Distributed Database Performance Data Interaction Method and System

Abstract

The present invention discloses a method and system for performance data interaction under a distributed database, belonging to the technical field of distributed databases. The technical problem to be solved is how to achieve performance monitoring data interaction on the premise of meeting the installation, deployment, operation and maintenance of the KaiWuDB database. The performance monitoring data under the KaiWuDB database is realized based on asymmetric encryption and the gossip protocol. The application of the openssl asymmetric encryption algorithm improves the security of the monitoring data interaction of the KaiWuDB database. The gossip protocol is used for the monitoring data interaction of the KaiWuDB database, ensuring the scalability, fault tolerance and consistency convergence of the cluster scale.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed databases, and in particular to a method and system for performance data interaction under a distributed database. Background Art

[0002] Monitoring data interaction is to meet the requirements of the performance monitoring function of the KaiWuDB database and is a key issue for ensuring the normal operation of the database. Although traditional performance data interaction management methods collect performance metric data as much as possible and then report the data to meet the monitoring requirements of cluster performance data, for the special customized performance monitoring requirements of the KaiWuDB database, such as multi-center in different regions, customized projects, customized project metrics, customized display methods, customized data reporting methods, etc., traditional performance monitoring methods cannot meet these customized requirements and cannot ensure the requirements of the performance monitoring function module of the KaiWuDB database for security, fault tolerance, scalability, consistent convergence, etc., which greatly limits the actual requirements such as the installation, deployment, operation, and maintenance of the KaiWuDB database.

[0003] How to achieve performance monitoring data interaction on the premise of meeting the installation, deployment, operation, and maintenance of the KaiWuDB database is a technical problem to be solved. Summary of the Invention

[0004] The technical task of the present invention is to provide a method and system for performance data interaction under a distributed database to solve the technical problem of how to achieve performance monitoring data interaction on the premise of meeting the installation, deployment, operation, and maintenance of the KaiWuDB database in view of the above deficiencies.

[0005] In a first aspect, a method for performance data interaction under a distributed database according to the present invention implements performance monitoring data under the KaiWuDB database based on asymmetric encryption and the gossip protocol. Based on the performance monitoring cluster of the KaiWuDB database, the following steps are performed:

[0006] For a new performance monitoring node, record its configuration information and openssl certificate in the system table;

[0007] For a new performance monitoring node, after encrypting the configuration information based on the openssl asymmetric encryption algorithm, load the encrypted configuration information and the openssl certificate into memory;

[0008] For a new performance monitoring node, add the performance monitoring node to the performance monitoring cluster and a monitoring sub-cluster. The performance monitoring cluster includes multiple levels of monitoring sub-clusters, and each monitoring sub-cluster includes performance monitoring nodes and other nodes. For each monitoring sub-cluster, the performance monitoring node that successfully joins the monitoring sub-cluster pushes its public key to other nodes in the monitoring sub-cluster through the gossip protocol. Other nodes asymmetrically encrypt their performance monitoring data with the received public key and periodically push the encrypted performance monitoring data to the corresponding performance monitoring node. After the performance monitoring node asymmetrically decrypts the received performance monitoring data based on its public key, it pushes the performance monitoring data to the upper-level performance monitoring data;

[0009] For the monitoring sub-cluster, regularly determine whether the performance monitoring node and the performance monitoring data are normal. If normal, synchronize and summarize the performance monitoring data of the monitoring sub-cluster where the performance monitoring node is located to the upper-level performance monitoring node through the performance monitoring node, and synchronize and summarize all performance monitoring data through the performance monitoring node at the highest level, and write the summarized performance monitoring data to disk;

[0010] For each monitoring sub-cluster, regularly determine whether the number of nodes in the monitoring sub-cluster meets the threshold. If it exceeds the threshold, add the performance monitoring node that joined the monitoring sub-cluster later to other monitoring sub-clusters or create a new monitoring sub-cluster according to the time data when the performance monitoring node joined the monitoring sub-cluster, and prompt the user;

[0011] Customize the UI interface based on the configuration information and display the performance monitoring data through the customized UI interface.

[0012] Preferably, for a new performance monitoring node, record its configuration information and the openssl certificate in the system table, including the following steps:

[0013] For a new performance monitoring node, determine whether its configuration information and the openssl certificate exist in the system table. If not, determine that the new performance monitoring node has not been started, and record its configuration information and the openssl certificate in the system table.

