A multi-service adjustment method based on intranet environment

Abstract

The present invention relates to the field of multi-service adjustment, and in particular to a multi-service adjustment method based on an intranet environment. The present invention obtains the operation characteristics between each service node during the execution of each type of running program, determines the dependency characterization parameters between each service node, analyzes each service node, obtains the dependency relationship diagram between the service nodes for a single type of running program and determines the dependency cluster, calculates the associated abnormal tendency characterization parameters of the dependency cluster according to the number of service nodes corresponding to the dependency cluster combined with the average response delay and average load of each service node during operation, divides the associated abnormal tendency categories of the dependency cluster, obtains the load change of each service node during the execution process, determines whether there is an abnormal service node, and adaptively controls the corresponding service nodes in the dependency cluster according to the associated abnormal tendency category of the corresponding dependency cluster, thereby alleviating the load surge and response delay of the service node and ensuring the stability of the execution of the running program.

Description

Technical Field

[0001] The present invention relates to the field of multi-service adjustment, and in particular to a multi-service adjustment method based on an intranet environment. Background Art

[0002] With the development of modern technology, in the intranet environment, the requirements for high resource utilization and high service availability are becoming more and more stringent. By dynamically allocating services and combining them with load balancing and other technical means to adjust multiple services, we can cope with changes in multiple aspects and high requirements for system performance.

[0003] Chinese Patent Publication No.: CN111614763A discloses a method, device, and storage medium for dynamically adjusting the load of an application server, including: integrating multiple application servers in a specified environment, adjusting the priority of all application servers to 255; monitoring the status of the application server to determine whether the performance index of the application server reaches a specified threshold; if so, the application server is a load server, and the fire wall is controlled to shield or enable VRRP protocol messages; if not, the application server is adjusted to a standby server, the fire wall is controlled by a script, the reception of VRRP protocol messages is enabled, and the priority of the standby server is adjusted to 0; it is determined whether the performance index of the standby server is restored to the specified threshold, if so, the standby server is enabled, the standby server is adjusted to a load server, the fire wall is controlled by a script, the reception of VRRP protocol messages is disabled, and the priority of the load server is adjusted to 255.

[0004] However, the following problems still exist in the prior art: during the program running process, there are characteristics of high response frequency and high interactivity between service nodes with strong dependencies. Especially under high load, the abnormality of a single service node will affect the remaining dependent service nodes, but due to the failure to take adjustment and control measures in time, it may cause a sudden increase in the load of the service node and a delay in response. Summary of the invention

[0005] To this end, the present invention provides a multi-service adjustment method based on an intranet environment to overcome the problem in the prior art that, during the execution of a running program, there are characteristics of high response frequency and high interactivity between service nodes with strong dependencies, especially under high load, the abnormality of a single service node will affect the remaining dependent service nodes, but due to the failure to take adjustment and control measures in time, it may cause a sudden increase in the load of the service node and a delay in response.

[0006] To achieve the above object, the present invention provides a method for adjusting multiple services based on an intranet environment, which comprises:

[0007] Obtaining the operation characteristics between the service nodes during the execution of various types of running programs, so as to determine the dependency representation parameters between the service nodes based on the operation characteristics;

[0008] Parsing each service node based on the dependency characterization parameters to obtain a dependency graph between service nodes for a single type of running program and determine a dependency cluster;

[0009] According to the category of the running program, the corresponding dependency graph is called, and the correlation abnormality tendency characterization parameter of the dependency cluster is calculated based on the number of service nodes corresponding to the dependency cluster combined with the average response delay and average load of each service node during operation, so as to classify the correlation abnormality tendency category of the dependency cluster;

[0010] Obtain the load change of each service node during the execution of the running program, determine whether there is an abnormal service node, and control each corresponding service node in the dependency cluster according to the associated abnormal tendency category of the dependency cluster corresponding to the abnormal service node, including:

[0011] Performing intra-cluster association adjustment, adjusting the network bandwidth allocation of the corresponding remaining service nodes in the dependency cluster according to the load increase rate of the abnormal service node;

[0012] Or, only the network bandwidth allocation of the abnormal service node is adjusted.

