power horizontal isolation cluster can be expanded or contracted. Then, the three-stage process of expansion and contraction can be executed: Silent management / pre-positioning stage: During expansion, the newly added nodes are pre-positioned into the silent buffer pool, which is silently managed and monitored by the load balancing nodes to avoid unsuitable nodes interfering with cluster scheduling; during contraction, redundant nodes are marked as silent and waiting to go offline, and no new tasks are assigned to them. Configuration Synchronization Phase: During expansion, pre-configured nodes automatically synchronize the collection configuration, forwarding rules, and scheduling strategies of existing nodes of the same type, requiring no manual intervention. During contraction, nodes to be taken offline synchronize task migration configurations. Task migration configurations include at least the scope of migration targets, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules. Gradual Entry / Decommissioning Phase: During expansion, nodes that have completed configuration synchronization are activated one by one according to load tiers, increasing task allocation and monitoring load in real time, distinguishing between instantaneous and continuous loads, and performing node recycling or solidification accordingly. During contraction, redundant nodes are gradually decommissioned after completing the migration of their executing tasks, ensuring continuity of power services.
[0123] This embodiment judges the frequency of business requests for power isolation requests from the perspective of a single power isolation node, and from the perspective of the cluster, it judges the overall load rate, the number of nodes exceeding the limit, the duration of exceeding the limit, the degree of exceeding the limit, etc., so as to avoid erroneously triggering the reduction of the number of power isolation nodes in the power horizontal isolation cluster.
[0124] In some embodiments, reducing the number of power isolation nodes in a power lateral isolation cluster includes:
[0125] Step S510: Mark redundant nodes in the power horizontal isolation cluster;
[0126] Step S520: Stop assigning tasks to the marked redundant nodes and perform task migration operation;
[0127] Step S530: After the redundant nodes complete the task migration operation, the redundant nodes are taken offline one by one.
[0128] In this context, redundant nodes are power isolation nodes that are expected to be taken offline from the power horizontal isolation cluster. The task migration operation involves migrating unfinished tasks on redundant nodes to power isolation nodes other than the initial redundant nodes in the power horizontal isolation cluster.
[0129] For example, this embodiment can determine the number of nodes to be reduced based on the overall load rate. The number of nodes to be reduced is negatively correlated with the overall load rate; that is, the lower the overall load rate, the greater the number of nodes to be reduced. This embodiment can also determine the number of nodes to be reduced based on the overall load rate, where the number of nodes to be reduced is also negatively correlated with the overall load rate; that is, the lower the overall load rate, the greater the number of nodes to be reduced.
[0130] This embodiment can select power isolation nodes with the smallest node reduction from the power horizontal isolation cluster as redundant nodes. For example, this embodiment can randomly select power isolation nodes with the smallest node reduction from the power horizontal isolation cluster as redundant nodes. Furthermore, to reduce task migration and improve migration efficiency, this embodiment can select the power isolation node with the smallest node load rate and smallest node reduction from the power horizontal isolation cluster as a redundant node.
[0131] This embodiment can mark redundant nodes in a power horizontal isolation cluster to facilitate accurate handling of these redundant nodes laterally. This involves stopping task allocation to the marked redundant nodes and performing task migration operations. After the redundant nodes complete their task migration, they are sequentially taken offline. Therefore, this embodiment can gradually take redundant nodes offline after they have completed their ongoing task migration, ensuring continuity of power services.
[0132] In some embodiments, performing a task migration operation includes:
[0133] Step S521: Determine the task migration configuration, which includes the migration target range of redundant nodes, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules.
[0134] Step S522: Based on the scope of the migration object, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules, perform task migration for redundant nodes.
