A method, system, device and medium for cluster node memory load balancing

Through regular detection and flag settings, dynamically adjusting the client connection of the cluster nodes, the problem of static load balancing being unable to adapt to dynamic load changes is solved, and the load balancing and stability improvement of the cluster nodes is achieved.

CN113806084BActive Publication Date: 2024-11-05JINAN INSPUR DATA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111045024.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-09-07
Publication Date
2024-11-05
Estimated Expiration
2041-09-07

AI Technical Summary

Technical Problem

The static load balancing strategy in the prior art cannot effectively and dynamically adjust the load of cluster nodes, resulting in the inability to maintain load balancing in real time when network and business factors change, affecting system stability.

Method used

Dynamic load balancing is achieved by regularly detecting the pressure values ​​of cluster nodes, setting healthy and non-health flags, resetting the client connection of the overloaded node and transferring it to a lighter load node.

Benefits of technology

Real-time adjustment of node load improves the stability and resource utilization of the cluster and ensures dynamic balance of the load.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113806084B_ABST
    Figure CN113806084B_ABST
Patent Text Reader

Abstract

The present invention provides a method, system, device and storage medium for cluster node memory load balancing, the method comprising: detecting the pressure value of each node at preset intervals, and setting an unhealthy flag for a node whose pressure value exceeds a first preset condition, and setting a healthy flag for a node whose pressure value is less than a second preset condition; detecting the flag of each node at preset intervals, and judging whether there is a node with a healthy flag in response to detecting that there is a node with an unhealthy flag in the current cluster; in response to the existence of a node with a healthy flag, resetting the transmission control protocol connection of the latest client of the node with the unhealthy flag; and returning the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the healthy flag to achieve reconnection of the latest client. The present invention can ensure that the pressure values ​​of each node tend to be balanced in real time, and improve the stability of the cluster.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of load balancing, and more specifically, to a method, system, device and storage medium for cluster node memory load balancing. Background Art

[0002] AS13000 is a four-in-one data service cabinet software-defined storage system that integrates blocks, files, objects, and big data. Currently, the AS13000 product uses a static load balancing solution, that is, when the client uses a domain name to mount the AS13000 storage system server through cifs (Common Internet File Systems) or nfs (Network File System), the server returns a virtual ip (Internet Protocol) to the client after calculating through a certain strategy, and the client uses this virtual ip to access the storage. In this way, the business pressure on a certain node is avoided, and a node with the least load is selected when the client connects.

[0003] Currently, there are five main server-side calculation strategies:

[0004] 1. Polling-based strategy: The server cluster will have multiple virtual IPs. When multiple clients access the server, the server will return the virtual IPs to the clients in turn.

[0005] 2. Strategy based on the number of connections: When a client accesses the server, the server returns the virtual IP on the node with the least number of client connections to the client.

[0006] 3. Bandwidth-based strategy: When a client accesses a server, the server returns the virtual IP address on the node with the least bandwidth pressure to the client.

[0007] 4. CPU-based strategy: When a client accesses a server, the server returns the virtual IP address on the node with the least CPU pressure to the client.

[0008] 5. Memory-based strategy: When a client accesses a server, the server returns the virtual IP address on the node with the least memory pressure to the client.

[0009] The above static load balancing strategy only determines the node to which the client connects based on the number of connections, CPU, memory, bandwidth and other information at the time of the client's first connection to achieve load balancing at the current moment. The obvious drawback of the static load balancing strategy is that it only balances the load at a certain moment, but the CPU, memory, bandwidth and other indicators of the cluster are affected by network, business and other factors and change dynamically. Summary of the invention

[0010] In view of this, the purpose of an embodiment of the present invention is to propose a method, system, computer device and computer-readable storage medium for cluster node memory load balancing. The present invention switches some clients on nodes with large pressure values ​​to nodes with small pressure values ​​according to the pressure conditions of various indicators of cluster nodes, thereby ensuring in real time that the pressure values ​​of various nodes tend to be balanced, thereby improving the stability of the cluster and making full use of cluster resources.

