A host scheduling method, device, equipment, medium and computer program

Abstract

The present invention discloses a host scheduling method, device, equipment, medium and computer program, which are applied to the field of intelligent scheduling technology, including: establishing a circular bull's eye model according to the weights of multiple indicators; determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator; determining the scheduling parameter value of each indicator using the indicator model according to the distance of each indicator from the bull's eye; determining the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determining the target scheduling host from multiple host machines according to the comprehensive scheduling parameter value of the host. The present invention uses the circular bull's eye model to perform calculations according to multiple indicators, and then determines the comprehensive scheduling parameter value, and selects a host from all host machines according to the comprehensive scheduling parameter value. Since the circular bull's eye model can comprehensively reflect the overall performance of the host, the accuracy of host determination can be improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent scheduling technology, and in particular to a host scheduling method, device, equipment, medium and computer program. Background Art

[0002] The traditional scheduling method of cloud platforms is to randomly select a host to create a virtual machine. If the resources on the host are insufficient, the creation will fail or the machine cannot be powered on after creation. Another scheduling strategy is to select a host based on a single indicator each time, such as the host with the most remaining memory. However, relying on a single indicator to select a host may be too simplistic and cannot fully reflect the actual performance and applicability of the host. Therefore, there is currently a problem of not being able to accurately select a host.

[0003] It can be seen that how to improve the accuracy of host machine scheduling is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a host scheduling method, device, equipment, medium and computer program, which can solve the technical problem of inaccurate host determination.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a host scheduling method, comprising:

[0006] A circular bull's-eye model is established according to the weights of multiple indicators; wherein the circular bull's-eye model is a model that uses the angle of a sector to determine the weight of each indicator and includes a maximum selection area and an optimal selection area, wherein the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area;

[0007] Determine the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator;

[0008] Determine the scheduling parameter value of each indicator by using an indicator model according to the distance of each indicator from the center of the target; wherein the indicator model is a model for determining the indicator parameter according to the distance of the indicator from the center of the target, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area;

[0009] The comprehensive scheduling parameter value of the host machine is determined according to the scheduling parameter value of each indicator corresponding to the host machine, and the target scheduling host machine is determined from multiple host machines according to the comprehensive scheduling parameter value of the host machine.

[0010] In some embodiments, the indicators include numerical indicators and non-numerical indicators; the non-numerical indicators are indicators that cannot be measured numerically.

[0011] In some embodiments, the numerical type indicators include at least one of CPU utilization, memory utilization, storage utilization, number of cloud hosts, number of network inputs and outputs, and number of processes, and the non-numerical type indicators include whether the CPU architecture is consistent with the virtual machine.

[0012] In some embodiments, the extreme value parameter is a maximum value and a minimum value, or the extreme value parameter is an optimal value and a worst value. In some embodiments, determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator includes:

[0013] The distance between each indicator and the bull's eye is determined by using the bull's eye distance formula according to the extreme value parameter and the indicator value corresponding to each indicator; wherein the bull's eye distance formula is: ; wherein said Indicates the indicator value, Indicates the maximum value, Indicates the minimum value, represents the optimal value, Indicates the worst value, represents the radius of the optimal selection area, and R represents the radius of the circular bull's eye model.

[0014] In some embodiments, the indicator model is ;in, Indicates the scheduling parameter value of each indicator, Indicates the distance of each indicator from the center of the target. Indicates the radius of the maximum selection area. represents the radius of the optimal selection area, Indicates the indicator angle.

[0015] In some embodiments, determining the scheduling parameter value of each indicator by using the indicator model according to the distance of each indicator from the target, includes:

[0016] When the distance between the indicator and the center of the target is greater than the radius of the maximum selection area, the indicator model is used to determine the scheduling parameter value of the indicator as a preset minimum value;

[0017] When the distance between the indicator and the bull's eye is greater than or equal to the radius of the optimal selection area and less than the radius of the maximum selection area, the indicator model is used to determine the scheduling parameter value of the indicator, which is the area of ​​the sector area of ​​the maximum selection area minus the area of ​​the sector area where the indicator value is located;

[0018] When the distance between the indicator and the center of the target is less than the radius of the optimal selection area, the indicator model is used to determine the scheduling parameter value of the indicator as the area of ​​the sector region where the indicator is located.

[0019] In some embodiments, determining a comprehensive scheduling parameter value of a host machine according to the scheduling parameter value of each indicator corresponding to the host machine, and determining a target scheduling host machine from a plurality of host machines according to the comprehensive scheduling parameter value of the host machine, comprises:

[0020] Performing a sum operation on the scheduling parameter value of each indicator corresponding to each of the host machines to determine the comprehensive scheduling parameter value of each of the host machines;

[0021] A maximum comprehensive scheduling parameter value is selected from the comprehensive scheduling parameter values ​​of each host machine, and a host machine corresponding to the maximum comprehensive scheduling parameter value is used as the target scheduling host machine.

[0022] In some embodiments, determining a comprehensive scheduling parameter value of a host machine according to the scheduling parameter value of each indicator corresponding to the host machine, and determining a target scheduling host machine from a plurality of host machines according to the comprehensive scheduling parameter value of the host machine, comprises:

[0023] When the multiple indicators do not include resource utilization parameters, the target scheduling host machine is determined according to the resource utilization and the comprehensive scheduling parameter value, so that the resource utilization of the target scheduling host machine is greater than a preset minimum threshold and less than a preset maximum threshold.

[0024] In some embodiments, when the multiple indicators do not include a resource utilization parameter, determining the target scheduling host according to the resource utilization and the comprehensive scheduling parameter value includes:

[0025] Determine the host machine corresponding to the maximum comprehensive scheduling parameter value, and when the resource utilization of the host machine is greater than the preset maximum threshold, or when the resource utilization of the host machine is less than the preset minimum threshold, determine to reselect the host machine as the target scheduling host machine.

