A load system scheduling method and device, electronic equipment and storage medium

Abstract

The application provides a load system scheduling method and device, electronic equipment and storage medium. The method comprises: obtaining service demand data corresponding to a target application service, wherein the service demand data comprises resource demand data corresponding to a plurality of network resources; determining a plurality of first load systems capable of running the target application service; for each first load system, obtaining maximum load characteristics of the first load system, wherein the maximum load characteristics comprise maximum load data corresponding to a plurality of network resources; determining initial performance scores corresponding to the first load systems according to the resource demand data and the maximum load data corresponding to a plurality of network resources; and determining the first load system with the highest initial performance score as an initial load system running the target application service. Thus, multi-dimensional and globally optimal load scheduling of network resources is achieved.

Description

Technical Field

[0001] This application relates to the field of network resource scheduling technology, and in particular to a load system scheduling method, apparatus, electronic device and storage medium. Background Technology

[0002] With the development of Internet technology, the load problem in network systems has become increasingly prominent. How to effectively schedule and manage network resources to achieve load balancing and optimization has become an urgent technical problem to be solved.

[0003] Currently, when implementing network system scheduling, most systems only consider a single load characteristic, such as CPU utilization. If the CPU utilization of a certain load system meets the requirements of the target application service, then the target application service will be run using that load system.

[0004] However, this approach, which only considers a single load characteristic, cannot meet the multi-dimensional network resource requirements of complex application services, and there is a problem of unreasonable resource allocation. Summary of the Invention

[0005] This application provides a load system scheduling method, apparatus, electronic device, and storage medium to solve the problem that the existing technology, which only considers a single load characteristic, cannot meet the multi-dimensional network resource requirements of complex application services and has the problem of unreasonable resource allocation.

[0006] Firstly, this application provides a load system scheduling method, including:

[0007] Obtain service requirement data corresponding to the target application service, wherein the service requirement data includes resource requirement data corresponding to multiple network resources;

[0008] Identify multiple first-load systems that can currently run the target application service;

[0009] For each first load system, the maximum load characteristic of the first load system is obtained, wherein the maximum load characteristic includes the maximum load data corresponding to multiple network resources;

[0010] Based on the resource demand data and maximum load data corresponding to the multiple network resources, determine the initial performance score corresponding to the first load system;

[0011] The first load system with the highest initial performance score is determined as the initial load system for running the target application service.

[0012] In one possible implementation, determining the initial performance score corresponding to the first load system based on resource demand data and maximum load data corresponding to the plurality of network resources includes:

[0013] For each network resource, calculate the ratio of the maximum load data to the resource demand data corresponding to that network resource;

[0014] Substituting the ratio into a preset function yields the resource performance data corresponding to the network resource, wherein the preset function is a monotonically increasing function;

[0015] Based on the resource performance data corresponding to the multiple network resources, the initial performance score corresponding to the first load system is determined.

[0016] In one possible implementation, determining the initial performance score of the first load system based on resource performance data corresponding to the plurality of network resources includes:

[0017] Obtain the weight corresponding to each network resource;

[0018] Based on the weight corresponding to each network resource, the resource performance data corresponding to multiple network resources are weighted and summed to obtain the initial performance score corresponding to the first load system.

[0019] In one possible implementation, the method further includes:

[0020] During the process of running the target application service using the initial load system, the first real-time load feature of the initial load system is obtained at preset time intervals. The first real-time load feature includes first real-time load data corresponding to multiple network resources.

[0021] Obtain the pre-set first and second thresholds;

[0022] Based on the first real-time load characteristic, the first threshold, and the second threshold, determine whether it is necessary to redeploy the load system running the target application service.

[0023] In one possible implementation, determining whether the load system running the target application service needs to be redeployed based on the first real-time load characteristic, the first threshold, and the second threshold includes:

[0024] Substituting the first real-time load characteristic, the first threshold, and the second threshold into a preset formula, the first real-time performance score of the initial load system is obtained, wherein the preset formula is as follows:

[0025] Q i (t)=[r ij (t)-θ 2j ][θ 1j -r ij (t)](Formula 1);

[0026] Among them, Q i (t) represents the first real-time performance score of the initial load system at time t, r ij θ(t) represents the first real-time load data of the j-th network resource in the initial load system at time t, where θ 1j Let θ represent the first threshold for the j-th network resource. 2j This represents the second threshold for the j-th network resource;

[0027] If the first real-time performance score is greater than zero, it is determined that the initial load system is operating normally and there is no need to redeploy the load system running the target application service;

[0028] If the first real-time performance score is less than or equal to zero, it is determined that the initial load system is operating abnormally and the load system for running the target application service needs to be redeployed.

[0029] In one possible implementation, the method further includes:

[0030] If it is determined that the load system that needs to be redeployed to run the target application service is required, multiple second load systems that can currently run the target application service are identified.