[0011] Based on the above-mentioned purpose, one aspect of an embodiment of the present invention provides a method for cluster node memory load balancing, comprising the following steps: detecting the pressure value of each node at preset time intervals, and setting an unhealthy flag for nodes whose pressure values ​​exceed a first preset condition, and setting a healthy flag for nodes whose pressure values ​​are less than a second preset condition; detecting the flag of each node at the preset time intervals, and in response to detecting the presence of a node with an unhealthy flag in the current cluster, determining whether there is a node with a healthy flag; in response to the presence of a node with a healthy flag, resetting the transmission control protocol connection of the latest client of the node with the unhealthy flag; and returning the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the healthy flag to achieve reconnection of the latest client.

[0012] In some embodiments, setting an unhealthy flag for a node whose stress value exceeds a first preset condition includes: in response to the stress value of the node exceeding a first threshold, increasing the unhealthy value of the node by one; in response to the stress value of the node being less than or equal to the first threshold, clearing the unhealthy value of the node to zero; and determining whether there is a node whose unhealthy value reaches a second threshold; and setting an unhealthy flag for the node in response to the existence of a node whose unhealthy value reaches the second threshold.

[0013] In some embodiments, setting a health flag for a node whose stress value is less than a second preset condition includes: in response to the stress value of the node being less than or equal to a third threshold, adding one to the health value of the node; in response to the stress value of the node exceeding the third threshold, clearing the health value of the node to zero; and determining whether there is a node whose health value reaches a fourth threshold, and setting a health flag for the node in response to the presence of a node whose health value reaches the fourth threshold.

[0014] In some implementations, resetting the transmission control protocol connection of the latest client of the node with the unhealthy mark includes: obtaining connection timestamps of all clients of the node with the unhealthy mark, and determining the latest client according to the connection timestamps.

[0015] Another aspect of an embodiment of the present invention provides a system for cluster node memory load balancing, including: a setting module, configured to detect the pressure value of each node at preset time intervals, and set an unhealthy flag for nodes whose pressure values ​​exceed a first preset condition, and set a healthy flag for nodes whose pressure values ​​are less than a second preset condition; a detection module, configured to detect the flag of each node at the preset time interval, and in response to detecting the presence of a node with an unhealthy flag in the current cluster, determine whether there is a node with a healthy flag; a reset module, configured to reset the transmission control protocol connection of the latest client of the node with the unhealthy flag in response to the presence of a node with a healthy flag; and a reconnection module, configured to return the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the healthy flag to achieve reconnection of the latest client.

[0016] In some embodiments, the setting module is configured to: in response to the stress value of the node exceeding a first threshold, increase the unhealthy value of the node by one; in response to the stress value of the node being less than or equal to the first threshold, clear the unhealthy value of the node to zero; and determine whether there is a node whose unhealthy value reaches a second threshold, and in response to the existence of a node whose unhealthy value reaches the second threshold, set an unhealthy flag for the node.

[0017] In some embodiments, the setting module is configured to: in response to the stress value of the node being less than or equal to a third threshold, increase the health value of the node by one; in response to the stress value of the node exceeding the third threshold, clear the health value of the node to zero; and determine whether there is a node whose health value reaches a fourth threshold; in response to the presence of a node whose health value reaches the fourth threshold, set a health flag for the node.

[0018] In some implementations, the reset module is configured to: obtain connection timestamps of all clients of the node with the unhealthy mark, and determine the latest client according to the connection timestamps.

[0019] According to another aspect of an embodiment of the present invention, a computer device is provided, comprising: at least one processor; and a memory, wherein the memory stores computer instructions executable on the processor, and the instructions implement the steps of the above method when executed by the processor.

[0020] According to another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, which stores a computer program that implements the above method steps when executed by a processor.

[0021] The present invention has the following beneficial technical effects: according to the pressure conditions of various indicators of cluster nodes, some clients on nodes with large pressure values ​​are switched to nodes with small pressure values, which can ensure in real time that the pressure values ​​of various nodes tend to be balanced, thereby improving the stability of the cluster and making full use of cluster resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying creative work.