[0013] Furthermore, the dependency characterization parameters between the service nodes are determined based on the operation characteristics, including:

[0014] Determine the operational characteristics between service nodes, including the synchronous response frequency and interaction frequency between service nodes;

[0015] Calculating a synchronous response frequency ratio between the synchronous response frequency and a preset synchronous response frequency threshold;

[0016] Calculating an interaction frequency ratio between the interaction frequency and a preset interaction frequency threshold;

[0017] The sum of the synchronous response frequency ratio and the interaction frequency ratio is determined as the dependency characterization parameter.

[0018] Furthermore, the process of parsing each service node to obtain a dependency graph between service nodes for a single type of running program includes:

[0019] Use the number of each service node as the node of the dependency graph, and connect the nodes through edges;

[0020] The dependency representation parameters between service nodes are used as edge weights;

[0021] Dependency clusters are determined based on edge weights.

[0022] Further, a dependency cluster is determined, wherein the weight of each edge in the determined dependency cluster is greater than a predetermined dependency characterization parameter standard threshold.

[0023] Furthermore, the correlation abnormal tendency characterization parameter of the dependency cluster is calculated according to formula (1):

[0024]

[0025] In formula (1), R represents the parameter characterizing the abnormal tendency of the dependency cluster, F represents the average response delay, F0 represents the preset standard threshold of the average response delay, H represents the average load, H0 represents the preset standard threshold of the average load, N represents the number of service nodes corresponding to the dependency cluster, and N0 represents the preset standard threshold of the number of service nodes corresponding to the dependency cluster.

[0026] Furthermore, the abnormal tendency categories of the dependency clusters are divided into:

[0027] If the correlation abnormality tendency characterization parameter of the dependency cluster is greater than or equal to the preset correlation abnormality tendency characterization parameter standard threshold of the dependency cluster, the correlation abnormality tendency category of the dependency cluster is a high abnormality tendency category;

[0028] If the correlation abnormality tendency characterization parameter of the dependency cluster is less than a preset standard threshold value of the correlation abnormality tendency characterization parameter of the dependency cluster, the correlation abnormality tendency category of the dependency cluster is a low abnormality tendency category.

[0029] Furthermore, the corresponding service nodes in the dependency cluster are controlled according to the associated abnormal tendency category of the dependency cluster corresponding to the abnormal service node, including:

[0030] If the association abnormality tendency category of the dependency cluster is a high abnormality tendency category, then perform intra-cluster association adjustment, and adjust the network bandwidth allocation of the corresponding remaining service nodes in the dependency cluster according to the load change amount and change trend of the abnormal service node;

[0031] If the associated abnormal tendency category of the dependency cluster is a low abnormal tendency category, only the network bandwidth allocation of the abnormal service node is adjusted.

[0032] Further, determining whether there is an abnormal service node includes:

[0033] If there is a service node whose load variation is greater than or equal to the load variation standard threshold, it is determined that there is an abnormal service node.

[0034] Further, adjusting the network bandwidth allocation corresponding to the remaining service nodes in the dependency cluster includes:

[0035] The network bandwidth is increased, and the increase in network bandwidth is positively correlated with the load increase rate.

[0036] Furthermore, the categories of running programs are divided according to the functions performed by each program.

[0037] Compared with the prior art, the present invention obtains the operating characteristics between each service node during the execution of each type of running program, determines the dependency characterization parameters between each service node, analyzes each service node, obtains the dependency relationship diagram between the service nodes for a single type of running program and determines the dependency cluster, calculates the associated abnormal tendency characterization parameters of the dependency cluster according to the number of service nodes corresponding to the dependency cluster combined with the average response delay and average load of each service node during operation, divides the associated abnormal tendency category of the dependency cluster, obtains the load change of each service node during the execution process, determines whether there is an abnormal service node, and adaptively controls the corresponding service nodes in the dependency cluster according to the associated abnormal tendency category of the corresponding dependency cluster, thereby alleviating the load surge and response delay of the service node and ensuring the stability of the execution of the running program.

[0038] In particular, the present invention determines the dependency characterization parameters between each service node through the operating characteristics between each service node during the execution of the running program. During the execution of the running program, there will be communication or data interaction between each service node. At the same time, when the running program is triggered to execute, each service node that needs to work may respond to the request synchronously. The above situation can characterize the dependency between each service node. Therefore, the present application determines the dependency characterization parameters between each service node through the synchronous response frequency and the interaction frequency, and provides data support for the subsequent analysis of each service node to determine the dependency cluster. Furthermore, the associated abnormal tendency of the dependency cluster is determined according to the operating conditions of the service nodes in the dependency cluster, and the associated abnormal tendency categories of the dependency cluster are divided to adaptively control the corresponding service nodes in the dependency cluster, thereby alleviating the load surge and response delay of the service node and ensuring the stability of the execution of the running program.