[0031] For each second load system, a second real-time load characteristic of the second load system is obtained, the second real-time load characteristic including second real-time load data corresponding to multiple network resources;

[0032] Based on the resource demand data and the second real-time load data corresponding to the multiple network resources, a second real-time performance score corresponding to the second load system is determined.

[0033] The second load system with the highest real-time performance score is identified as the target load system for running the target application service.

[0034] Migrate the target application service to the target load system.

[0035] In one possible implementation, the method further includes:

[0036] The first load system whose initial performance score is greater than the preset score threshold is identified as a candidate load system;

[0037] Obtain the system priority for each candidate load system;

[0038] The candidate load system with the highest priority is determined as the initial load system for running the target application service.

[0039] Secondly, this application provides a load system scheduling device, comprising:

[0040] The data acquisition module is used to acquire service requirement data corresponding to the target application service, wherein the service requirement data includes resource requirement data corresponding to multiple network resources;

[0041] The first determining module is used to determine multiple first load systems that can currently run the target application service;

[0042] The first feature acquisition module is used to acquire the maximum load feature of each first load system, wherein the maximum load feature includes the maximum load data corresponding to multiple network resources.

[0043] The second determining module is used to determine the initial performance score corresponding to the first load system based on the resource demand data and maximum load data corresponding to the multiple network resources.

[0044] The third determining module is used to determine the first load system with the highest initial performance score as the initial load system for running the target application service.

[0045] In one possible implementation, the second determining module includes:

[0046] The calculation unit is used to calculate the ratio of the maximum load data to the resource demand data corresponding to each network resource.

[0047] The substitution unit is used to substitute the ratio into a preset function to obtain the resource performance data corresponding to the network resource, wherein the preset function is a monotonically increasing function;

[0048] The determining unit is used to determine the initial performance score corresponding to the first load system based on the resource performance data corresponding to the multiple network resources.

[0049] In one possible implementation, the determining unit includes:

[0050] Obtain sub-units to obtain the weights corresponding to each network resource;

[0051] The computation subunit is used to perform a weighted summation operation on the resource performance data corresponding to multiple network resources according to the weight corresponding to each network resource, so as to obtain the initial performance score corresponding to the first load system.

[0052] In one possible implementation, the device further includes:

[0053] The second feature acquisition module is used to acquire the first real-time load feature of the initial load system at preset time intervals during the process of running the target application service using the initial load system. The first real-time load feature includes first real-time load data corresponding to multiple network resources.

[0054] The threshold acquisition module is used to acquire a pre-set first threshold and a second threshold.

[0055] The fourth determining module is used to determine whether the load system for running the target application service needs to be redeployed based on the first real-time load characteristics, the first threshold, and the second threshold.

[0056] In one possible implementation, the fourth determining module includes:

[0057] The scoring calculation unit is used to substitute the first real-time load characteristic, the first threshold, and the second threshold into a preset formula to obtain the first real-time performance score of the initial load system, wherein the preset formula is as follows:

[0058] Q i (t)=[r ij (t)-θ 2j ][θ 1j -r ij (t)](Formula 1);

[0059] Among them, Q i (t) represents the first real-time performance score of the initial load system at time t, r ij θ(t) represents the first real-time load data of the j-th network resource in the initial load system at time t, where θ 1j Let θ represent the first threshold for the j-th network resource. 2j This represents the second threshold for the j-th network resource;

[0060] The first determining subunit is used to determine that the initial load system is operating normally and does not need to be redeployed to run the load system of the target application service when the first real-time performance score is greater than zero.

[0061] The second determining subunit is used to determine that the initial load system is operating abnormally and that the load system for running the target application service needs to be redeployed when the first real-time performance score is less than or equal to zero.

[0062] In one possible implementation, the device further includes:

[0063] The fifth determining module is used to determine multiple second load systems that can currently run the target application service when it is determined that the load system needs to be redeployed to run the target application service.

[0064] The third feature acquisition module is used to acquire the second real-time load feature of each second load system, wherein the second real-time load feature includes second real-time load data corresponding to multiple network resources.

[0065] The sixth determining module is used to determine the second real-time performance score corresponding to the second load system based on the resource demand data and the second real-time load data corresponding to the multiple network resources.

[0066] The seventh determination module is used to determine the second load system with the highest corresponding second real-time performance score as the target load system running the target application service;

[0067] The migration module is used to migrate the target application service to the target load system.

[0068] In one possible implementation, the device further includes:

[0069] The eighth determination module is used to determine the first load system whose initial performance score is greater than a preset score threshold as a candidate load system;

[0070] The priority acquisition module is used to acquire the system priority of each candidate load system;

[0071] The ninth determining module is used to determine the candidate load system with the highest priority of the corresponding system as the initial load system for running the target application service.