[0135] This embodiment can pre-build a task migration configuration to standardize task migration operations. The task migration configuration includes the migration target scope of redundant nodes, migration priority, maximum number of tasks to be migrated per batch, migration integrity verification rules, and exception rollback rules. The migration target scope describes the tasks to be migrated from the redundant nodes to be taken offline, such as all currently executing tasks, including tasks that have completed some processing but are not yet finished. The migration priority is the order in which tasks are migrated. For example, the migration priority can be determined according to task type (e.g., real-time control tasks are prioritized, followed by batch data tasks) and / or task execution time (tasks with shorter execution time are migrated first). The maximum number of tasks to be migrated per batch is the maximum number of tasks that can be migrated in a single batch, for example, no more than 10, to avoid a large number of tasks migrating to the target node (power isolation nodes not marked as redundant nodes) simultaneously, causing instantaneous overload of the target node. The migration integrity verification rules are the rules for verifying whether the task migration is complete. For example, after a task is migrated to the target node, the target node returns takeover confirmation information to the load balancing node. Only after the load balancing node confirms the confirmation can the task migration be marked as complete. The abnormal rollback rule is a rule for rolling back in case of an abnormality during the task migration process. For example, if the target node exceeds the load threshold or the task takeover fails during the migration process, the current batch migration will be stopped, the migrated tasks will be rolled back to the original node, and an alarm will be triggered.
[0136] Therefore, this embodiment can perform task migration on redundant nodes according to the scope of migration objects, migration priority, upper limit of the number of migration tasks in each batch, migration integrity verification rules and abnormal rollback rules, thereby standardizing the task migration process and ensuring the stable execution of tasks.
[0137] In some embodiments, such as Figure 5 As shown, the power horizontal isolation cluster includes a load balancing node, a status-aware node, and at least two power isolation nodes;
[0138] The load balancing node controls the status sensing node and at least two power isolation nodes to implement the steps of the above method.
[0139] For example, this embodiment deploys a load balancing node, at least two power isolation nodes, and one status awareness node. Each node establishes a communication connection to form a horizontally isolated power cluster. The load balancing node performs global scheduling, the power isolation node performs power service isolation forwarding, and the status awareness node performs cluster status collection. Figure 6As shown, the status-aware node actively detects and collects the operating parameters of the power isolation nodes, receives passively reported information, calculates the real-time load rate of each power isolation node and compares it with a preset load threshold, and synchronizes the integrated status load data to the load balancing node according to a preset time window. The load balancing node performs a global situational prediction of the power service load, splits service requests exceeding the preset volume threshold, allocates tasks preferentially according to the load margin of each power isolation node, monitors the task execution status in real time and updates the node load, forming a load scheduling closed loop. The load balancing node uses a three-level progressive judgment logic of request frequency warning, overall load confirmation, and final review of nodes exceeding the limit to trigger cluster expansion and contraction, which is executed according to the process of silent management / pre-configuration, configuration synchronization, and gradient entry / exit. During expansion, pre-configured nodes are activated one by one in the buffer pool to distinguish between instantaneous and continuous load to avoid resource waste. During contraction, nodes are gradually taken offline to ensure service continuity.
[0140] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.
[0141] Based on the same inventive concept, this application also provides an adjustment system for a power lateral isolation cluster to implement the adjustment method for the power lateral isolation cluster described above. The solution provided by this system is similar to the implementation described in the above method. Therefore, the specific limitations of one or more power lateral isolation cluster adjustment system embodiments provided below can be found in the limitations of the power lateral isolation cluster adjustment method described above, and will not be repeated here.
[0142] In one exemplary embodiment, such as Figure 7 As shown, a power horizontal isolation cluster adjustment system is provided, the power horizontal isolation cluster including at least two power isolation nodes; including: an information acquisition module 610, a task allocation module 620, a load determination module 630, and a cluster adjustment module 640, wherein:
[0143] The information acquisition module 610 is used to acquire the operating load information of each power isolation node when a power isolation request is received.
[0144] The task allocation module 620 is used to allocate power isolation requests to at least one power isolation node based on the running load information, and update the running load information of the power isolation node to which the power isolation request is allocated.
[0145] The load determination module 630 is used to determine the node load status information of each power isolation node and the cluster load status information of the power horizontal isolation cluster based on the updated operating load information.
[0146] The cluster adjustment module 640 is used to adjust the number of power isolation nodes in the power horizontal isolation cluster based on the load status information of each node and the cluster load status information.
[0147] In some embodiments, the cluster adjustment module 640 is further configured to:
[0148] Based on the node load status information, the service request frequency and node load rate of each power isolation node are determined. If at least one service request frequency is greater than or equal to a first frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information. If the cluster load rate is greater than or equal to a first overall load threshold, the number of nodes in each power isolation node whose node load rate is greater than or equal to the first node load threshold, the first duration of the node load rate being greater than or equal to the first node load threshold, and the first degree of exceedance of the node load rate being greater than or equal to the first node load threshold are determined. If the number of nodes is greater than or equal to the first number threshold, the first duration is greater than or equal to the first duration threshold, and the first degree of exceedance is greater than or equal to the first degree threshold, the number of power isolation nodes in the power horizontal isolation cluster is increased.