[0039] In particular, the present invention calculates the associated abnormal tendency characterization parameters of the dependency cluster by combining the number of service nodes corresponding to the dependency cluster with the average response delay and the average load when the service node is running. During the actual execution of the running program, several service nodes in the dependency cluster will run simultaneously according to the request. If any service node is abnormal, it may cause the response of other service nodes with which it has a dependency relationship to be delayed. Not only that, in order to maintain the continuity of the overall operation, other service nodes may need to bear more workloads, and then, resource competition occurs, resulting in a sudden increase in the load of the service node. Therefore, the present application characterizes the associated abnormal tendency of the dependency cluster by the degree of associated abnormality of the dependency cluster, and then provides data support for the subsequent classification of the associated abnormal tendency category of the dependency cluster. Then, it is determined whether there is an abnormal service node based on the load change of each service node during the execution of the running program. The two are combined to adaptively control the corresponding service nodes in the dependency cluster, and then alleviate the sudden increase in the load and response delay of the service node, and ensure the stability of the execution of the running program.

[0040] In particular, the present invention takes into account the load increase rate of the abnormal service node, and adjusts the remaining service nodes in the dependency cluster through the load increase rate, or adjusts the abnormal service node in a targeted manner. When an abnormality occurs in a service node, the load of the service node suddenly increases, thereby affecting the overall execution of the running program. Therefore, the present application adaptively allocates the network bandwidth in advance based on the load increase rate of the abnormal service node, so as to ensure the normal response of other service nodes that have a dependent relationship with the abnormal service node and the normal interaction between the service nodes, thereby alleviating the sudden increase in load and response delay of the service node and ensuring the stability of the execution of the running program. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the steps of a method for adjusting multiple services in an intranet environment according to an embodiment of the invention;

[0042] Figure 2 A logical decision diagram for classifying the associated abnormal tendency categories of dependency clusters according to an embodiment of the invention;

[0043] Figure 3 A logic decision diagram corresponding to each service node in a control dependency cluster of an embodiment of the invention;

[0044] Figure 4 This is a logical decision diagram for determining whether there is an abnormal service node according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0046] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0047] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0048] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] See also Figures 1 to 4 As shown, Figure 1 The following is a schematic diagram of the steps of the method for adjusting multiple services in an intranet environment according to an embodiment of the present invention. Figure 2 A logical decision diagram for classifying the associated abnormal tendency categories of dependency clusters according to an embodiment of the invention, Figure 3 The logical decision diagram corresponding to each service node in the control dependency cluster of the embodiment of the invention is as follows: Figure 4 The logical determination diagram for determining whether there is an abnormal service node in the embodiment of the invention is provided. The multi-service adjustment method based on the intranet environment in the embodiment of the invention includes:

[0050] Step S1, obtaining operation characteristics between various service nodes during the execution of various types of running programs, so as to determine dependency representation parameters between various service nodes based on the operation characteristics;

[0051] Step S2, parsing each service node based on the dependency characterization parameter to obtain a dependency graph between service nodes for a single type of running program, and determining a dependency cluster;

[0052] Step S3, according to the category of the running program, calling the corresponding dependency graph, and calculating the correlation abnormality tendency characterization parameter of the dependency cluster based on the number of service nodes corresponding to the dependency cluster combined with the average response delay and average load of each service node during operation, so as to classify the correlation abnormality tendency category of the dependency cluster;

[0053] Step S4, obtaining the load change of each service node during the execution of the running program, determining whether there is an abnormal service node, and controlling each corresponding service node in the dependency cluster according to the associated abnormal tendency category of the dependency cluster corresponding to the abnormal service node, including:

[0054] Performing intra-cluster association adjustment, adjusting the network bandwidth allocation of the corresponding remaining service nodes in the dependency cluster according to the load increase rate of the abnormal service node;

[0055] Or, only the network bandwidth allocation of the abnormal service node is adjusted.

[0056] Specifically, the specific structure of the service node is not limited, and it only needs to provide services for running programs and realizing interaction between services. It can be a physical server or a virtual server, which will not be elaborated here.