[0026] In some embodiments, after determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator, the method further includes:

[0027] Determine whether the distance between the current indicator and the center of the target is greater than the radius of the maximum selection area;

[0028] When it is determined that the distance between the current indicator and the center of the target is greater than the radius of the maximum selection area, determining to filter the host machines corresponding to the indicators greater than the radius of the maximum selection area to obtain the host machines that are not filtered out;

[0029] Correspondingly, determining the comprehensive scheduling parameter value of the host machine according to the scheduling parameter value of each indicator corresponding to the host machine includes:

[0030] The comprehensive scheduling parameter value of the host machine that has not been filtered out is determined according to the scheduling parameter value of each indicator corresponding to the host machine that has not been filtered out.

[0031] The present invention also provides a host scheduling device, comprising:

[0032] A circular bull's-eye model determination module, used to establish a circular bull's-eye model according to the weights of multiple indicators; wherein the circular bull's-eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, wherein the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area;

[0033] A distance determination module, used to determine the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator;

[0034] A scheduling parameter value determination module, used to determine the scheduling parameter value of each indicator using an indicator model according to the distance of each indicator from the center of the target; wherein the indicator model is a model for determining the indicator parameter according to the distance of the indicator from the center of the target, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area;

[0035] The target scheduling host determination module is used to determine the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determine the target scheduling host from multiple hosts according to the comprehensive scheduling parameter value of the host.

[0036] The present invention also provides a host scheduling device, comprising:

[0037] Memory for storing computer programs;

[0038] The processor is used to execute the computer program to implement the steps of the host scheduling method as described above.

[0039] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the host scheduling method described above are implemented.

[0040] The present invention also provides a computer program product, including a computer program / instruction, which implements the steps of the above host scheduling method when executed by a processor.

[0041] The host scheduling method provided by the embodiment of the present invention may include: establishing a circular bull's eye model according to the weights of multiple indicators; wherein the circular bull's eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area; determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator; determining the scheduling parameter value of each indicator according to the distance of each indicator from the bull's eye using the indicator model; wherein the indicator model is a model that determines the indicator parameter according to the distance of the indicator from the bull's eye, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area; determining the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determining the target scheduling host from multiple host machines according to the comprehensive scheduling parameter value of the host.

[0042] It can be seen from the above technical scheme that the beneficial effect of the present invention is that compared with the original scheduling index for determining the host machine, which is single and makes it impossible to accurately select the host machine, the present invention will use the circular bull's-eye model to calculate according to multiple indicators when selecting the host machine, and then determine the comprehensive scheduling parameter value, and select the host machine from all the host machines according to the comprehensive scheduling parameter value. Since the circular bull's-eye model can comprehensively reflect the overall performance of the host machine, the accuracy of the host machine determination can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A flowchart of a host scheduling method provided by an embodiment of the present invention;

[0045] Figure 2 A flowchart of a host scheduling method provided by an embodiment of the present invention;

[0046] Figure 3 A topology diagram of host machine scheduling provided by an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of a circular bull's eye model provided by an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of the distance between an indicator and a bull's eye provided by an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of a comprehensive scheduling parameter value provided by an embodiment of the present invention;

[0050] Figure 7 A schematic diagram of the structure of a host scheduling device provided by an embodiment of the present invention;

[0051] Figure 8 A schematic diagram of the structure of a host scheduling device provided by an embodiment of the present invention;

[0052] Attached Figure 7-Figure 8 In the figure, the reference numerals are explained as follows: 100-circular bull's eye model determination module; 200-distance determination module; 300-scheduling parameter value determination module; 400-target scheduling host determination module 60-memory; 61-processor; 62-display screen; 63-input and output interface; 64-communication interface; 65-power supply; 66-communication bus; 601-computer program; 602-operating system; 603-data. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] The terms "including" and "having" in the specification of the present invention and the above-mentioned drawings, as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may include steps or units that are not listed.

[0055] Some of the terms that appear in the description of the embodiments of the present invention may be interpreted as follows:

[0056] Cloud platform: A cloud resource management platform on which virtual machines can be created.

[0057] Virtual machine: refers to the virtual machine in the cloud management platform. It is not a real server or host machine, but a virtual machine based on some virtualization technology that can achieve the same functions as a physical machine.

[0058] Bull's-eye algorithm: In data processing, especially when processing unbalanced data, the concept of "bull's-eye" may be used to refer to a certain center or key point. Different indicator values ​​can be marked on the target. The closer to the bull's-eye, the more it meets the requirements.

[0059] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0060] Next, a flowchart of a host scheduling method provided by an embodiment of the present invention is described in detail. Figure 1 A flowchart of a host scheduling method provided by an embodiment of the present invention, the host scheduling method may include:

[0061] S101, establishing a circular bull's-eye model according to the weights of multiple indicators; wherein the circular bull's-eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area.

[0062] The execution subject of this embodiment is an electronic device, for example, the electronic device can be a cloud platform, a server, etc. This embodiment does not limit specific indicators, and the indicators are performance indicators of the host machine that affect the resource scheduling of the cloud platform, such as CPU and memory utilization. All indicators are displayed in the shape of a sector in the circular bull's eye model. The size of the sector is the weight of different indicators. For example, if the memory utilization rate is more important than the number of cloud hosts, the angle of the memory utilization rate indicator is larger than the angle of the number of cloud hosts. For example, the multiple indicators in this embodiment can be CPU (central processing unit) utilization, memory utilization, storage utilization; or the multiple indicators in this embodiment can also be memory utilization, storage utilization, network IO (input and output), number of cloud hosts, number of processes, etc. The circular bull's eye model in this embodiment is a circular model with a radius of R. The circular bull's eye model has two dotted circular rings, the outer one is the selected area max (the maximum selected area), the radius is Rmax, and the inner ring is the optimal line of the selected area (optimal selected area), the radius is Rbest. For example, the outermost circle is the entire model, with a radius of R, and there are two rings inside. The outer one is the maximum selection area. If the host exceeds the index of this selected area, it will not be scheduled; the inner one is the optimal selection area. The scheduling parameter value of the index smaller than this selected area will be lower, and the scheduling parameter value of the index closer to the optimal selection area will be higher. The entire circle is divided into multiple sectors according to the number of indicators scheduled by the cloud host, and the size of the sector angle is defined according to the weight of the indicator. The maximum selection area and the optimal selection area in this embodiment are set according to demand.