[0014] Preferably, when adding the performance monitoring node to the performance monitoring cluster and a monitoring sub-cluster, if the addition fails, there are the following scenarios:

[0015] Prompt that the addition to the performance monitoring cluster fails, determine that the new performance monitoring node is offline, and record it in the system table;

[0016] If the new performance monitoring node does not receive the performance monitoring data pushed by other nodes in the monitoring sub-cluster within a predetermined time period, determine that other nodes in the monitoring sub-cluster are offline, and record it in the system table.

[0017] Preferably, when periodically determining whether the performance monitoring data is normal, it is determined whether the metrics of the performance monitoring data exceed the thresholds specified in the configuration information. If so, it is prompted that the performance monitoring data is abnormal and recorded in the system table.

[0018] In a second aspect, a performance data interaction system under a distributed database according to the present invention is used to implement performance monitoring data under the KaiWuDB database through a performance data interaction method under a distributed database as described in any item of the first aspect, based on asymmetric encryption and the gossip protocol. The system includes:

[0019] A configuration module, which is used to perform the following: for a new performance monitoring node, record its configuration information and the openssl certificate in the system table;

[0020] A loading module, which is used to perform the following: for a new performance monitoring node, after encrypting the configuration information based on the openssl asymmetric encryption algorithm, load the encrypted configuration information and the openssl certificate into the memory;

[0021] A cluster update module, which is used to perform the following: for a new performance monitoring node, add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. The performance monitoring cluster includes multiple levels of monitoring sub-clusters, and each monitoring sub-cluster includes performance monitoring nodes and other nodes. For each monitoring sub-cluster, the performance monitoring node that successfully joins the monitoring sub-cluster pushes its public key to other nodes in the monitoring sub-cluster through the gossip protocol. Other nodes asymmetrically encrypt their performance monitoring data with the received public key and periodically push the encrypted performance monitoring data to the corresponding performance monitoring node. After the performance monitoring node asymmetrically decrypts the received performance monitoring data based on its public key, it pushes the performance monitoring data to the upper-level performance monitoring data;

[0022] A performance data collection module, which is used to perform the following: for the monitoring sub-cluster, periodically determine whether the performance monitoring node and the performance monitoring data are normal. If normal, synchronize and summarize the performance monitoring data of the monitoring sub-cluster where the performance monitoring node is located to the upper-level performance monitoring node through the performance monitoring node, and synchronize and summarize all the performance monitoring data through the performance monitoring node at the highest level, and store the summarized performance monitoring data on disk;

[0023] A sub-cluster detection module, which is used to perform the following: for each monitoring sub-cluster, periodically determine whether the number of nodes in the monitoring sub-cluster meets the threshold. If it exceeds the threshold, according to the time data when the performance monitoring node joins the monitoring sub-cluster, add the performance monitoring node that joined the monitoring sub-cluster later to other monitoring sub-clusters or create a new monitoring sub-cluster, and prompt the user;

[0024] A data display module, which is used to perform the following: customize the UI interface based on configuration information, and display performance monitoring data through the customized UI interface.

[0025] Preferably, for a new performance monitoring node, the configuration module is used to perform the following to record its configuration information and openssl certificate in the system table:

[0026] For a new performance monitoring node, check whether its configuration information and openssl certificate exist in the system table. If not, determine that the new performance monitoring node has not been started, and record its configuration information and openssl certificate in the system table.

[0027] Preferably, when adding a performance monitoring node to a performance monitoring cluster and adding it to a monitoring sub-cluster, if the addition fails, the cluster update module is used to perform the following:

[0028] Prompt that the addition to the performance monitoring cluster fails, determine that the new performance monitoring node is offline, and record it in the system table;

[0029] If a new performance monitoring node does not receive performance monitoring data pushed by other nodes in the monitoring sub-cluster within a predetermined time period, determine that other nodes in the monitoring sub-cluster are offline, and record it in the system table.

[0030] Preferably, when periodically determining whether the performance monitoring data is normal, the performance data collection module is used to perform the following: check whether the metrics of the performance monitoring data exceed the threshold specified in the configuration information. If so, prompt that the performance monitoring data is abnormal, and record it in the system table.

[0031] The performance data interaction method and system under the distributed database of the present invention have the following advantages: The performance monitoring data interaction method under KaiWuDB based on asymmetric encryption and gossip protocol meets the functional module requirements for performance data monitoring of the KaiWuDB database. The application of the openssl asymmetric encryption algorithm improves the security of the monitoring data interaction of the KaiWuDB database. The gossip protocol is used for the monitoring data interaction of the KaiWuDB database, which ensures the scalability, fault tolerance, and consistency convergence of the cluster scale, and the message propagation speed can reach log(N). Brief Description of the Drawings

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for use in the embodiments or the description of the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0033] The present invention will be further described below with reference to the accompanying drawings.