[0072] Thirdly, this application provides an apparatus comprising: a processor and a memory, the processor being configured to execute a load system scheduler stored in the memory to implement the load system scheduling method described in any one of the first aspects.

[0073] Fourthly, this application provides a storage medium storing one or more programs that can be executed by one or more processors to implement the load system scheduling method described in any one aspect.

[0074] Compared with the prior art, the technical solution provided in this application has the following advantages: The method provided in this application obtains service demand data corresponding to the target application service, wherein the service demand data includes resource demand data corresponding to multiple network resources, determines multiple first load systems that can currently run the target application service, and for each first load system, obtains the maximum load characteristic of the first load system, wherein the maximum load characteristic includes the maximum load data corresponding to multiple network resources, and determines the initial performance score corresponding to the first load system based on the resource demand data and maximum load data corresponding to multiple network resources, and determines the first load system with the highest initial performance score as the initial load system for running the target application service. Through this application, the load data of multiple network resources (such as CPU, memory, bandwidth, etc.) can be comprehensively considered to determine the initial load system for running the target application service, realizing multi-dimensional and globally optimal load scheduling of network resources, effectively avoiding the problem of unreasonable resource allocation caused by only considering a single load characteristic in the prior art, and improving the overall utilization efficiency of network resources. Attached Figure Description

[0075] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0076] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0077] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0078] Figure 1 A flowchart illustrating an embodiment of a load system scheduling method provided in this application;

[0079] Figure 2 A flowchart illustrating an embodiment of this application for determining an initial performance score for a first load system based on resource demand data and maximum load data corresponding to multiple network resources;

[0080] Figure 3 A flowchart illustrating an embodiment of another load system scheduling method provided in this application;

[0081] Figure 4 A block diagram illustrating an embodiment of a load system scheduling device provided in this application;

[0082] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0083] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0084] The following disclosure provides numerous different embodiments or examples for implementing various structures of this application. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the scope of this application. Furthermore, reference numerals and / or letters may be repeated in different examples. Such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed.

[0085] To address the problem that existing technologies, which only consider a single load characteristic, cannot meet the multi-dimensional network resource requirements of complex application services and suffer from unreasonable resource allocation, this application provides a load system scheduling method that can achieve multi-dimensional and globally optimal load scheduling of network resources. This effectively avoids the problem of unreasonable resource allocation caused by considering only a single load characteristic in existing technologies and improves the overall utilization efficiency of network resources.

[0086] Figure 1 This is a flowchart illustrating an embodiment of a load system scheduling method provided in this application. Figure 1 As shown, the method includes the following steps:

[0087] Step 101: Obtain the service requirement data corresponding to the target application service. The service requirement data includes resource requirement data corresponding to multiple network resources.

[0088] Application services refer to services that provide a specific function or solution through networks or mobile devices. In practical applications, application services are run through load balancing systems (e.g., server clusters, cloud service instances, etc.) to provide users with the specific function or solution corresponding to that application service.

[0089] The target application service refers to the application service of the system to which the load is to be allocated. In practice, it can be specified by the user.

[0090] Resource requirement data is used to characterize the target application service's demand for corresponding network resources (such as CPU resources, memory resources, and bandwidth resources), and is generally input by the user.

[0091] For example, the service demand data is D = (80, 120, 30), where 80 represents the demand for CPU resources, 120 represents the demand for memory resources, and 30 represents the demand for bandwidth resources. That is, the resource demand data corresponding to CPU resources is 80, the resource demand data corresponding to memory resources is 120, and the resource demand data corresponding to bandwidth resources is 30.

[0092] Step 102: Identify multiple primary load systems that can currently run the target application service.

[0093] The first load system refers to a load system that is in a working state and can provide services normally within a given time interval.

[0094] In practical applications, users can specify multiple primary load systems, or they can use automated tools or scripts to automatically scan networks or cloud platforms to discover multiple currently available primary load systems.

[0095] Step 103: For each first load system, obtain the maximum load characteristics of the first load system, wherein the maximum load characteristics include the maximum load data corresponding to multiple network resources.

[0096] The maximum load characteristic is used to characterize the maximum load of multiple network resources in the first load system, including the maximum load data corresponding to multiple network resources. The maximum load data is used to characterize the maximum load of the corresponding network resource in the first load system.

[0097] For example, the maximum load characteristic of load system A is LA = (100, 200, 50), where 100 represents the maximum CPU resource of load system A, 200 represents the maximum memory resource of load system A, and 50 represents the maximum bandwidth resource of load system A. That is, the maximum load data corresponding to CPU resource is 100, the maximum load data corresponding to memory resource is 200, and the maximum load data corresponding to bandwidth resource is 50.