[0063] It should be further explained that in order to make non-numerical indicators also be used as indicators for selecting the host machine and improve the accuracy of the construction of the circular bull's eye model, the above indicators may include numerical indicators and non-numerical indicators; non-numerical indicators are indicators that cannot be measured numerically. This embodiment does not limit specific numerical indicators. For example, the numerical indicator in this embodiment may be memory utilization; or the numerical indicator in this embodiment may also be storage utilization; or the numerical indicator in this embodiment may also include at least one of CPU utilization, memory utilization, storage utilization, number of cloud hosts, number of network inputs and outputs, and number of processes. This embodiment does not limit specific non-numerical indicators. For example, the non-numerical indicator in this embodiment may be whether the CPU architecture of the host machine is consistent with that of the virtual machine; or the non-numerical indicator in this embodiment may also be whether stains can be tolerated to see which cloud hosts of which services can tolerate this stain. The stain allows cluster managers to mark nodes, thereby limiting the scheduling of ordinary containers, while tolerance allows certain containers to ignore these marks and run on specific nodes. It can be understood that, currently, when determining the host machine based on parameters, the selection is mainly based on the utilization parameters of the host machine, and non-numerical types cannot be used as parameters for selecting the host machine. The indicators in the present invention include not only numerical indicators, but also non-numerical indicators. Since the indicators in the circular bull's-eye model are diverse, the accuracy of constructing the circular bull's-eye model can be improved.

[0064] S102, determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator.

[0065] The extreme value parameter in this embodiment refers to the extreme value corresponding to each indicator, which can be the maximum value, minimum value, optimal value and worst value. For example, the maximum value of memory utilization is 100%, and the minimum value of memory utilization is 0%; or the CPU architecture indicator has no range, and the host machine that meets the architecture must be selected, that is, the optimal value is its own CPU architecture, and the worst value is other architectures.

[0066] It should be further explained that, when the extreme value parameters in this embodiment are the maximum value and the minimum value, or the extreme value parameters are the optimal value and the worst value, the above-mentioned determination of the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameters corresponding to each indicator and the indicator value corresponding to each indicator may include: using the bull's eye distance formula to determine the distance of each indicator from the bull's eye according to the extreme value parameters and the indicator value corresponding to each indicator; wherein the bull's eye distance formula is ;in, Indicates the indicator value, Indicates the maximum value, Indicates the minimum value, represents the optimal value, represents the worst value, Represents the radius of the optimal selection area, and R represents the radius of the circular bull's eye model. This embodiment provides the corresponding formulas for determining the bull's eye distance when the extreme value parameters are of different types. According to the above formula, when the indicator is of numerical type, when the indicator value is the optimal value, the distance of the indicator from the bull's eye is the radius of the optimal selection area; when the indicator value is the worst value, the distance of the indicator from the bull's eye is the radius of the circular bull's eye model, and it will not be selected in the periphery of the maximum selection area; when the indicator type is of numerical type, the distance of the indicator from the bull's eye is the radius x (current value / (maximum value-minimum value)) of the bull's eye model, that is, if the indicator score is between the maximum and minimum values, the distance of the indicator from the bull's eye is calculated according to the ratio. This embodiment can directly use the bull's eye distance formula to determine the distance of each indicator from the bull's eye, so the efficiency and accuracy of determining the distance of the indicator from the bull's eye can be improved.

[0067] S103, determining the scheduling parameter value of each indicator according to the distance of each indicator from the center of the target using the indicator model; wherein the indicator model is a model for determining the indicator parameters according to the distance of the indicator from the center of the target, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area.

[0068] The indicator model in this embodiment is a method for determining the scheduling parameter value of each indicator based on the relationship between the distance of each indicator from the center of the target and the radius of the maximum selection area and the radius of the optimal selection area.

[0069] It should be further explained that in order to improve the accuracy of determining the scheduling parameter value of the indicator, the above indicator model is ;in, Indicates the scheduling parameter value of each indicator, Indicates the distance of each indicator from the bull's eye. Indicates the radius of the maximum selection area, represents the radius of the optimal selection area, Indicates the indicator angle. This embodiment provides a specific formula for determining the scheduling parameter value of the indicator, so that after obtaining the relevant parameters, calculation can be performed directly according to the formula, thereby improving the accuracy of determining the scheduling parameter value of the indicator.

[0070] It should be further explained that the above-mentioned determination of the scheduling parameter value of each indicator according to the distance of each indicator from the bull's eye using the indicator model may include: when the distance of the indicator from the bull's eye is greater than the radius of the maximum selection area, using the indicator model to determine the scheduling parameter value of the indicator as a preset minimum value; when the distance of the indicator from the bull's eye is greater than or equal to the radius of the optimal selection area, and less than the radius of the maximum selection area, using the indicator model to determine the scheduling parameter value of the indicator as the area of ​​the fan-shaped area of ​​the maximum selection area minus the area of ​​the fan-shaped area where the indicator value is located; when the distance of the indicator from the bull's eye is less than the radius of the optimal selection area, using the indicator model to determine the scheduling parameter value of the indicator as the area of ​​the fan-shaped area where the indicator value is located. The indicator model in this embodiment is a specific corresponding relationship. If the distance between the indicator and the bull's eye is greater than the maximum selection area, the scheduling parameter value is the minimum value of the budget, for example, the preset minimum value is negative infinity; or the preset minimum value in this embodiment is other negative numbers or zero. If the distance between the indicator and the bull's eye is between the maximum selection area and the optimal selection area, the area between the maximum selection areas is calculated. The closer to the maximum selection area, the smaller the area, and the closer to the optimal selection area, the larger the area; if the distance between the indicator and the bull's eye is within the optimal selection area, the fan-shaped area from the distance between the indicator and the bull's eye to the center of the circle is calculated. The closer to the center of the circle, the lower the resource utilization, the lower the score, and the closer to the optimal selection area, the higher the score.