[0034] Figure 1 It is the logical architecture of the performance monitoring function of KaiwuDB;

[0035] Figure 2 It is the flowchart of the performance data interaction method under a distributed database in Embodiment 1;

[0036] Figure 3 It is the distribution diagram of performance monitoring nodes in the performance data interaction method under a distributed database in Embodiment 1. Specific embodiments

[0037] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it. However, the specific embodiments cited are not intended to limit the present invention. Without conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0038] The embodiments of the present invention provide a performance data interaction method and system under a distributed database, which are used to solve the technical problem of how to realize the interaction of performance monitoring data on the premise of meeting the installation, deployment, operation and maintenance of the KaiWuDB database.

[0039] Embodiment 1:

[0040] For the interaction of performance monitoring data under the KaiWuDB database, the performance index data of KaiWuDB is captured and integrated through the system resource tool and the built-in health monitoring module of KaiWuDB. The performance monitoring data of KaiWuDB is divided into operation index data (cluster scale, cluster resource occupancy, resource operation time, etc.), performance index data (data scale, insertion speed per unit time, query speed per unit time, P95, P99, etc.), health index data (whether the node is offline, whether the node is stuck, whether the node port is blocked, etc.), and customized index data (meeting the data reporting requirements of customized projects). The key steps are as Figure 1 shown:

[0041] (1) The nodes for performance monitoring join the performance monitoring cluster of KaiWuDB;

[0042] (2) Collect and report the performance data of the nodes;

[0043] (3) The performance monitoring nodes summarize and negotiate the performance data within the cluster;

[0044] (4) Disk the performance data within the KaiwuDB cluster.

[0045] A method for performance data interaction under a distributed database according to the present invention realizes performance monitoring data under the KaiWuDB database based on asymmetric encryption and the gossip protocol. Based on the performance monitoring cluster of the KaiWuDB database, the method includes the following steps:

[0046] S100. For a new performance monitoring node, record its configuration information and the openssl certificate in the system table;

[0047] S200. For a new performance monitoring node, after encrypting the configuration information based on the openssl asymmetric encryption algorithm, load the encrypted configuration information and the openssl certificate into the memory;

[0048] S300. For a new performance monitoring node, add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. The performance monitoring cluster includes multiple levels of monitoring sub-clusters. Each monitoring sub-cluster includes performance monitoring nodes and other nodes. For each monitoring sub-cluster, the performance monitoring node that successfully joins the monitoring sub-cluster pushes its public key to other nodes within the monitoring sub-cluster through the gossip protocol. Other nodes asymmetrically encrypt their performance monitoring data with the received public key and periodically push the encrypted performance monitoring data to the corresponding performance monitoring node. After asymmetrically decrypting the received performance monitoring data based on its public key, the performance monitoring node pushes the performance monitoring data to the upper-level performance monitoring data;

[0049] S400. For the monitoring sub-cluster, regularly determine whether the performance monitoring node and the performance monitoring data are normal. If normal, synchronize and summarize the performance monitoring data of the monitoring sub-cluster where the performance monitoring node is located to the upper-level performance monitoring node through the performance monitoring node, and synchronize and summarize all the performance monitoring data through the performance monitoring node at the highest level, and disk the summarized performance monitoring data;

[0050] S500. For each monitoring sub-cluster, regularly determine whether the number of nodes within the monitoring sub-cluster meets the threshold. If it exceeds the threshold, according to the time data when the performance monitoring node joins the monitoring sub-cluster, add the performance monitoring node that joined the monitoring sub-cluster later to other monitoring sub-clusters or create a new monitoring sub-cluster, and prompt the user;

[0051] S600. Customize the UI interface based on the configuration information and display the performance monitoring data through the customized UI interface.

[0052] Among them, in step S100, for a new performance monitoring node, its configuration information and openssl certificate are recorded in the system table, including the following steps: For a new performance monitoring node, it is judged whether its configuration information and openssl certificate exist in the system table. If not, it is determined that the new performance monitoring node has not been started, and its configuration information and openssl certificate are recorded in the system table.