[0149] In some embodiments, the cluster adjustment module 640 is further configured to:
[0150] New nodes are pre-configured in the silent buffer pool. The configuration parameters of the new nodes in the silent buffer pool are synchronized with those of the same type of power isolation nodes in the power horizontal isolation cluster. The load level of the power horizontal isolation cluster is determined. Based on the load level, the new nodes with synchronized configuration parameters are activated sequentially.
[0151] In some embodiments, the cluster adjustment module 640 is further configured to:
[0152] Based on the load level, the task allocation amount is increased incrementally for newly added nodes after the configuration parameters are synchronized. The load difference between the node load rate of the newly added node and the load threshold of the first node is determined. If the load difference is less than or equal to the preset load difference, the increment of the task allocation amount for the newly added node is stopped.
[0153] In some embodiments, the cluster adjustment module 640 is further configured to:
[0154] If at least one service request frequency is less than or equal to a second frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information, wherein the second frequency threshold is less than the first frequency threshold. If the cluster load rate is less than or equal to a second overall load threshold, the number of second nodes in each power isolation node whose node load rate is less than or equal to the second node load threshold, the second duration of the node load rate being less than or equal to the second node load threshold, and the second degree of exceedance of the node load rate being less than or equal to the second node load threshold are determined, wherein the second overall load threshold is less than the first overall load threshold, and the second node load threshold is less than the first node load threshold. If the number of second nodes is greater than or equal to the second number threshold, the second duration is greater than or equal to the second duration threshold, and the second degree of exceedance is greater than or equal to the second degree threshold, the number of power isolation nodes in the power horizontal isolation cluster is reduced.
[0155] In some embodiments, the cluster adjustment module 640 is further configured to:
[0156] Mark redundant nodes in the power horizontal isolation cluster, stop assigning tasks to the marked redundant nodes, and perform task migration operations. After the redundant nodes complete the task migration operations, take the redundant nodes offline one by one.
[0157] In some embodiments, the cluster adjustment module 640 is further configured to:
[0158] Determine the task migration configuration, which includes the migration target scope of redundant nodes, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules. Perform task migration on redundant nodes based on the migration target scope, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules.
[0159] In some embodiments, the task allocation module 620 is further configured to:
[0160] If the data volume of a power isolation request is greater than or equal to a preset data volume threshold, the power isolation request is split into multiple sharded tasks with a data volume less than the preset data volume threshold. Based on the operating load information, the node load rate of each power isolation node is determined. According to the node load rate, each sharded task is allocated to each power isolation node. The number of sharded tasks allocated to any power isolation node is negatively correlated with the node load rate of the power isolation node.
[0161] The modules in the aforementioned power horizontal isolation cluster adjustment system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.
[0162] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
[0163] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.
[0164] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.
[0165] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.
[0166] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0167] The above embodiments merely illustrate several implementation methods of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.
Claims
1. A method for adjusting a power horizontal isolation cluster, characterized in that, The power horizontal isolation cluster includes at least two power isolation nodes; the method includes: Upon receiving a power isolation request, obtain the operating load information of each of the power isolation nodes; Based on the aforementioned operating load information, the power isolation request is allocated to at least one power isolation node, and the operating load information of the power isolation node to which the power isolation request is allocated is updated; Based on the updated operational load information, determine the node load status information of each of the power isolation nodes, as well as the cluster load status information of the power horizontal isolation cluster. The number of power isolation nodes in the power horizontal isolation cluster is adjusted based on the load status information of each node and the load status information of the cluster.
2. The method according to claim 1, characterized in that, The step of adjusting the number of power isolation nodes in the power horizontal isolation cluster based on the load status information of each node and the cluster load status information includes: Based on the node load status information, determine the service request frequency and node load rate of each of the power isolation nodes; If at least one service request frequency is greater than or equal to a first frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information. When the cluster load rate is greater than or equal to the first overall load threshold, determine the number of first nodes in each of the power isolation nodes whose node load rate is greater than or equal to the first node load threshold, the first duration of the node load rate being greater than or equal to the first node load threshold, and the first degree of exceedance of the node load rate being greater than or equal to the first node load threshold. If the number of the first nodes is greater than or equal to the first quantity threshold, the first duration is greater than or equal to the first duration threshold, and the first exceedance value is greater than or equal to the first degree threshold, the number of power isolation nodes in the power horizontal isolation cluster is increased.