[0071] It should be further explained that in order to improve the efficiency of subsequent determination of the target scheduling host, after determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameters corresponding to each indicator and the indicator value corresponding to each indicator, it can also include: judging whether the distance of the current indicator from the bull's eye is greater than the radius of the maximum selection area; when it is determined that the distance of the current indicator from the bull's eye is greater than the radius of the maximum selection area, filtering the host corresponding to the indicator greater than the radius of the maximum selection area to obtain the host that is not filtered out; accordingly, determining the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, including: determining the comprehensive scheduling parameter value of the host that is not filtered out according to the scheduling parameter value of each indicator corresponding to the host that is not filtered out. This embodiment will filter the host before determining the comprehensive scheduling parameter value of the host, reducing the number of hosts for subsequent parameter calculation, thereby improving the efficiency of subsequent determination of the target scheduling host.

[0072] S104, determining a comprehensive scheduling parameter value of the host machine according to the scheduling parameter value of each indicator corresponding to the host machine, and determining a target scheduling host machine from multiple host machines according to the comprehensive scheduling parameter value of the host machine.

[0073] This embodiment needs to determine the scheduling parameter value of each indicator corresponding to each host machine, so as to sum the scheduling parameter values ​​of each indicator to determine the comprehensive scheduling parameter value of the host machine, and then determine the target scheduling parameter value according to the comprehensive scheduling parameter value corresponding to each host machine.

[0074] It should be further explained that, in order to determine the accuracy of the target host, the above-mentioned determination of the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and the determination of the target scheduling host from multiple hosts according to the comprehensive scheduling parameter value of the host, can include: summing the scheduling parameter value of each indicator corresponding to each host to determine the comprehensive scheduling parameter value of each host; selecting the largest comprehensive scheduling parameter value from the comprehensive scheduling parameter value of each host, and taking the host corresponding to the largest comprehensive scheduling parameter value as the target scheduling host. According to the setting of the circular bull's eye model, this embodiment displays multiple indicators in the form of a fan on the model, and calculates the position of each indicator in the model, and calculates the final comprehensive scheduling parameter value in combination with the selected area and weight in the model, and the host with the largest comprehensive scheduling parameter value is scheduled. After calculating the scheduling parameter value of each indicator in this embodiment, the comprehensive scheduling parameter value of the host is the sum of the scheduling parameter values ​​of all indicators. If there is an indicator that does not meet the requirements, the score of the indicator is negative infinity, and the comprehensive score of the host will also be negative infinity, and it will be directly eliminated. The remaining hosts with higher scores greater than 0 are close to the optimal selection area in terms of indicators, and are the most suitable hosts, which will eventually be scheduled. The comprehensive scheduling parameter value corresponding to the host is , Indicates each indicator. This embodiment uses the host machine with the largest comprehensive scheduling parameter value as the target host machine, ensuring that the host machine that best meets the requirements is selected from multiple host machines, thereby optimizing resource utilization and improving scheduling efficiency.

[0075] It should be further explained that, in order to improve the overall performance and efficiency of the system, the above-mentioned determination of the comprehensive scheduling parameter value of the host machine according to the scheduling parameter value of each indicator corresponding to the host machine, and the determination of the target scheduling host machine from multiple host machines according to the comprehensive scheduling parameter value of the host machine may include: when the resource utilization parameter is not included in the multiple indicators, the target scheduling host machine is determined according to the resource utilization and the comprehensive scheduling parameter value, so that the resource utilization of the target scheduling host machine is greater than the preset minimum threshold and less than the preset maximum threshold. This embodiment does not limit the specific resource utilization. For example, the resource utilization in this embodiment can be the current usage of the CPU, memory, storage and network of each host machine. When the resource utilization parameter is not included in the multiple indicators of this embodiment, the method of determining the target scheduling host machine according to the resource utilization and the comprehensive scheduling parameter value has many advantages. It can ensure that the selected host machine reaches the optimal state in terms of comprehensive performance and resource utilization, thereby meeting business needs and improving the overall performance and efficiency of the system.

[0076] It should be further explained that in order to improve the accuracy of the target host machine determination, when the resource utilization parameter is not included in the above multiple indicators, the target scheduling host machine is determined according to the resource utilization and the comprehensive scheduling parameter value, which may include: determining the host machine corresponding to the largest comprehensive scheduling parameter value, and when the resource utilization of the host machine is greater than the preset maximum threshold, or when the resource utilization of the host machine is less than the preset minimum threshold, determining to reselect the host machine as the target scheduling host machine. In this embodiment, if the resource utilization of the target host machine is lower than the preset minimum threshold, it means that the resources of the host machine are not fully utilized and cannot meet the business needs. In this case, it is necessary to reselect the host machine to ensure the effective use of resources. If the resource utilization of the target host machine is higher than the preset maximum threshold, it means that the resources of the host machine are overused, which may lead to performance degradation or stability problems. In order to avoid these problems, it is necessary to reselect the host machine to ensure the rational use of resources.

[0077] It should be further explained that, in order to improve the utilization rate of the host machine, the comprehensive scheduling parameter value of the host machine is determined according to the scheduling parameter value of each indicator corresponding to the host machine, and the target scheduling host machine is determined from multiple host machines according to the comprehensive scheduling parameter value of the host machine, which may include: determining multiple target scheduling host machines from multiple host machines according to the comprehensive scheduling parameter value of the host machine; determining the scheduled host machines among the multiple target scheduling host machines according to the historical scheduling information corresponding to each host machine; judging the number of scheduled host machines; when the number of scheduled host machines is single, directly taking the scheduled host machine as the target scheduling host machine; when the number of scheduled host machines is multiple, determining the scheduled host machine whose resource surplus is within the preset resource surplus interval as the target scheduling host machine. This embodiment preferentially calls the scheduled host machine, and when the number of scheduled host machines is multiple, preferentially selects the host machine with moderate resource surplus, so as to improve the utilization rate of the host machine as much as possible, reduce the number of host machines used, and improve resource utilization.