[0053] Step S300 adds the performance monitoring node to the performance monitoring cluster and adds it to a monitoring sub-cluster. If the addition fails, there are the following scenarios:

[0054] Scenario 1: It is prompted that the addition to the performance monitoring cluster fails, it is determined that the new performance monitoring node is offline, and it is recorded in the system table;

[0055] Scenario 2: The new performance monitoring node does not receive the performance monitoring data pushed by other nodes in the monitoring sub-cluster within a predetermined time period. It is determined that other nodes in the monitoring sub-cluster are offline, and it is recorded in the system table.

[0056] When step S400 periodically determines whether the performance monitoring data is normal, it is judged whether the indicators of the performance monitoring data exceed the threshold specified in the configuration information. If so, it is prompted that the performance monitoring data is abnormal and it is recorded in the system table.

[0057] The method of this embodiment assumes that a performance data monitoring cluster is deployed, and the following steps are executed:

[0058] The first step is to judge whether the configuration information and openssl certificate of the performance monitoring node exist in the system table. If not, it is prompted that the startup of the performance monitoring node fails, it is prompted that the configuration information is missing, and the configuration information and openssl certificate of the performance monitoring node are recorded in the system table;

[0059] The second step is to load the encrypted configuration information and openssl certificate into the running memory, where the configuration information includes cluster information, monitoring items, display methods, customized monitoring items, etc. The encryption of the configuration information is based on the openssl asymmetric encryption algorithm.

[0060] The third step is to add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. If the addition fails, it is divided into two scenarios: Scenario 1: It is prompted that the addition to the performance monitoring cluster fails, and the added performance monitoring node is suspected of being offline. And it is recorded in the system table; Scenario 2: If the monitoring data from the other party is not received for a long time, it is prompted that other nodes in the monitoring sub-cluster are suspected of being offline, and it is recorded in the system table.

[0061] Step 4: The performance monitoring nodes that join the performance monitoring cluster push their public keys through the gossip protocol. First, the performance monitoring nodes within the monitoring sub-cluster use gossip to broadcast their public key certificates. The other nodes within the monitoring sub-cluster periodically push their encrypted performance monitoring data to the performance monitoring nodes within the monitoring sub-cluster. If other nodes are newly joined nodes, they can obtain the public keys of the performance monitoring nodes that join the monitoring sub-cluster by actively pushing. After the performance monitoring nodes verify the identities of other nodes, they will push their public keys through the gossip protocol. All performance monitoring data is encrypted with the public key certificates of the performance monitoring nodes within the performance cluster. The performance monitoring nodes within the monitoring sub-cluster decrypt the performance monitoring data within the monitoring sub-cluster with their own private key certificates and synchronize and summarize the performance data to the higher-level performance monitoring nodes.

[0062] Step 5: The highest-level performance monitoring node synchronizes and summarizes all the performance monitoring data, and then stores the relevant performance monitoring data on physical storage media such as disks.

[0063] Step 6: Periodically determine whether the number of nodes within the performance monitoring sub-cluster is less than or equal to three. According to the order of joining time, adjust the scale of the performance monitoring sub-cluster, let the newly joined performance monitoring nodes join other sub-clusters or create a new performance monitoring sub-cluster, and prompt the user.

[0064] Step 7: Customize the UI to display the performance monitoring data.

[0065] Embodiment 2:

[0066] A performance data interaction system under a distributed database according to the present invention includes a configuration module, a loading module, a cluster update module, a performance data collection module, a sub-cluster detection module, and a data display module. This system executes the method disclosed in Embodiment 1 and realizes the performance monitoring data under the KaiWuDB database based on asymmetric encryption and the gossip protocol.

[0067] The configuration module is used to perform the following: For new performance monitoring nodes, record their configuration information and openssl certificates in the system table.

[0068] In this embodiment, for new performance monitoring nodes, the configuration module is used to record their configuration information and openssl certificates in the system table, including the following steps: For new performance monitoring nodes, determine whether their configuration information and openssl certificates exist in the system table. If not, determine that the new performance monitoring node has not been started, and record their configuration information and openssl certificates in the system table.

[0069] The loading module is used to perform the following: for a new performance monitoring node, after encrypting the configuration information based on the openssl asymmetric encryption algorithm, the encrypted configuration information and the openssl certificate are loaded into the memory.