[0057] Specifically, the dependency characterization parameters between service nodes are determined based on the operation characteristics, including:

[0058] Determine the operational characteristics between service nodes, including the synchronous response frequency and interaction frequency between service nodes;

[0059] Calculating a synchronous response frequency ratio between the synchronous response frequency and a preset synchronous response frequency threshold;

[0060] Calculating an interaction frequency ratio between the interaction frequency and a preset interaction frequency threshold;

[0061] The sum of the synchronous response frequency ratio and the interaction frequency ratio is determined as the dependency characterization parameter.

[0062] In actual situations, each service node may be configured with different service functions. Therefore, different types of running programs call service nodes differently during operation. For different task execution stages, different service nodes or service node combinations may need to be called to complete the corresponding tasks.

[0063] The synchronous response frequency is the frequency at which the service nodes simultaneously respond to the running programs within a predetermined time and provide computing services for the corresponding running programs.

[0064] The interaction frequency is the frequency of data interaction between service nodes within a predetermined time.

[0065] Specifically, in this example, the synchronous response frequency threshold C0 and the interaction frequency threshold D0 are obtained in advance, and the relevant data of each service node in several running program execution processes are obtained. The synchronous response frequency and interaction frequency data between the service nodes are called, and the synchronous response frequency average value ΔC0 and the interaction frequency average value ΔD0 are solved. It is set that C0=r1×ΔC0, D0=r2×ΔD0, wherein r1 represents the first precision coefficient, and r2 represents the second precision coefficient, wherein 1.05<r1<1.1, 1.15<r2<1.2.

[0066] The present invention determines the dependency characterization parameters between each service node through the operating characteristics between each service node during the execution of the running program. During the execution of the running program, there will be communication or data interaction between each service node. At the same time, when the running program is triggered to execute, each service node that needs to work may respond to the request synchronously. The above situation can characterize the dependency between each service node. Therefore, the present application determines the dependency characterization parameters between each service node through the synchronous response frequency and the interaction frequency, and provides data support for the subsequent analysis of each service node to determine the dependency cluster. Furthermore, the association abnormality tendency of the dependency cluster is determined according to the operating conditions of the service nodes in the dependency cluster, and the association abnormality tendency categories of the dependency cluster are divided to adaptively control the corresponding service nodes in the dependency cluster, thereby alleviating the load surge and response delay of the service node and ensuring the stability of the execution of the running program.

[0067] Specifically, the process of parsing each service node to obtain a dependency graph between service nodes for a single type of running program includes:

[0068] Use the number of each service node as the node of the dependency graph, and connect the nodes through edges;

[0069] The dependency representation parameters between service nodes are used as edge weights;

[0070] Dependency clusters are determined based on edge weights.

[0071] Specifically, a dependency cluster is determined, wherein the weight of each edge in the determined dependency cluster is greater than a predetermined dependency characterization parameter standard threshold.

[0072] The standard threshold value M0 of the dependency characterization parameter is selected in the interval [2.1, 2.2].

[0073] Specifically, the correlation abnormality tendency characterization parameter of the dependency cluster is calculated according to formula (1):

[0074]

[0075] In formula (1), R represents the parameter characterizing the abnormal tendency of the dependency cluster, F represents the average response delay, F0 represents the preset standard threshold of the average response delay, H represents the average load, H0 represents the preset standard threshold of the average load, N represents the number of service nodes corresponding to the dependency cluster, and N0 represents the preset standard threshold of the number of service nodes corresponding to the dependency cluster.

[0076] Specifically, in this example, the standard threshold value F0 for average response delay, the standard threshold value H0 for average load, and the standard threshold value N0 for the number of service nodes corresponding to the dependency cluster are obtained in advance, and the relevant data of each service node in several execution processes of the program are obtained. The data of the average response delay, load, and the number of service nodes corresponding to the dependency cluster when each service node is running are called to solve the average response delay Fe0, the average load He0, and the number of service nodes corresponding to the dependency cluster Ne0. F0=Fe0, H0=He0, N0=Ne0 are set, and the standard threshold values ​​are determined by the above solutions.