[0078] The host scheduling method provided by the embodiment of the present invention may include: S101, establishing a circular bull's eye model according to the weights of multiple indicators; wherein the circular bull's eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area; S102, determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator; S103, determining the scheduling parameter value of each indicator according to the distance of each indicator from the bull's eye using the indicator model; wherein the indicator model is a model that determines the indicator parameter according to the distance of the indicator from the bull's eye, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area; S104, determining the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determining the target scheduling host from multiple host machines according to the comprehensive scheduling parameter value of the host. Compared with the original host scheduling index being single and making it impossible to accurately select a host, the present invention, when selecting a host, will use a circular bull's-eye model to perform calculations based on multiple indexes, and then determine a comprehensive scheduling parameter value, and select a host from all host machines based on the comprehensive scheduling parameter value. Since the circular bull's-eye model can comprehensively reflect the overall performance of the host, the accuracy of host determination is improved, and when the multiple indexes include multiple resource utilization indexes, the multiple indexes can be balanced, the utilization of the host can be improved as much as possible, the number of host machines in use can be reduced, and resource utilization can be improved. Furthermore, the indicators in the present invention include not only numerical indicators but also non-numerical indicators. Since the indicators in the circular bull's-eye model are diverse, the accuracy of constructing the circular bull's-eye model can be improved; and, this embodiment can directly use the bull's-eye distance formula to determine the distance of each indicator from the bull's-eye, so the efficiency and accuracy of determining the distance of the indicator from the bull's-eye can be improved; and, in this embodiment, a specific formula for determining the scheduling parameter value of the indicator is given, so that after obtaining the relevant parameters, calculation can be directly performed according to the formula, thereby improving the accuracy of determining the scheduling parameter value of the indicator; and, before determining the comprehensive scheduling parameter value of the host machine, this embodiment will perform a The filtering process is performed to reduce the number of hosts for subsequent parameter calculation, thereby improving the efficiency of subsequent determination of the target scheduling host; and, this embodiment uses the host with the largest comprehensive scheduling parameter value as the target host, ensuring that the host that best meets the requirements is selected from multiple host machines, thereby optimizing resource utilization and improving scheduling efficiency; and, when the resource utilization parameter is not included in multiple indicators, the method of determining the target scheduling host according to the resource utilization and the comprehensive scheduling parameter value has many advantages, which can ensure that the selected host reaches the optimal state in terms of comprehensive performance and resource utilization, thereby meeting business needs and improving the overall performance and efficiency of the system.

[0079] In order to make the present invention easier to understand, please refer to Figure 2 , Figure 2 An example flow chart of a host scheduling method provided by an embodiment of the present invention may specifically include:

[0080] S201, establishing a circular bull's-eye model according to multiple indicators; wherein the multiple indicators are indicators corresponding to each host machine; the radius of the circular bull's-eye model is R, and the circular bull's-eye model includes a ring with a radius of Rmax and a ring with a radius of Rbest, and the size of the sector angle is determined according to the weight of each indicator.

[0081] The topology diagram of the virtual machine scheduling host machine corresponding to this embodiment is as follows: Figure 3 As shown, Figure 3 A topology diagram of a host scheduling provided by an embodiment of the present invention. In this embodiment, Rmax is the maximum selection area, and Rbest is the optimal selection area. The entire circle is divided into multiple sectors according to the cloud host scheduling indicators, and the size of the sector angle is defined according to the weight of the indicator. For ease of understanding, please refer to Figure 4 , Figure 4 A schematic diagram of a circular bull's eye model provided by an embodiment of the present invention, from Figure 4 It can be seen that the indicators include central processing unit (CPU) utilization, memory utilization, storage utilization, number of cloud hosts, number of network input and output (IO), number of processes and many other indicators. Figure 4 The selected area max and the selected area best in correspond to the maximum selected area and the optimal selected area in the above text respectively.

[0082] S202, defining a calculated value of each indicator, and determining the distance of each indicator from the center of the target according to the measured value; wherein the measured value includes at least one of a maximum value, a minimum value, an optimal value, and a worst value.

[0083] This embodiment defines the measurement value of each indicator (corresponding to the extreme value parameter above) to calculate the distance from each indicator to the bull's eye. Each indicator has four measurement values, the maximum value (Vmax), the minimum value (Vmin), the optimal value (Vbest), and the worst value (Vbad). For example, for CPU utilization, the maximum value is 100%, and the minimum value is 0%. For example, for indicators such as CPU architecture, there is no range, and the host machine that meets the architecture must be screened. That is, the optimal value is its own CPU architecture, and the worst value is other architectures. This embodiment calculates the position of the indicator in the circular bull's eye model for each indicator of each host machine. For example, if the CPU utilization of host machine A is 50%, then its distance from the bull's eye is the radius of the bull's eye model * (current value / (maximum value-minimum value)) is R multiplied by (50% / (Vmax-Vmin)). For example, if the CPU architecture of the host machine is x86, and the CPU architecture of the virtual machine to be created is also x86, which happens to be in line with it, then his distance from the bull's eye is Rbest, which has the highest score. If you want to create an arm virtual machine, it does not meet the distance from the bull's eye. The score is the least. In summary, if the indicator value is x, then the calculation formula for the distance between the indicator and the center of the target is ;in, Indicates the distance of each indicator from the bull's eye. represents the radius of the circular bull's-eye model, Indicates the maximum value of the indicator. Indicates the minimum value of the indicator. represents the optimal value of the indicator, Indicates the worst value of the indicator, represents the radius of the optimal selection area. According to this formula, the distance from the center of the target is calculated for each indicator, such as Figure 5 As shown, Figure 5 A schematic diagram of the distance between an indicator and a bull's eye provided by an embodiment of the present invention, Figure 5 The black dot is the index value.

[0084] S203, determining the scheduling parameter value of each indicator according to the distance of each indicator from the center of the target, and determining the comprehensive scheduling parameter value of each host machine according to the scheduling parameter value of each indicator.

[0085] In order to maximize the resource utilization of the host machine, this embodiment does not cause downtime due to excessive load. If the distance of the target indicator exceeds Vmax, the host machine will be filtered out. Brand new host machines with low utilization of various indicators or cloud hosts with utilization close to the Rmax threshold are not the first choice. Combined with the punctuation in the previous step, the host machine with more indicators around Rbest has a high probability of being selected, and the closer to Rbest within a reasonable range, the higher the score should be. For points inside Rbest, resource utilization is relatively low and the priority is lower. Combining the above ideas, the method for finally calculating the comprehensive scheduling parameter value of the host machine can be abstracted as Figure 6 , Figure 6 A schematic diagram of a comprehensive scheduling parameter value provided in an embodiment of the present invention. The final comprehensive scheduling parameter value of the host machine is the total area of ​​the green part in the figure. In the selected area, the lower the utilization rate, the higher the score. If an indicator exceeds the selected area, the host machine will be excluded. If it is less than the selected area, the closer to the center of the target, the smaller the scheduling parameter value.