3. The method according to claim 2, characterized in that, Increasing the number of power isolation nodes in the power horizontal isolation cluster includes: New nodes are pre-configured in the silent buffer pool, and the configuration parameters of new nodes in the silent buffer pool are synchronized with those of the same type of power isolation nodes in the power horizontal isolation cluster. Determine the load level of the power horizontal isolation cluster; Based on the load level, newly added nodes after configuration parameter synchronization are activated sequentially.
4. The method according to claim 3, characterized in that, The step of sequentially activating newly added nodes after configuration parameter synchronization according to the load level includes: According to the load level, the task allocation amount is increased for newly added nodes after the configuration parameters are synchronized. Determine the load difference between the node load rate of the newly added node and the load threshold of the first node; If the load difference is less than or equal to a preset load difference, the increment of the task allocation for the newly added node is stopped.
5. The method according to claim 2, characterized in that, The step of adjusting the number of power isolation nodes in the power horizontal isolation cluster based on the node load status information and the cluster load status information further includes: If at least one service request frequency is less than or equal to a second frequency threshold, the cluster load rate of the power horizontal isolation cluster is determined based on the cluster load status information, wherein the second frequency threshold is less than the first frequency threshold. When the cluster load rate is less than or equal to the second overall load threshold, the number of second nodes in each of the power isolation nodes whose node load rate is less than or equal to the second node load threshold, the second duration of the node load rate being less than or equal to the second node load threshold, and the second degree of exceedance of the node load rate being less than or equal to the second node load threshold are determined, wherein the second overall load threshold is less than the first overall load threshold, and the second node load threshold is less than the first node load threshold. If the number of the second node is greater than or equal to the second number threshold, the second duration is greater than or equal to the second duration threshold, and the second exceedance value is greater than or equal to the second degree threshold, the number of power isolation nodes in the power horizontal isolation cluster shall be reduced.
6. The method according to claim 5, characterized in that, Reducing the number of power isolation nodes in the power horizontal isolation cluster includes: Mark redundant nodes in the power lateral isolation cluster; Stop assigning tasks to the marked redundant nodes and perform task migration operations; After the redundant nodes complete the task migration operation, the redundant nodes are taken offline in sequence.
7. The method according to claim 6, characterized in that, The execution task migration operation includes: Determine the task migration configuration, wherein the task migration configuration includes the migration target range of the redundant nodes, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and abnormal rollback rules; Based on the migration target range, migration priority, maximum number of migration tasks per batch, migration integrity verification rules, and exception rollback rules, the redundant nodes are migrated.
8. The method according to any one of claims 1 to 7, characterized in that, The step of allocating the power isolation request to at least one power isolation node based on the operational load information includes: If the data volume of the power isolation request is greater than or equal to a preset data volume threshold, the power isolation request is split into multiple fragmented tasks with a data volume less than the preset data volume threshold. Based on the operating load information, determine the node load rate of each of the power isolation nodes; According to the load rate of each node, each sharding task is allocated to each power isolation node, wherein the number of sharding tasks allocated to any power isolation node is negatively correlated with the node load rate of the power isolation node.
9. An adjustment system for a power lateral isolation cluster, characterized in that, The power horizontal isolation cluster includes at least two power isolation nodes; the system includes: The information acquisition module is used to acquire the operating load information of each of the power isolation nodes when a power isolation request is received. The task allocation module is used to allocate the power isolation request to at least one power isolation node according to the running load information, and update the running load information of the power isolation node to which the power isolation request is allocated; The load determination module is used to determine the node load status information of each of the power isolation nodes and the cluster load status information of the power horizontal isolation cluster based on the updated running load information. The cluster adjustment module is used to adjust the number of power isolation nodes in the power horizontal isolation cluster based on the load status information of each node and the cluster load status information.
10. A power lateral isolation cluster, characterized in that, The power horizontal isolation cluster includes a load balancing node, a status-aware node, and at least two power isolation nodes; The load balancing node controls the state-aware node and at least two power isolation nodes to implement the steps of the method according to any one of claims 1 to 8.