[0077] The present invention calculates the associated abnormal tendency characterization parameters of the dependency cluster by combining the number of service nodes corresponding to the dependency cluster with the average response delay and the average load when the service node is running. During the actual execution of the running program, several service nodes in the dependency cluster will run simultaneously according to the request. If any service node is abnormal, it may cause the response of other service nodes with which it has a dependency relationship to be delayed. Not only that, in order to maintain the continuity of the overall operation, other service nodes may need to bear more workloads, and then, resource competition occurs, resulting in a sudden increase in the load of the service node. Therefore, the present application characterizes the associated abnormal tendency of the dependency cluster by the degree of associated abnormality of the dependency cluster, and then provides data support for the subsequent classification of the associated abnormal tendency category of the dependency cluster. Furthermore, it determines whether there is an abnormal service node based on the load change of each service node during the execution of the running program. The two are combined to adaptively control the corresponding service nodes in the dependency cluster, and then alleviate the sudden increase in the load and response delay of the service node, and ensure the stability of the execution of the running program.

[0078] Specifically, the abnormal tendency categories of associations for dividing the dependency clusters include:

[0079] If the correlation abnormality tendency characterization parameter of the dependency cluster is greater than or equal to the preset correlation abnormality tendency characterization parameter standard threshold of the dependency cluster, the correlation abnormality tendency category of the dependency cluster is a high abnormality tendency category;

[0080] If the correlation abnormality tendency characterization parameter of the dependency cluster is less than a preset standard threshold value of the correlation abnormality tendency characterization parameter of the dependency cluster, the correlation abnormality tendency category of the dependency cluster is a low abnormality tendency category.

[0081] Among them, the standard threshold value R0 of the parameter characterizing the tendency of associated abnormality is selected in the interval [1.35, 1.45].

[0082] Specifically, the corresponding service nodes in the dependency cluster are controlled according to the associated abnormal tendency category of the dependency cluster corresponding to the abnormal service node, including:

[0083] If the association abnormality tendency category of the dependency cluster is a high abnormality tendency category, then perform intra-cluster association adjustment, and adjust the network bandwidth allocation of the corresponding remaining service nodes in the dependency cluster according to the load change amount and change trend of the abnormal service node;

[0084] If the associated abnormal tendency category of the dependency cluster is a low abnormal tendency category, only the network bandwidth allocation of the abnormal service node is adjusted.

[0085] Specifically, determining whether there is an abnormal service node includes:

[0086] If there is a service node whose load variation is greater than or equal to the load variation standard threshold, it is determined that there is an abnormal service node.

[0087] Specifically, adjusting the network bandwidth allocation corresponding to the remaining service nodes in the dependency cluster includes:

[0088] The network bandwidth is increased, and the increase in network bandwidth is positively correlated with the load increase rate.

[0089] In this embodiment, optionally,

[0090] The load increase rate threshold S is compared with the first load increase rate threshold S1 and the second load increase rate S2.

[0091] If S>S2, the first network bandwidth increase amount k1 is determined, and k1=1.9k0 is set;

[0092] If S1≤S≤S2, the second network bandwidth increase amount k2 is determined, and k2=1.7k0 is set;

[0093] If S<S1, the third network bandwidth increase amount k3 is determined, and k3=1.5k0 is set;

[0094] Wherein, k0 represents the initial network bandwidth, S1=1.4S0, S2=1.6S0.

[0095] Specifically, in this embodiment, the load increase rate threshold S0 is obtained in advance, and the load-related data of each service node in several running program execution processes are obtained to solve the load increase rate mean ΔS0, and set S0=v1×ΔS0, where v1 represents the first deviation coefficient, and 1.02<v1<1.04.

[0096] Specifically, the initial network bandwidth is configured according to the access volume and data transmission requirements, which will not be described in detail here.

[0097] Specifically, the categories of running programs are divided according to the functions performed by each program. For example, the functions of running programs are different, and their categories are also different. Each function corresponds to a category. Those skilled in the art can divide the categories according to the specific functions of the running programs, which will not be repeated here.

[0098] The present invention takes into account the load increase rate of the abnormal service node, and adjusts the remaining service nodes in the dependency cluster through the load increase rate, or specifically adjusts the abnormal service node. When an abnormality occurs in a service node, the load of the service node suddenly increases, thereby affecting the overall execution of the running program. Therefore, the present application adaptively allocates the network bandwidth in advance according to the load increase rate of the abnormal service node, so as to ensure the normal response of other service nodes that have a dependent relationship with the abnormal service node and the normal interaction between the service nodes, thereby alleviating the sudden increase in load and response delay of the service node and ensuring the stability of the execution of the running program.