[0086] Based on the above derivation, assuming that there are n indicators, each indicator value is xn, and the indicator angle is k, combined with the above bull's eye distance formula f(x), the calculation formula for the scheduling parameter value of each indicator is:

[0087] ;

[0088] Combined by formula Figure 6 It can be seen that, ① when the distance of the center of the target is greater than the maximum selection area, it takes negative infinity, and the value of this target is infinitely small (such as Figure 6 ② When the bull's eye distance of the indicator is between the maximum selection area and the optimal selection area, take the area of ​​the sector area of ​​the maximum selection area minus the area of ​​the sector area where the indicator value is located. ③ When the bull's eye distance of the indicator is less than the optimal selection area, take the area of ​​the sector area where the indicator is located (such as Figure 6 The number of cloud hosts in the index).

[0089] S204: Determine a target scheduling host from all host machines according to the comprehensive scheduling parameter value.

[0090] In this embodiment, the comprehensive scheduling parameter value of the host machine is the sum of the scheduling parameter values ​​of each indicator. If the host machine has an indicator that does not meet the requirements, its final comprehensive scheduling parameter value is negative infinity. Among them, the comprehensive scheduling parameter value of the host machine is the sum of the scheduling parameter values ​​of each indicator, and the formula is:

[0091] ; When scheduling, the cloud platform can select the host machine with the largest comprehensive scheduling parameter value as the target scheduling host machine. The basic idea of ​​this embodiment is: define multiple indicators on a circular model, present them in the form of a sector, and use the angle of the sector to identify the weight of the indicator. Based on the bull's eye algorithm and improved on this basis, the values ​​of different indicators are uniformly calculated as the bull's eye distance, and the bull's eye prototype is punctuated to extract the bull's eye distance formula. The distance of each indicator from the center of the circle is used for punctuation. The closer the indicator falls to the optimal selection area, the higher the score and the greater the probability of selection. The formula for determining the comprehensive parameter value of the host machine is extracted.

[0092] The present invention abandons the disadvantages of the original cloud resource scheduling with a single indicator and excess scheduling resources without better resource utilization. It adopts multi-indicator balanced scheduling. When scheduling, it does not schedule to a host machine whose resource utilization rate exceeds a predetermined threshold. It tries to schedule to a host machine with a moderate resource utilization rate first, fully utilizes the existing resources, and then schedules to new host machine resources, so that the overall cloud resources are fully utilized and energy consumption is saved. The hosts of this algorithm are relatively independent and can perform multi-threaded concurrent operations, which speeds up the scheduling of virtual machines and improves the usability of the cloud platform.

[0093] The host scheduling device provided by an embodiment of the present invention is introduced below. The host scheduling device described below and the host scheduling method described above can be referred to each other.

[0094] Figure 7 A schematic diagram of the structure of a host scheduling device provided by an embodiment of the present invention may include:

[0095] The circular bull's-eye model determination module 100 is used to establish a circular bull's-eye model according to the weights of multiple indicators; wherein the circular bull's-eye model is a model that uses the angle of a sector to determine the weight of each indicator and includes a maximum selection area and an optimal selection area, wherein the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area;

[0096] A distance determination module 200, for determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator;

[0097] The scheduling parameter value determination module 300 is used to determine the scheduling parameter value of each indicator using an indicator model according to the distance of each indicator from the center of the target; wherein the indicator model is a model for determining the indicator parameter according to the distance of the indicator from the center of the target, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area;

[0098] The target scheduling host determination module 400 is used to determine the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determine the target scheduling host from multiple hosts according to the comprehensive scheduling parameter value of the host.

[0099] Further, based on the above embodiment, the indicators may include numerical indicators and non-numerical indicators; the non-numerical indicators are indicators that cannot be measured by numerical values.

[0100] Further, based on the above embodiments, the numerical type indicators include at least one of central processing unit utilization, memory utilization, storage utilization, number of cloud hosts, number of network inputs and outputs, and number of processes, and the non-numerical type indicators include whether the central processing unit architecture is consistent with the virtual machine.

[0101] Further, based on any of the above embodiments, the extreme value parameters are maximum values ​​and minimum values, or the extreme value parameters are optimal values ​​and worst values.

[0102] Further, based on any of the above embodiments, the distance determination module 200 may include:

[0103] A distance determination unit is used to determine the distance of each indicator from the bull's eye according to the extreme value parameter corresponding to each indicator and the indicator value using a bull's eye distance formula; wherein the bull's eye distance formula is: ; wherein said Indicates the indicator value, Indicates the maximum value, Indicates the minimum value, represents the optimal value, Indicates the worst value, represents the radius of the optimal selection area, and R represents the radius of the circular bull's eye model.

[0104] Further, based on any of the above embodiments, the above indicator model is

[0105] ;in, Indicates the scheduling parameter value of each indicator, Indicates the distance of each indicator from the center of the target. Indicates the radius of the maximum selection area. represents the radius of the optimal selection area, Indicates the indicator angle.

[0106] Further, based on any of the above embodiments, the above scheduling parameter value determination module 300 may include:

[0107] A first parameter value determination unit, configured to determine, by using the indicator model, a scheduling parameter value of the indicator as a preset minimum value when the distance between the indicator and the center of the target is greater than the radius of the maximum selection area;

[0108] A second parameter value determination unit is used to determine, when the distance between the indicator and the center of the target is greater than or equal to the radius of the optimal selection area and less than the radius of the maximum selection area, the scheduling parameter value of the indicator is determined by using the indicator model to be the area of ​​the sector area of ​​the maximum selection area minus the area of ​​the sector area where the indicator value is located;

[0109] The third parameter value determination unit is used to determine the scheduling parameter value of the indicator as the area of ​​the fan-shaped region where the indicator is located by using the indicator model when the distance between the indicator and the center of the target is less than the radius of the optimal selection area.