[0098] In this embodiment of the application, the maximum load characteristics of each first load system are determined by collecting basic information of each first load system, such as configuration, location, network bandwidth, etc.

[0099] Step 104: Determine the initial performance score corresponding to the first load system based on the resource demand data and maximum load data corresponding to multiple network resources.

[0100] Step 105: Select the first load system with the highest initial performance score as the initial load system for running the target application service.

[0101] For ease of understanding, steps 104 and 105 will be explained uniformly below:

[0102] The initial performance score is used to characterize the performance of the target application service running on the corresponding first load system. The higher the initial performance score, the better the performance of the target application service running on the corresponding first load system.

[0103] In this embodiment, by setting and calculating the resource demand data and maximum load data corresponding to multiple network resources, an initial performance score corresponding to the first load system is obtained. Thus, the performance of each first load system can be evaluated based on the load characteristics of multi-dimensional network resources. Furthermore, by determining the first load system with the highest initial performance score as the initial load system for running the target application service, globally optimal load scheduling is achieved, improving the overall utilization efficiency of network resources.

[0104] In another embodiment of this application, after step 104, the following steps may be included: determining the first load system with an initial performance score greater than a preset score threshold as a candidate load system, obtaining the system priority of each candidate load system, and determining the candidate load system with the highest system priority as the initial load system for running the target application service.

[0105] A preset scoring threshold is used to select the first load system that can guarantee the running effect of the target application service. That is, when the initial performance score of the first load system is greater than the preset scoring threshold, it means that the running effect of the target application service can be guaranteed when running the target application service through the first load system.

[0106] System priority is used to characterize the usage priority of the corresponding first load system, and is generally set by the user according to actual needs.

[0107] Based on this, in this embodiment of the application, after selecting candidate load systems through a preset scoring threshold, the candidate load system with the highest priority can be determined as the initial load system for running the target application service. In this way, while ensuring the operational performance of the target application service, the determined initial load system can better meet user needs.

[0108] The technical solution provided in this application obtains service requirement data corresponding to a target application service, including resource requirement data corresponding to multiple network resources. It then determines multiple first load systems currently capable of running the target application service. For each first load system, it obtains the maximum load characteristic, which includes the maximum load data corresponding to multiple network resources. Based on the resource requirement data and maximum load data corresponding to multiple network resources, it determines the initial performance score of the first load system and identifies the first load system with the highest initial performance score as the initial load system for running the target application service. This application comprehensively considers the load data of multiple network resources (such as CPU, memory, and bandwidth) to determine the initial load system for running the target application service, achieving multi-dimensional, globally optimal load scheduling of network resources. This effectively avoids the problem of unreasonable resource allocation caused by considering only a single load characteristic in existing technologies, improving the overall utilization efficiency of network resources.

[0109] Figure 2 This application provides a flowchart of an embodiment for determining the initial performance score of a first load system based on resource demand data and maximum load data corresponding to multiple network resources. Figure 2 The process shown is in Figure 1 Based on the illustrated process, the following steps are included:

[0110] Step 201: For each network resource, calculate the ratio of the maximum load data to the resource demand data corresponding to the network resource.

[0111] Step 202: Substitute the ratio into a preset function to obtain the resource performance data corresponding to the network resource, wherein the preset function is a monotonically increasing function.

[0112] Step 203: Determine the initial performance score corresponding to the first load system based on the resource performance data corresponding to multiple network resources.

[0113] For ease of understanding, steps 201 to 203 will be explained uniformly below:

[0114] In this embodiment of the application, firstly, the ratio of the maximum load data to the resource demand data corresponding to each network resource is calculated. This ratio is used to characterize the load characteristics of the corresponding network resource on the target application service in the first load system.

[0115] Then, by substituting this ratio into a preset function (a monotonically increasing function), the corresponding resource performance data is obtained. Thus, the load characteristic (i.e., the ratio of the maximum load data to the resource demand data) is converted into a performance index (i.e., resource performance data). It can be understood that the larger the load characteristic, the higher the performance index and the larger the resource performance data.

[0116] In applications, the preset function can be f(x) = log(1+x). This function satisfies the condition of monotonically increasing, and as x increases, the growth rate of f(x) will gradually slow down. In this way, the growth trend of resource utilization in actual applications can be simulated, thereby making the obtained data more accurate.

[0117] Finally, the resource performance data corresponding to multiple network resources are set and calculated to obtain the initial performance score corresponding to the first load system.

[0118] Determining the initial performance score of the first load system based on resource performance data corresponding to multiple network resources may include the following steps:

[0119] Step A1: Obtain the weight corresponding to each network resource;

[0120] Step A2: Based on the weight corresponding to each network resource, perform a weighted summation operation on the resource performance data corresponding to multiple network resources to obtain the initial performance score corresponding to the first load system.