[0023] Figure 1 A schematic diagram of an embodiment of a cluster node memory load balancing method provided by the present invention;

[0024] Figure 2 A schematic diagram of an embodiment of a system for cluster node memory load balancing provided by the present invention;

[0025] Figure 3 A schematic diagram of the hardware structure of an embodiment of a computer device for cluster node memory load balancing provided by the present invention;

[0026] Figure 4 A schematic diagram of an embodiment of a computer storage medium for cluster node memory load balancing provided by the present invention. DETAILED DESCRIPTION

[0027] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0028] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two non-identical entities with the same name or non-identical parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be understood as limitations on the embodiments of the present invention. The subsequent embodiments will not explain this one by one.

[0029] A first aspect of an embodiment of the present invention provides an embodiment of a method for cluster node memory load balancing. Figure 1 FIG. 1 is a schematic diagram of an embodiment of a method for cluster node memory load balancing provided by the present invention. Figure 1 As shown, the embodiment of the present invention includes the following steps:

[0030] S1. Detect the pressure value of each node at a preset time interval, and set an unhealthy flag for a node whose pressure value exceeds a first preset condition, and set a healthy flag for a node whose pressure value is less than a second preset condition;

[0031] S2, detecting the flag of each node at the preset time interval, and in response to detecting that there is a node with an unhealthy flag in the current cluster, determining whether there is a node with a healthy flag;

[0032] S3. In response to the existence of a node with a healthy mark, resetting the transmission control protocol connection of the latest client of the node with an unhealthy mark; and

[0033] S4. Return the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the health mark to achieve reconnection of the latest client.

[0034] According to the pressure conditions of various indicators of cluster nodes, the embodiment of the present invention switches some clients on nodes with large pressure values ​​to nodes with small pressure values, thereby achieving dynamic load balancing.

[0035] The pressure value of each node is detected at a preset time interval, and an unhealthy flag is set for a node whose pressure value exceeds a first preset condition, and a healthy flag is set for a node whose pressure value is less than a second preset condition.

[0036] In some embodiments, the setting of an unhealthy flag for a node whose stress value exceeds a first preset condition includes: in response to the stress value of the node exceeding a first threshold, adding one to the unhealthy value of the node, and in response to the stress value of the node being less than or equal to the first threshold, clearing the unhealthy value of the node; and determining whether there is a node whose unhealthy value reaches a second threshold, and in response to the existence of a node whose unhealthy value reaches the second threshold, setting an unhealthy flag for the node. For example, a process detects a stress value once every 10 seconds, and when the stress value is greater than 80%, the unhealthy value is increased by 1, and if it is less than 80%, the unhealthy value is cleared. When the unhealthy value reaches 10 times, it means that the stress value is greater than 80% for 10 consecutive times after being detected once every 10 seconds, and the node is set to an unhealthy flag.

[0037] In some embodiments, setting a health flag for a node whose stress value is less than a second preset condition includes: in response to the stress value of the node being less than or equal to a third threshold, adding one to the health value of the node, and in response to the stress value of the node exceeding the third threshold, clearing the health value of the node; and determining whether there is a node whose health value reaches a fourth threshold, and in response to the existence of a node whose health value reaches the fourth threshold, setting a health flag for the node. For example, a process detects the stress value once every 10 seconds, and when the stress value is less than 50%, the healthier value is increased by 1, and if it is greater than 50%, the healthier value is cleared. When the healthier value reaches 10 times, it means that the stress value is less than 50% for 10 consecutive times after being detected once every 10 seconds, and the node is set to the healthier state.

[0038] The flag of each node is detected at the preset time interval, and in response to detecting the presence of a node with an unhealthy flag in the current cluster, it is determined whether there is a node with a healthy flag. In response to the presence of a node with a healthy flag, the transmission control protocol connection of the latest client of the node with the unhealthy flag is reset. The process detects the flag of the node every 10 seconds, and the client switching condition is met if and only if a node is marked as unhealthy and the other node is marked as healthier. When a node is marked as unhealthy but there is no node with a healthier flag, the client switching cannot be completed.

[0039] In some embodiments, resetting the transmission control protocol connection of the latest client of the node with the unhealthy mark includes: obtaining the connection timestamps of all clients of the node with the unhealthy mark, determining the latest client according to the connection timestamps, determining the time when the client connected to the node according to the connection timestamps, and determining the client that last connected to the node as the latest client.