[0099] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for adjusting multiple services based on an intranet environment, characterized in that: include: Obtaining the operation characteristics between the service nodes during the execution of various types of running programs, so as to determine the dependency representation parameters between the service nodes based on the operation characteristics; Parsing each service node based on the dependency characterization parameters to obtain a dependency graph between service nodes for a single type of running program and determine a dependency cluster; According to the category of the running program, the corresponding dependency graph is called, and the correlation abnormality tendency characterization parameter of the dependency cluster is calculated based on the number of service nodes corresponding to the dependency cluster combined with the average response delay and average load of each service node during operation, so as to classify the correlation abnormality tendency category of the dependency cluster; Obtain the load change of each service node during the execution of the running program, determine whether there is an abnormal service node, and control the corresponding service nodes in the dependency cluster according to the associated abnormal tendency category of the dependency cluster corresponding to the abnormal service node. include, Performing intra-cluster association adjustment, adjusting the network bandwidth allocation of the corresponding remaining service nodes in the dependency cluster according to the load increase rate of the abnormal service node; Or, only adjust the network bandwidth allocation of the abnormal service node; Determine the dependency characterization parameters between service nodes based on operation characteristics, including: Determine the operational characteristics between service nodes, including the synchronous response frequency and interaction frequency between service nodes; Calculating a synchronous response frequency ratio between the synchronous response frequency and a preset synchronous response frequency threshold; Calculating an interaction frequency ratio between the interaction frequency and a preset interaction frequency threshold; Determining the sum of the synchronous response frequency ratio and the interaction frequency ratio as the dependency characterization parameter; According to formula (1), the correlation abnormal tendency characterization parameter of the dependency cluster is calculated. （1） In formula (1), R represents the parameter characterizing the abnormal tendency of the dependency cluster, F represents the average response delay, F0 represents the preset standard threshold of the average response delay, H represents the average load, H0 represents the preset standard threshold of the average load, N represents the number of service nodes corresponding to the dependency cluster, and N0 represents the preset standard threshold of the number of service nodes corresponding to the dependency cluster.

2. The method for adjusting multiple services based on an intranet environment according to claim 1, characterized in that: The process of parsing each service node to obtain a dependency graph between service nodes for a single type of running program includes: Use the number of each service node as the node of the dependency graph, and connect the nodes through edges; The dependency representation parameters between service nodes are used as edge weights; Dependency clusters are determined based on edge weights.

3. The method for adjusting multiple services based on an intranet environment according to claim 2, characterized in that: A dependency cluster is determined, wherein the weight of each edge in the determined dependency cluster is greater than a predetermined dependency characterization parameter standard threshold.

4. The method for adjusting multiple services based on an intranet environment according to claim 1, characterized in that: The associated abnormal tendency categories for dividing the dependency clusters include: If the correlation abnormality tendency characterization parameter of the dependency cluster is greater than or equal to the preset correlation abnormality tendency characterization parameter standard threshold of the dependency cluster, the correlation abnormality tendency category of the dependency cluster is a high abnormality tendency category; If the correlation abnormality tendency characterization parameter of the dependency cluster is less than a preset standard threshold value of the correlation abnormality tendency characterization parameter of the dependency cluster, the correlation abnormality tendency category of the dependency cluster is a low abnormality tendency category.

5. The method for adjusting multiple services based on an intranet environment according to claim 1, characterized in that: Control each service node in the dependency cluster according to the abnormal tendency category of the abnormal service node corresponding to the dependency cluster, including: If the association abnormality tendency category of the dependency cluster is a high abnormality tendency category, then perform intra-cluster association adjustment, and adjust the network bandwidth allocation of the corresponding remaining service nodes in the dependency cluster according to the load change amount and change trend of the abnormal service node; If the associated abnormal tendency category of the dependency cluster is a low abnormal tendency category, only the network bandwidth allocation of the abnormal service node is adjusted.

6. The method for adjusting multiple services based on an intranet environment according to claim 1, characterized in that: Determine whether there are abnormal service nodes. include, If there is a service node whose load variation is greater than or equal to the load variation standard threshold, it is determined that there is an abnormal service node.

7. The method for adjusting multiple services based on an intranet environment according to claim 1, characterized in that: Adjust the network bandwidth allocation corresponding to the remaining service nodes in the dependency cluster, include, The network bandwidth is increased, and the increase in network bandwidth is positively correlated with the load increase rate.

8. The method for adjusting multiple services based on an intranet environment according to claim 1, characterized in that: The categories of running programs are divided according to the functions performed by each program.

Images