[0110] Further, based on any of the above embodiments, the target scheduling host determination module 400 may include:

[0111] A comprehensive scheduling parameter value determination unit, used to perform a sum operation on the scheduling parameter value of each indicator corresponding to each of the host machines to determine the comprehensive scheduling parameter value of each of the host machines;

[0112] The target scheduling host machine determination unit is used to select the maximum comprehensive scheduling parameter value from the comprehensive scheduling parameter values ​​of each host machine, and use the host machine corresponding to the maximum comprehensive scheduling parameter value as the target scheduling host machine.

[0113] Further, based on any of the above embodiments, the target scheduling host determination module 400 may include:

[0114] The host unit is determined according to the resource utilization, and is used to determine the target scheduling host according to the resource utilization and the comprehensive scheduling parameter value when the resource utilization parameter is not included in the multiple indicators, so that the resource utilization of the target scheduling host is greater than the preset minimum threshold and less than the preset maximum threshold.

[0115] Further, based on the above embodiment, determining the host unit according to the resource utilization rate may include:

[0116] The host subunit is determined according to the resource utilization, and is used to determine the host corresponding to the maximum comprehensive scheduling parameter value. When the resource utilization of the host is greater than the preset maximum threshold, or when the resource utilization of the host is less than the preset minimum threshold, it is determined to reselect the host as the target scheduling host.

[0117] Further, based on any of the above embodiments, the host scheduling method may further include:

[0118] Determine whether the distance between the current indicator and the center of the target is greater than the maximum selection area;

[0119] When it is determined that the distance between the current indicator and the center of the target is greater than the maximum selection area, determining to filter the host machines corresponding to the indicators greater than the maximum selection area;

[0120] Accordingly, the target scheduling host determination module 400 includes:

[0121] A host unit is determined according to the filtering result, and is used to determine the comprehensive scheduling parameter value of the host machine that has not been filtered out according to the scheduling parameter value of each indicator corresponding to the host machine that has not been filtered out.

[0122] A judgment module, used to judge whether the distance between the current indicator and the center of the target is greater than the radius of the maximum selection area;

[0123] A filtering module, for filtering host machines corresponding to the indicators greater than the radius of the maximum selection area to obtain unfiltered host machines when it is determined that the distance between the current indicator and the target center is greater than the radius of the maximum selection area;

[0124] Accordingly, the target scheduling host determination module 400 includes:

[0125] A host unit is determined according to the filtering result, and is used to determine the comprehensive scheduling parameter value of the host machine that has not been filtered out according to the scheduling parameter value of each indicator corresponding to the host machine that has not been filtered out.

[0126] It should be noted that the order of the modules and units in the above host scheduling device can be changed without affecting the logic.

[0127] Figure 7 The description of the features in the corresponding embodiments can be found in Figure 1 The relevant descriptions of the corresponding embodiments will not be repeated here one by one.

[0128] The host scheduling device provided by the embodiment of the present invention may include: a circular bull's eye model determination module 100, which is used to establish a circular bull's eye model according to the weights of multiple indicators; wherein the circular bull's eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, wherein the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area; a distance determination module 200, which is used to determine the distance of each indicator according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator. The distance from the center of the circular bull's eye model; a scheduling parameter value determination module 300, used to determine the scheduling parameter value of each indicator using the indicator model according to the distance of each indicator from the center of the bull's eye; wherein the indicator model is a model for determining indicator parameters according to the distance from the indicator to the center of the bull's eye, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area; a target scheduling host determination module 400, used to determine the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determine the target scheduling host from multiple host machines according to the comprehensive scheduling parameter value of the host. Compared with the original host scheduling index being single and making it impossible to accurately select a host, the present invention, when selecting a host, will use a circular bull's-eye model to perform calculations based on multiple indexes, and then determine a comprehensive scheduling parameter value, and select a host from all host machines based on the comprehensive scheduling parameter value. Since the circular bull's-eye model can comprehensively reflect the overall performance of the host, the accuracy of host determination is improved, and when the multiple indexes include multiple resource utilization indexes, the multiple indexes can be balanced, the utilization of the host can be improved as much as possible, the number of host machines in use can be reduced, and resource utilization can be improved.Furthermore, the indicators in the present invention include not only numerical indicators but also non-numerical indicators. Since the indicators in the circular bull's-eye model are diverse, the accuracy of constructing the circular bull's-eye model can be improved; and, this embodiment can directly use the bull's-eye distance formula to determine the distance of each indicator from the bull's-eye, so the efficiency and accuracy of determining the distance of the indicator from the bull's-eye can be improved; and, in this embodiment, a specific formula for determining the scheduling parameter value of the indicator is given, so that after obtaining the relevant parameters, calculation can be directly performed according to the formula, thereby improving the accuracy of determining the scheduling parameter value of the indicator; and, before determining the comprehensive scheduling parameter value of the host machine, this embodiment will perform a The filtering process is performed to reduce the number of hosts for subsequent parameter calculation, thereby improving the efficiency of subsequent determination of the target scheduling host; and, this embodiment uses the host with the largest comprehensive scheduling parameter value as the target host, ensuring that the host that best meets the requirements is selected from multiple host machines, thereby optimizing resource utilization and improving scheduling efficiency; and, when the resource utilization parameter is not included in multiple indicators, the method of determining the target scheduling host according to the resource utilization and the comprehensive scheduling parameter value has many advantages, which can ensure that the selected host reaches the optimal state in terms of comprehensive performance and resource utilization, thereby meeting business needs and improving the overall performance and efficiency of the system.

[0129] A host scheduling device provided by an embodiment of the present invention is introduced below. The host scheduling device described below and the host scheduling method described above can refer to each other.

[0130] Figure 8 A schematic diagram of the structure of a host scheduling device provided by an embodiment of the present invention is shown in FIG. Figure 8 As shown, the host scheduling device includes: a memory 60 for storing computer programs;

[0131] The processor 61 is used to implement the steps of the host scheduling method in the above embodiment when executing a computer program.