[0121] The aforementioned weights characterize the degree of impact of corresponding network resources on application service performance. In practical applications, users can flexibly configure these weights according to the specific needs of their application services to adapt to different application scenarios and service requirements.

[0122] As can be seen from the description of steps A1 and A2 above, in the technical solution of this application embodiment, the resource performance data corresponding to multiple network resources can be weighted and summed according to the weight corresponding to each network resource, so as to make the calculated initial performance score more accurate.

[0123] Specifically, assume that Li = (li1, li2, ..., lin) is the maximum load characteristic of the i-th load system, where lij represents the maximum load data of the j-th resource of the i-th load system (such as maximum CPU resource, maximum memory resource, maximum bandwidth resource, etc.); D = (d1, d2, ..., dn) is the indicator vector of application service requirements (i.e., service requirement data), where dj represents the resource requirement data of the target application service for the j-th resource; wj represents the weight of the j-th resource in the target application service; the preset function is f(x), which is a monotonically increasing function.

[0124] Then the initial performance score In other words, the initial performance score is obtained by comparing the maximum load data lij of each network resource with the service demand data dj of the target application service for that network resource, converting the comparison result into a performance index through the function f(x), multiplying it by the weight wj of the network resource, and finally summing the performance indices of all network resources.

[0125] Taking the preset function f(x) = log(1+x) as an example, assume there are two load systems A and B, whose maximum load characteristics include maximum CPU resources, maximum memory resources, and maximum bandwidth resources, which are LA = (100, 200, 50) and LB = (150, 100, 100) respectively. The application service requirement of the target application service is D = (80, 120, 30). Assume that the weights of the three network resources are equal, that is, w1 = w2 = w3 = 1 / 3.

[0126] The initial performance score of load system A is:

[0127] PA=1 / 3[log(1+80 / 100)+log(1+120 / 200)+log(1+30 / 50)];

[0128] The initial performance score of load system A is:

[0129] PB=1 / 3[log(1+80 / 150)+log(1+120 / 100)+log(1+30 / 100)];

[0130] After calculating the values ​​of PA and PB, the load system with the higher initial performance score can be selected as the initial load system by comparing the magnitudes of PA and PB.

[0131] Figure 2 The process shown is in Figure 1 Based on the process shown, the ratio of the maximum load data to the resource demand data corresponding to each network resource is calculated. This ratio is then substituted into a monotonically increasing function to obtain the corresponding resource performance data. Based on the resource performance data corresponding to multiple network resources, the initial performance score corresponding to the first load system is determined. This enables the performance evaluation of each first load system through the load characteristics of multi-dimensional network resources. Furthermore, in this scheme, by converting load characteristics into performance indicators, a quantitative evaluation of the processing capacity of the load system is achieved, which helps to more accurately select the load system that meets the application service requirements.

[0132] Figure 3 A flowchart illustrating another embodiment of a load system scheduling method provided in this application. Figure 3 As shown, the process includes the following steps:

[0133] Step 301: During the process of running the target application service using the initial load system, the first real-time load feature of the initial load system is obtained at preset time intervals. The first real-time load feature includes the first real-time load data corresponding to multiple network resources.

[0134] Step 302: Obtain the pre-set first threshold and second threshold;

[0135] Step 303: Based on the first real-time load characteristics, the first threshold, and the second threshold, determine whether it is necessary to redeploy the load system running the target application service.

[0136] For ease of understanding, steps 301 to 303 will be explained uniformly below:

[0137] The first real-time load characteristic is used to characterize the real-time load status of multiple network resources in the initial load system, including the first real-time load data corresponding to the multiple network resources. The first real-time load data is used to characterize the real-time load of the corresponding network resources in the initial load system.

[0138] The first threshold and the second threshold are set by the user based on experience. When the first real-time load characteristic is greater than the first threshold, the current load system is considered unsustainable and a target load system needs to be deployed and configured. When the first real-time load characteristic is less than the second threshold, the current load system is considered to be wasting performance and a target load system also needs to be deployed and configured.

[0139] In this embodiment of the application, in order to promptly detect and handle overloaded or idle resources, the first real-time load characteristics of the initial load system are obtained at preset time intervals. By comparing the first real-time load characteristics, the first threshold, and the second threshold, it is determined whether the current initial load system is overloaded or idle. Then, the determination result is used to determine whether the load system for running the target application service needs to be redeployed.

[0140] Specifically, determining whether the load system running the target application service needs to be redeployed based on the first real-time load characteristics, the first threshold, and the second threshold may include the following steps:

[0141] Step B1: Substitute the first real-time load characteristic, the first threshold, and the second threshold into a preset formula to obtain the first real-time performance score of the initial load system. The preset formula is as follows:

[0142] Q i (t)=[r ij (t)-θ 2j ][θ 1j -r ij (t)] (Formula 1);

[0143] Among them, Qi (t) represents the first real-time performance score of the initial load system at time t, r ij θ(t) represents the first real-time load data of the j-th network resource in the initial load system at time t, where θ 1j Let θ represent the first threshold for the j-th network resource. 2j This represents the second threshold for the j-th network resource;

[0144] Step B2: If the first real-time performance score is greater than zero, it is determined that the initial load system is operating normally and there is no need to redeploy the load system running the target application service.