[0070] The cluster update module is used to perform the following: for a new performance monitoring node, add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. The performance monitoring cluster includes multiple levels of monitoring sub-clusters, and each monitoring sub-cluster includes performance monitoring nodes and other nodes. For each monitoring sub-cluster, the performance monitoring node that successfully joins the monitoring sub-cluster pushes its public key to other nodes in the monitoring sub-cluster through the gossip protocol. Other nodes asymmetrically encrypt their performance monitoring data with the received public key and periodically push the encrypted performance monitoring data to the corresponding performance monitoring node. After the performance monitoring node asymmetrically decrypts the received performance monitoring data based on its public key, it pushes the performance monitoring data to the upper-level performance monitoring data.

[0071] In this embodiment, when adding a performance monitoring node to the performance monitoring cluster and adding it to a monitoring sub-cluster, if the addition fails, the cluster update module is used to perform the following:

[0072] Scenario 1: Prompt that the addition to the performance monitoring cluster fails, determine that the new performance monitoring node is offline, and record it in the system table;

[0073] Scenario 2: If the new performance monitoring node does not receive the performance monitoring data pushed by other nodes in the monitoring sub-cluster within a predetermined time period, determine that other nodes in the monitoring sub-cluster are offline, and record it in the system table.

[0074] When step S400 periodically determines whether the performance monitoring data is normal, it determines whether the metrics of the performance monitoring data exceed the threshold specified in the configuration information. If it exceeds, it prompts that the performance monitoring data is abnormal and records it in the system table.

[0075] The performance data collection module is used to perform the following: for a monitoring sub-cluster, periodically determine whether the performance monitoring node and the performance monitoring data are normal. If normal, synchronize and summarize the performance monitoring data of the monitoring sub-cluster where the performance monitoring node is located to the upper-level performance monitoring node through the performance monitoring node, and synchronize and summarize all performance monitoring data through the performance monitoring node at the highest level, and write the summarized performance monitoring data to disk.

[0076] The sub-cluster detection module is used to perform the following: for each monitoring sub-cluster, periodically determine whether the number of nodes in the monitoring sub-cluster meets the threshold. If it exceeds the threshold, according to the time data when the performance monitoring node joins the monitoring sub-cluster, add the performance monitoring node that joined the monitoring sub-cluster later to other monitoring sub-clusters or create a new monitoring sub-cluster, and prompt the user.

[0077] The data display module is used to perform the following: customize the UI interface based on the configuration information, and display the performance monitoring data through the customized UI interface.

[0078] The present invention has been described in detail above with reference to the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above-mentioned multiple embodiments, those skilled in the art can know that more embodiments of the present invention can be obtained by combining the means in the above different embodiments, and these embodiments are also within the protection scope of the present invention.

Claims

1. A method for performance data interaction under a distributed database, characterized in that, Implement performance monitoring data under KaiWuDB database based on asymmetric encryption and gossip protocol. For the performance monitoring cluster based on KaiWuDB database, execute the following steps: For a new performance monitoring node, record its configuration information and openssl certificate in the system table; For a new performance monitoring node, after encrypting the configuration information based on the openssl asymmetric encryption algorithm, load the encrypted configuration information and openssl certificate into memory; For a new performance monitoring node, add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. The performance monitoring cluster includes multiple levels of monitoring sub-clusters, and each monitoring sub-cluster includes performance monitoring nodes and other nodes. For each monitoring sub-cluster, the performance monitoring node that successfully joins the monitoring sub-cluster pushes its public key to other nodes in the monitoring sub-cluster through the gossip protocol. Other nodes asymmetrically encrypt their performance monitoring data with the received public key and periodically push the encrypted performance monitoring data to the corresponding performance monitoring node. After asymmetrically decrypting the received performance monitoring data based on its public key, the performance monitoring node pushes the performance monitoring data to the upper-level performance monitoring data; For the monitoring sub-cluster, regularly determine whether the performance monitoring node and the performance monitoring data are normal. If normal, synchronize and summarize the performance monitoring data of the monitoring sub-cluster where the performance monitoring node is located to the upper-level performance monitoring node through the performance monitoring node, and synchronize and summarize all performance monitoring data through the performance monitoring node at the highest level, and write the summarized performance monitoring data to disk; For each monitoring sub-cluster, regularly determine whether the number of nodes in the monitoring sub-cluster meets the threshold. If it exceeds the threshold, according to the time data when the performance monitoring node joins the monitoring sub-cluster, add the performance monitoring node that joined the monitoring sub-cluster later to other monitoring sub-clusters or create a new monitoring sub-cluster, and prompt the user; Customize the UI interface based on the configuration information and display the performance monitoring data through the customized UI interface.