[0040] The reset transmission control protocol connection of the latest client is returned to the virtual IP address of the node with the health mark to achieve the reconnection of the latest client. The TCP (Transmission Control Protocol) connection of the latest client on the unhealthy node is reset, and then returned to the virtual IP of the client's healthier node, and the client achieves reconnection.

[0041] The embodiment of the present invention switches some clients on nodes with high pressure values ​​to nodes with low pressure values ​​according to the pressure conditions of various indicators of cluster nodes, which can ensure in real time that the pressure values ​​of various nodes tend to be balanced, improve the stability of the cluster, and make full use of cluster resources.

[0042] It should be pointed out in particular that the various steps in the various embodiments of the above-mentioned cluster node memory load balancing method can be cross-linked, replaced, added, and deleted. Therefore, these reasonable permutations, combinations and transformations of the cluster node memory load balancing method should also fall within the scope of protection of the present invention, and the scope of protection of the present invention should not be limited to the embodiments.

[0043] Based on the above purpose, the second aspect of the embodiment of the present invention provides a system for cluster node memory load balancing. Figure 2 As shown, the system 200 includes the following modules: a setting module, configured to detect the stress value of each node at preset time intervals, and set an unhealthy flag for a node whose stress value exceeds a first preset condition, and set a healthy flag for a node whose stress value is less than a second preset condition; a detection module, configured to detect the flag of each node at the preset time interval, and in response to detecting the presence of a node with an unhealthy flag in the current cluster, determine whether there is a node with a healthy flag; a reset module, configured to reset the transmission control protocol connection of the latest client of the node with the unhealthy flag in response to the presence of a node with a healthy flag; and a reconnection module, configured to return the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the healthy flag to achieve reconnection of the latest client.

[0044] In some embodiments, the setting module is configured to: in response to the stress value of the node exceeding a first threshold, increase the unhealthy value of the node by one; in response to the stress value of the node being less than or equal to the first threshold, clear the unhealthy value of the node to zero; and determine whether there is a node whose unhealthy value reaches a second threshold, and in response to the existence of a node whose unhealthy value reaches the second threshold, set an unhealthy flag for the node.

[0045] In some embodiments, the setting module is configured to: in response to the stress value of the node being less than or equal to a third threshold, increase the health value of the node by one; in response to the stress value of the node exceeding the third threshold, clear the health value of the node to zero; and determine whether there is a node whose health value reaches a fourth threshold; in response to the presence of a node whose health value reaches the fourth threshold, set a health flag for the node.

[0046] In some implementations, the reset module is configured to: obtain connection timestamps of all clients of the node with the unhealthy mark, and determine the latest client according to the connection timestamps.

[0047] Based on the above purpose, the third aspect of an embodiment of the present invention proposes a computer device, comprising: at least one processor; and a memory, the memory storing computer instructions that can be run on the processor, the instructions being executed by the processor to implement the following steps: S1, detecting the pressure value of each node at preset time intervals, and setting an unhealthy flag for nodes whose pressure values ​​exceed a first preset condition, and setting a healthy flag for nodes whose pressure values ​​are less than a second preset condition; S2, detecting the flag of each node at the preset time intervals, and in response to detecting the presence of a node with an unhealthy flag in the current cluster, determining whether there is a node with a healthy flag; S3, in response to the presence of a node with a healthy flag, resetting the transmission control protocol connection of the latest client of the node with the unhealthy flag; and S4, returning the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the healthy flag to achieve reconnection of the latest client.

[0048] In some embodiments, setting an unhealthy flag for a node whose stress value exceeds a first preset condition includes: in response to the stress value of the node exceeding a first threshold, increasing the unhealthy value of the node by one; in response to the stress value of the node being less than or equal to the first threshold, clearing the unhealthy value of the node to zero; and determining whether there is a node whose unhealthy value reaches a second threshold; and setting an unhealthy flag for the node in response to the existence of a node whose unhealthy value reaches the second threshold.