[0145] Step B3: If the first real-time performance score is less than or equal to zero, it is determined that the initial load system is operating abnormally and the load system running the target application service needs to be redeployed.

[0146] As described in steps B1 to B3, according to Formula 1, when the first real-time load characteristic is between the first threshold and the second threshold, the first real-time performance score Q is... i (t) is positive; otherwise, Q i (t) is less than or equal to 0, that is, in Q i If (t) is greater than zero, it is determined that the initial load system is operating normally and there is no need to redeploy the load system running the target application service. i If (t) is less than or equal to zero, it is determined that the initial load system is operating abnormally (i.e., there is overload or idle situation), and the load system running the target application service needs to be redeployed.

[0147] In this scheme, Formula 1 can be used to calculate whether the first real-time performance score exceeds the threshold range or falls within the threshold range, thereby using the calculated value to assist in further judging the initial load system.

[0148] Furthermore, in yet another embodiment of this application, after step 303, the following steps may also be included:

[0149] Step C1: If it is determined that the load system running the target application service needs to be redeployed, identify multiple second load systems that can currently run the target application service;

[0150] Step C2: For each second load system, obtain the second real-time load characteristics of the second load system. The second real-time load characteristics include the second real-time load data corresponding to multiple network resources.

[0151] Step C3: Determine the second real-time performance score corresponding to the second load system based on the resource demand data and second real-time load data corresponding to multiple network resources.

[0152] Step C4: The second load system with the highest real-time performance score is identified as the target load system for running the target application service.

[0153] Step C5: Migrate the target application service to the target load system.

[0154] For ease of understanding, steps C1 to C5 will be explained uniformly below:

[0155] The second load system refers to a load system that is in a working state and can provide services normally within a given time interval.

[0156] The second real-time load feature is used to characterize the real-time load status of multiple network resources in the corresponding second load system, including the second real-time load data corresponding to the multiple network resources. The second real-time load data is used to characterize the real-time load of the corresponding network resources in the corresponding second load system.

[0157] In this embodiment of the application, when it is determined that the load system that needs to redeploy and run the target application service is to be redeployed, firstly, based on the resource demand data corresponding to multiple network resources and the second real-time load data of each second load system, the second real-time performance score corresponding to the second load system is determined. The process of determining the second real-time performance score is the same as the process of determining the initial performance score in step 104, and will not be described again here.

[0158] Furthermore, the second load system with the highest real-time performance score is selected as the target load system for running the target application service. Finally, the target application service is migrated to the target load system and continues to run on the target load system. This ensures the stable operation of the target application service and improves system stability.

[0159] Figure 3 The process described involves acquiring the first real-time load characteristics of the initial load system at preset time intervals. By comparing these first real-time load characteristics with a first threshold and a second threshold, it is determined whether the initial load system is overloaded or idle. The result then determines whether the load system for running the target application service needs to be redeployed. This allows for real-time monitoring of the initial load system's operational status, enabling timely detection and handling of overloaded or idle resources. This effectively avoids risks such as system performance degradation and crashes, enhancing the stability and reliability of the network system.

[0160] Figure 4 This is a block diagram illustrating an embodiment of a load system scheduling device provided in this application. Figure 4 As shown, the device includes:

[0161] The data acquisition module 41 is used to acquire service requirement data corresponding to the target application service, wherein the service requirement data includes resource requirement data corresponding to multiple network resources;

[0162] The first determining module 42 is used to determine multiple first load systems that can currently run the target application service;

[0163] The first feature acquisition module 43 is used to acquire the maximum load feature of each first load system, wherein the maximum load feature includes the maximum load data corresponding to multiple network resources.

[0164] The second determining module 44 is used to determine the initial performance score corresponding to the first load system based on the resource demand data and maximum load data corresponding to the multiple network resources.

[0165] The third determining module 45 is used to determine the first load system with the highest initial performance score as the initial load system for running the target application service.

[0166] In one possible implementation, the second determining module 44 includes:

[0167] The calculation unit is used to calculate the ratio of the maximum load data to the resource demand data corresponding to each network resource.

[0168] The substitution unit is used to substitute the ratio into a preset function to obtain the resource performance data corresponding to the network resource, wherein the preset function is a monotonically increasing function;

[0169] The determining unit is used to determine the initial performance score corresponding to the first load system based on the resource performance data corresponding to the multiple network resources.