2. The performance data interaction method under a distributed database according to claim 1, characterized in that For a new performance monitoring node, record its configuration information and openssl certificate in the system table, including the following steps: For a new performance monitoring node, determine whether its configuration information and openssl certificate exist in the system table. If not, determine that the new performance monitoring node has not been started, and record its configuration information and openssl certificate in the system table.

3. The performance data interaction method under a distributed database according to claim 1, characterized in that, Add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. If the addition fails, there are the following scenarios: Prompt that the addition to the performance monitoring cluster fails, determine that the new performance monitoring node is offline, and record it in the system table; If the new performance monitoring node does not receive the performance monitoring data pushed by other nodes in the monitoring sub-cluster within a predetermined time period, determine that other nodes in the monitoring sub-cluster are offline, and record it in the system table.

4. The performance data interaction method under a distributed database according to claim 1, wherein When regularly determining whether the performance monitoring data is normal, determine whether the metrics of the performance monitoring data exceed the threshold specified in the configuration information. If so, prompt that the performance monitoring data is abnormal and record it in the system table.

5. A performance data interaction system under a distributed database, characterized in that, A method for performance data interaction under a distributed database as described in any one of claims 1-4, implementing performance monitoring data under the KaiWuDB database based on asymmetric encryption and the gossip protocol. The system includes: A configuration module, which is used to perform the following: for a new performance monitoring node, record its configuration information and the openssl certificate in the system table; A loading module, which is used to perform the following: for a new performance monitoring node, after encrypting the configuration information based on the openssl asymmetric encryption algorithm, load the encrypted configuration information and the openssl certificate into memory; A cluster update module, which is used to perform the following: for a new performance monitoring node, add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. The performance monitoring cluster includes multiple levels of monitoring sub-clusters, and each monitoring sub-cluster includes performance monitoring nodes and other nodes. For each monitoring sub-cluster, the performance monitoring node that successfully joins the monitoring sub-cluster pushes its public key to other nodes in the monitoring sub-cluster through the gossip protocol. Other nodes asymmetrically encrypt their performance monitoring data with the received public key and periodically push the encrypted performance monitoring data to the corresponding performance monitoring node. After the performance monitoring node asymmetrically decrypts the received performance monitoring data based on its public key, it pushes the performance monitoring data to the upper-level performance monitoring data; A performance data collection module, which is used to perform the following: for the monitoring sub-cluster, regularly determine whether the performance monitoring node and the performance monitoring data are normal. If normal, synchronize and summarize the performance monitoring data of the monitoring sub-cluster where the performance monitoring node is located to the upper-level performance monitoring node through the performance monitoring node, and synchronize and summarize all performance monitoring data through the performance monitoring node at the highest level, and write the summarized performance monitoring data to disk; A sub-cluster detection module, which is used to perform the following: for each monitoring sub-cluster, regularly determine whether the number of nodes in the monitoring sub-cluster meets the threshold. If it exceeds the threshold, according to the time data when the performance monitoring node joins the monitoring sub-cluster, add the performance monitoring node that joined the monitoring sub-cluster later to other monitoring sub-clusters or create a new monitoring sub-cluster, and prompt the user; A data display module, which is used to perform the following: customize the UI interface based on the configuration information and display the performance monitoring data through the customized UI interface.

6. The performance data interaction system under a distributed database according to claim 5, wherein, For a new performance monitoring node, the configuration module is used to perform the following to record its configuration information and the openssl certificate in the system table: For a new performance monitoring node, determine whether its configuration information and the openssl certificate exist in the system table. If not, determine that the new performance monitoring node has not been started, and record its configuration information and the openssl certificate in the system table.

7. The performance data interaction system under a distributed database according to claim 5, wherein Add the performance monitoring node to the performance monitoring cluster and add it to a monitoring sub-cluster. If the addition fails, the cluster update module is used to perform the following: Failed to prompt the addition of the performance monitoring cluster. Determine that the new performance monitoring node is offline and record it in the system table; If the new performance monitoring node does not receive the performance monitoring data pushed by other nodes in the monitoring sub-cluster within the predetermined time period, determine that other nodes in the monitoring sub-cluster are offline and record it in the system table.

8. The performance data interaction system under the distributed database according to claim 5, characterized in that When periodically determining whether the performance monitoring data is normal, the performance data acquisition module is used to perform the following: determine whether the indicators of the performance monitoring data exceed the threshold specified in the configuration information. If so, prompt that the performance monitoring data is abnormal and record it in the system table.

Images