[0049] In some embodiments, setting a health flag for a node whose stress value is less than a second preset condition includes: in response to the stress value of the node being less than or equal to a third threshold, adding one to the health value of the node; in response to the stress value of the node exceeding the third threshold, clearing the health value of the node to zero; and determining whether there is a node whose health value reaches a fourth threshold, and setting a health flag for the node in response to the presence of a node whose health value reaches the fourth threshold.

[0050] In some implementations, resetting the transmission control protocol connection of the latest client of the node with the unhealthy mark includes: obtaining connection timestamps of all clients of the node with the unhealthy mark, and determining the latest client according to the connection timestamps.

[0051] like Figure 3 FIG. 1 is a schematic diagram of the hardware structure of an embodiment of the computer device for cluster node memory load balancing provided by the present invention.

[0052] As Figure 3 Taking the device shown as an example, the device includes a processor 301 and a memory 302.

[0053] The processor 301 and the memory 302 may be connected via a bus or other means. Figure 3 The example of connecting through bus is taken in the following.

[0054] The memory 302 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the cluster node memory load balancing method in the embodiment of the present application. The processor 301 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 302, that is, implementing the cluster node memory load balancing method.

[0055] The memory 302 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the cluster node memory load balancing method, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 302 may optionally include a memory remotely arranged relative to the processor 301, and these remote memories may be connected to the local module via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0056] Computer instructions 303 corresponding to one or more cluster node memory load balancing methods are stored in the memory 302 , and when executed by the processor 301 , the cluster node memory load balancing method in any of the above method embodiments is executed.

[0057] Any embodiment of a computer device that executes the above-mentioned cluster node memory load balancing method can achieve the same or similar effect as any corresponding embodiment of the above-mentioned method.

[0058] The present invention also provides a computer-readable storage medium, which stores a computer program for executing a method for cluster node memory load balancing when executed by a processor.

[0059] The method for cluster node memory load balancing includes the following steps: detecting the pressure value of each node at preset time intervals, and setting an unhealthy flag for nodes whose pressure values ​​exceed a first preset condition, and setting a healthy flag for nodes whose pressure values ​​are less than a second preset condition; detecting the flag of each node at the preset time intervals, and in response to detecting the presence of a node with an unhealthy flag in the current cluster, determining whether there is a node with a healthy flag; in response to the presence of a node with a healthy flag, resetting the transmission control protocol connection of the latest client of the node with the unhealthy flag; and returning the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the healthy flag to achieve reconnection of the latest client.

[0060] In some embodiments, setting an unhealthy flag for a node whose stress value exceeds a first preset condition includes: in response to the stress value of the node exceeding a first threshold, increasing the unhealthy value of the node by one; in response to the stress value of the node being less than or equal to the first threshold, clearing the unhealthy value of the node to zero; and determining whether there is a node whose unhealthy value reaches a second threshold; and setting an unhealthy flag for the node in response to the existence of a node whose unhealthy value reaches the second threshold.

[0061] In some embodiments, setting a health flag for a node whose stress value is less than a second preset condition includes: in response to the stress value of the node being less than or equal to a third threshold, adding one to the health value of the node; in response to the stress value of the node exceeding the third threshold, clearing the health value of the node to zero; and determining whether there is a node whose health value reaches a fourth threshold, and setting a health flag for the node in response to the presence of a node whose health value reaches the fourth threshold.

[0062] In some implementations, resetting the transmission control protocol connection of the latest client of the node with the unhealthy mark includes: obtaining connection timestamps of all clients of the node with the unhealthy mark, and determining the latest client according to the connection timestamps.

[0063] like Figure 4 As shown in FIG. 1 , a schematic diagram of an embodiment of the computer storage medium for cluster node memory load balancing provided by the present invention is shown. Figure 4 Taking the computer storage medium shown as an example, the computer readable storage medium 401 stores a computer program 402 that performs the above method when executed by a processor.

[0064] Finally, it should be noted that a person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program of the method for cluster node memory load balancing can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium of the program can be a disk, an optical disk, a read-only storage memory (ROM) or a random access memory (RAM), etc. The above-mentioned computer program embodiments can achieve the same or similar effects as the corresponding aforementioned arbitrary method embodiments.

[0065] The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope disclosed in the embodiments of the present invention as defined in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any particular order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless explicitly limited to the singular.