[0170] In one possible implementation, the determining unit includes:

[0171] Obtain sub-units to obtain the weights corresponding to each network resource;

[0172] The computation subunit is used to perform a weighted summation operation on the resource performance data corresponding to multiple network resources according to the weight corresponding to each network resource, so as to obtain the initial performance score corresponding to the first load system.

[0173] In one possible implementation, the device further includes:

[0174] The second feature acquisition module is used to acquire the first real-time load feature of the initial load system at preset time intervals during the process of running the target application service using the initial load system. The first real-time load feature includes first real-time load data corresponding to multiple network resources.

[0175] The threshold acquisition module is used to acquire a pre-set first threshold and a second threshold.

[0176] The fourth determining module is used to determine whether the load system for running the target application service needs to be redeployed based on the first real-time load characteristics, the first threshold, and the second threshold.

[0177] In one possible implementation, the fourth determining module includes:

[0178] The scoring calculation unit is used to substitute the first real-time load characteristic, the first threshold, and the second threshold into a preset formula to obtain the first real-time performance score of the initial load system, wherein the preset formula is as follows:

[0179] Q i (t)=[r ij (t)-θ 2j ][θ 1j -r ij (t)](Formula 1);

[0180] Among them, Q i (t) represents the first real-time performance score of the initial load system at time t, r ij θ(t) represents the first real-time load data of the j-th network resource in the initial load system at time t, where θ 1j Let θ represent the first threshold for the j-th network resource. 2j This represents the second threshold for the j-th network resource;

[0181] The first determining subunit is used to determine that the initial load system is operating normally and does not need to be redeployed to run the load system of the target application service when the first real-time performance score is greater than zero.

[0182] The second determining subunit is used to determine that the initial load system is operating abnormally and that the load system for running the target application service needs to be redeployed when the first real-time performance score is less than or equal to zero.

[0183] In one possible implementation, the device further includes:

[0184] The fifth determining module is used to determine multiple second load systems that can currently run the target application service when it is determined that the load system needs to be redeployed to run the target application service.

[0185] The third feature acquisition module is used to acquire the second real-time load feature of each second load system, wherein the second real-time load feature includes second real-time load data corresponding to multiple network resources.

[0186] The sixth determining module is used to determine the second real-time performance score corresponding to the second load system based on the resource demand data and the second real-time load data corresponding to the multiple network resources.

[0187] The seventh determination module is used to determine the second load system with the highest corresponding second real-time performance score as the target load system running the target application service;

[0188] The migration module is used to migrate the target application service to the target load system.

[0189] In one possible implementation, the device further includes:

[0190] The eighth determination module is used to determine the first load system whose initial performance score is greater than a preset score threshold as a candidate load system;

[0191] The priority acquisition module is used to acquire the system priority of each candidate load system;

[0192] The ninth determining module is used to determine the candidate load system with the highest priority of the corresponding system as the initial load system for running the target application service.

[0193] like Figure 5 As shown in the figure, this application provides a device including a processor 111, a communication interface 112, a memory 113, and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114.

[0194] Memory 113 is used to store computer programs;

[0195] In one embodiment of this application, when the processor 111 executes a program stored in the memory 113, it implements the load system scheduling method provided in any of the foregoing method embodiments, including:

[0196] Obtain service requirement data corresponding to the target application service, wherein the service requirement data includes resource requirement data corresponding to multiple network resources;

[0197] Identify multiple first-load systems that can currently run the target application service;

[0198] For each first load system, the maximum load characteristic of the first load system is obtained, wherein the maximum load characteristic includes the maximum load data corresponding to multiple network resources;

[0199] Based on the resource demand data and maximum load data corresponding to the multiple network resources, determine the initial performance score corresponding to the first load system;

[0200] The first load system with the highest initial performance score is determined as the initial load system for running the target application service.

[0201] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the load system scheduling method provided in any of the foregoing method embodiments.