[0066] It should be understood that, as used herein, the singular forms "a", "an" are intended to include the plural forms as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations including one or more of the associated listed items.

[0067] The serial numbers of the embodiments disclosed in the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0068] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0069] A person skilled in the art should understand that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples; under the concept of the embodiments of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes in different aspects of the embodiments of the present invention as above, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.

Claims

1. A method for cluster node memory load balancing, It is characterized in that The steps include: Detect the pressure value of each node at a preset time interval, and set an unhealthy flag for a node whose pressure value exceeds a first preset condition, and set a healthy flag for a node whose pressure value is less than a second preset condition; Detecting a flag of each node at a predetermined interval, and in response to detecting that there is a node with an unhealthy flag in the current cluster, determining whether there is a node with a healthy flag; In response to the existence of a node with a healthy mark, resetting a transmission control protocol connection of a latest client of the node with an unhealthy mark; as well as The reset transmission control protocol connection of the latest client is returned to the virtual IP address of the node with the health mark to achieve reconnection of the latest client.

2. The method according to claim 1, It is characterized in that The step of setting an unhealthy flag for a node whose pressure value exceeds a first preset condition comprises: In response to a stress value of a node exceeding a first threshold, increasing an unhealthy value of the node by one, and in response to a stress value of a node being less than or equal to the first threshold, clearing the unhealthy value of the node to zero; and It is determined whether there is a node whose unhealthy value reaches a second threshold value, and in response to the node whose unhealthy value reaches the second threshold value, an unhealthy flag is set for the node.

3. The method according to claim 1, It is characterized in that The step of setting a health flag for a node whose pressure value is less than a second preset condition comprises: In response to the stress value of the node being less than or equal to a third threshold, increasing the health value of the node by one, and in response to the stress value of the node exceeding the third threshold, clearing the health value of the node to zero; and It is determined whether there is a node whose health value reaches a fourth threshold value, and in response to a node whose health value reaches the fourth threshold value, a health flag is set for the node.

4. The method according to claim 1, It is characterized in that The resetting of the transmission control protocol connection of the latest client of the node with the unhealthy mark includes: The connection timestamps of all clients of the node with the unhealthy mark are obtained, and the latest client is determined according to the connection timestamps.

5. A cluster node memory load balancing system, It is characterized in that include: A setting module configured to detect the pressure value of each node at a preset time interval, and set an unhealthy flag for a node whose pressure value exceeds a first preset condition, and set a healthy flag for a node whose pressure value is less than a second preset condition; A detection module configured to detect a flag of each node at intervals of the preset time, and in response to detecting that there is a node with an unhealthy flag in the current cluster, determine whether there is a node with a healthy flag; A reset module, configured to reset the transmission control protocol connection of the latest client of the node with the unhealthy mark in response to the existence of the node with the healthy mark; as well as A reconnection module is configured to return the reset transmission control protocol connection of the latest client to the virtual IP address of the node with the health mark to achieve reconnection of the latest client.

6. The system according to claim 5, It is characterized in that The setting module is configured to: In response to a stress value of a node exceeding a first threshold, increasing an unhealthy value of the node by one; and in response to a stress value of a node being less than or equal to the first threshold, clearing the unhealthy value of the node to zero; as well as It is determined whether there is a node whose unhealthy value reaches a second threshold value, and in response to the node whose unhealthy value reaches the second threshold value, an unhealthy flag is set for the node.

7. The system according to claim 5, It is characterized in that The setting module is configured to: In response to the stress value of the node being less than or equal to a third threshold, increasing the health value of the node by one, and in response to the stress value of the node exceeding the third threshold, clearing the health value of the node to zero; as well as It is determined whether there is a node whose health value reaches a fourth threshold value, and in response to a node whose health value reaches the fourth threshold value, a health flag is set for the node.

8. The system according to claim 5, It is characterized in that The reset module is configured to: The connection timestamps of all clients of the node with the unhealthy mark are obtained, and the latest client is determined according to the connection timestamps.

9. A computer device, It is characterized in that include: at least one processor; as well as A memory storing computer instructions executable on the processor, wherein the instructions, when executed by the processor, implement the steps of the method according to any one of claims 1 to 4.

10. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.