[0202] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0203] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0204] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also include the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0205] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A load system scheduling method, characterized by, The method includes: Obtain service requirement data corresponding to the target application service, wherein the service requirement data includes resource requirement data corresponding to multiple network resources; Identify multiple first-load systems that can currently run the target application service; For each first load system, the maximum load characteristic of the first load system is obtained, wherein the maximum load characteristic includes the maximum load data corresponding to multiple network resources; Based on the resource demand data and maximum load data corresponding to the multiple network resources, determine the initial performance score corresponding to the first load system; The first load system with the highest initial performance score is determined as the initial load system for running the target application service; The step of determining the initial performance score of the first load system based on the resource demand data and maximum load data corresponding to the multiple network resources includes: For each network resource, calculate the ratio of the maximum load data to the resource demand data corresponding to that network resource; Substituting the ratio into a preset function yields the resource performance data corresponding to the network resource, wherein the preset function is a monotonically increasing function; Based on the resource performance data corresponding to the multiple network resources, determine the initial performance score corresponding to the first load system; The method further includes: During the process of running the target application service using the initial load system, the first real-time load feature of the initial load system is obtained at preset time intervals. The first real-time load feature includes first real-time load data corresponding to multiple network resources. Obtain the pre-set first and second thresholds; Based on the first real-time load characteristic, the first threshold, and the second threshold, determine whether it is necessary to redeploy the load system running the target application service; If it is determined that the load system that needs to be redeployed to run the target application service is required, multiple second load systems that can currently run the target application service are identified. For each second load system, a second real-time load characteristic of the second load system is obtained, the second real-time load characteristic including second real-time load data corresponding to multiple network resources; Based on the resource demand data and the second real-time load data corresponding to the multiple network resources, a second real-time performance score corresponding to the second load system is determined. The second load system with the highest real-time performance score is identified as the target load system for running the target application service. Migrate the target application service to the target load system.

2. The method of claim 1, wherein, The step of determining the initial performance score of the first load system based on the resource performance data corresponding to the multiple network resources includes: Obtain the weight corresponding to each network resource; Based on the weight corresponding to each network resource, the resource performance data corresponding to multiple network resources are weighted and summed to obtain the initial performance score corresponding to the first load system.

3. The method of claim 1, wherein, The step of determining whether the load system for running the target application service needs to be redeployed based on the first real-time load characteristic, the first threshold, and the second threshold includes: Substituting the first real-time load characteristic, the first threshold, and the second threshold into a preset formula, the first real-time performance score of the initial load system is obtained, wherein the preset formula is as follows: (Formula 1); in, r represents the first real-time performance score of the initial load system at time t. ij θ(t) represents the first real-time load data of the j-th network resource in the initial load system at time t, where θ 1j Let θ represent the first threshold for the j-th network resource. 2j This represents the second threshold for the j-th network resource; If the first real-time performance score is greater than zero, it is determined that the initial load system is operating normally and there is no need to redeploy the load system running the target application service; If the first real-time performance score is less than or equal to zero, it is determined that the initial load system is operating abnormally and the load system for running the target application service needs to be redeployed.

4. The method according to claim 1, characterized in that, The method further includes: The first load system whose initial performance score is greater than the preset score threshold is identified as a candidate load system; Obtain the system priority for each candidate load system; The candidate load system with the highest priority is determined as the initial load system for running the target application service.

5. A load system scheduling device, characterized in that, The device includes: The data acquisition module is used to acquire service requirement data corresponding to the target application service, wherein the service requirement data includes resource requirement data corresponding to multiple network resources; The first determining module is used to determine multiple first load systems that can currently run the target application service; The feature acquisition module is used to acquire the maximum load feature of each first load system, wherein the maximum load feature includes the maximum load data corresponding to multiple network resources. The second determining module is used to determine the initial performance score corresponding to the first load system based on the resource demand data and maximum load data corresponding to the multiple network resources. The third determining module is used to determine the first load system with the highest initial performance score as the initial load system for running the target application service; The second determining module includes: The calculation unit is used to calculate the ratio of the maximum load data to the resource demand data corresponding to each network resource. The substitution unit is used to substitute the ratio into a preset function to obtain the resource performance data corresponding to the network resource, wherein the preset function is a monotonically increasing function; The determining unit is configured to determine the initial performance score corresponding to the first load system based on the resource performance data corresponding to the multiple network resources. The device further includes: The second feature acquisition module is used to acquire the first real-time load feature of the initial load system at preset time intervals during the process of running the target application service using the initial load system. The first real-time load feature includes first real-time load data corresponding to multiple network resources. The threshold acquisition module is used to acquire a pre-set first threshold and a second threshold. The fourth determining module is used to determine whether it is necessary to redeploy the load system running the target application service based on the first real-time load characteristics, the first threshold, and the second threshold. The fifth determining module is used to determine multiple second load systems that can currently run the target application service when it is determined that the load system needs to be redeployed to run the target application service. The third feature acquisition module is used to acquire the second real-time load feature of each second load system, wherein the second real-time load feature includes second real-time load data corresponding to multiple network resources. The sixth determining module is used to determine the second real-time performance score corresponding to the second load system based on the resource demand data and the second real-time load data corresponding to the multiple network resources. The seventh determination module is used to determine the second load system with the highest corresponding second real-time performance score as the target load system running the target application service; The migration module is used to migrate the target application service to the target load system.

6. A device, characterized in that, include: A processor and a memory, the processor being configured to execute a load system scheduler stored in the memory to implement the load system scheduling method according to any one of claims 1-4.

7. A storage medium, characterized in that, The storage medium stores one or more programs, which can be executed by one or more processors to implement the load system scheduling method according to any one of claims 1-4.

Images