[0132] The host machine scheduling provided in this embodiment may include but is not limited to a smart phone, a tablet computer, a laptop computer, or a desktop computer.

[0133] Among them, the processor 61 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 61 can be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 61 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 61 may be integrated with a graphics processing unit (GPU), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 61 may also include an artificial intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0134] The memory 60 may include one or more computer-readable storage media, which may be non-transitory. The memory 60 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 60 is at least used to store the following computer program 601, wherein, after the computer program is loaded and executed by the processor 61, it can implement the relevant steps of the host scheduling method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 60 may also include an operating system 602 and data 603, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 602 may include Windows, Unix, Linux, etc. Data 603 may include but is not limited to host scheduling data, etc.

[0135] In some embodiments, the host scheduling device may further include a display screen 62 , an input / output interface 63 , a communication interface 64 , a power supply 65 , and a communication bus 66 .

[0136] Those skilled in the art will understand that Figure 8 The structure shown in does not constitute a limitation on the host scheduling device and may include more or fewer components than shown in the figure.

[0137] The present invention also provides a computer program product, including a computer program / instruction, which implements the steps of the above host scheduling method when executed by a processor.

[0138] It is understandable that if the host scheduling method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or part of the contribution to the current technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, magnetic disk or optical disk and other media that can store program code.

[0139] Based on this, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the host scheduling method described above are implemented.

[0140] The above is a detailed introduction to a host scheduling method, device, equipment, medium and computer program provided by an embodiment of the present invention. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0141] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0142] The above is a detailed introduction to a host scheduling method, device, equipment, medium and computer program provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, the present invention can also be improved and modified, and these improvements and modifications also fall within the scope of protection of the present invention.

Claims

1. A host scheduling method, characterized in that: include: A circular bull's-eye model is established according to the weights of multiple indicators; wherein the circular bull's-eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, wherein the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area; the outermost circle of the circular bull's-eye model is the entire circular bull's-eye model, and the entire circular bull's-eye model includes two circular rings, the outer one is the maximum selection area, and the host machine with indicators exceeding the maximum selection area will not be scheduled; the inner one is the optimal selection area, and the scheduling parameter value of the indicator closer to the center of the circular bull's-eye model is lower, and the scheduling parameter value of the indicator closer to the optimal selection area is higher; Determine the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator; The scheduling parameter value of each indicator is determined by using an indicator model according to the distance of each indicator from the center of the target; wherein the indicator model is a model for determining the indicator parameter according to the distance of the indicator from the center of the target, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area; the indicator model is ;in, Indicates the scheduling parameter value of each indicator, Indicates the distance of each indicator from the center of the target. Indicates the radius of the maximum selection area. represents the radius of the optimal selection area, Indicates the indicator angle; The comprehensive scheduling parameter value of the host machine is determined according to the scheduling parameter value of each indicator corresponding to the host machine, and the target scheduling host machine is determined from multiple host machines according to the comprehensive scheduling parameter value of the host machine.

2. The host scheduling method according to claim 1, characterized in that: The indicators include numerical indicators and non-numerical indicators; the non-numerical indicators are indicators that cannot be measured by numerical values.

3. The host scheduling method according to claim 2, characterized in that: The numerical type indicators include at least one of central processing unit utilization, memory utilization, storage utilization, number of cloud hosts, number of network inputs and outputs, and number of processes. The non-numerical type indicators include whether the central processing unit architecture is consistent with the virtual machine.

4. The host scheduling method according to claim 1, characterized in that: The extreme value parameters are the maximum value and the minimum value, or the extreme value parameters are the optimal value and the worst value.

5. The host scheduling method according to claim 4, characterized in that: Determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator includes: The distance between each indicator and the bull's eye is determined by using the bull's eye distance formula according to the extreme value parameter and the indicator value corresponding to each indicator; wherein the bull's eye distance formula is: ; wherein said Indicates the indicator value, Indicates the maximum value, Indicates the minimum value, represents the optimal value, Indicates the worst value, represents the radius of the optimal selection area, and R represents the radius of the circular bull's eye model.

6. The host scheduling method according to any one of claims 1 to 5, characterized in that: The indicator model is used to determine the scheduling parameter value of each indicator based on the distance of each indicator from the bull's eye, including: When the distance between the indicator and the center of the target is greater than the radius of the maximum selection area, the indicator model is used to determine the scheduling parameter value of the indicator as a preset minimum value; When the distance between the indicator and the bull's eye is greater than or equal to the radius of the optimal selection area and less than the radius of the maximum selection area, the indicator model is used to determine the scheduling parameter value of the indicator, which is the area of ​​the sector area of ​​the maximum selection area minus the area of ​​the sector area where the indicator value is located; When the distance between the indicator and the center of the target is less than the radius of the optimal selection area, the indicator model is used to determine the scheduling parameter value of the indicator as the area of ​​the sector region where the indicator is located.

7. The host scheduling method according to claim 1, characterized in that: Determining a comprehensive scheduling parameter value of the host machine according to the scheduling parameter value of each indicator corresponding to the host machine, and determining a target scheduling host machine from multiple host machines according to the comprehensive scheduling parameter value of the host machine, including: Performing a sum operation on the scheduling parameter value of each indicator corresponding to each of the host machines to determine the comprehensive scheduling parameter value of each of the host machines; A maximum comprehensive scheduling parameter value is selected from the comprehensive scheduling parameter values ​​of each host machine, and a host machine corresponding to the maximum comprehensive scheduling parameter value is used as the target scheduling host machine.

8. The host scheduling method according to claim 1, characterized in that: Determining a comprehensive scheduling parameter value of the host machine according to the scheduling parameter value of each indicator corresponding to the host machine, and determining a target scheduling host machine from multiple host machines according to the comprehensive scheduling parameter value of the host machine, including: When the multiple indicators do not include resource utilization parameters, the target scheduling host is determined according to the resource utilization and the comprehensive scheduling parameter value, so that the resource utilization of the target scheduling host is greater than a preset minimum threshold and less than a preset maximum threshold.

9. The host scheduling method according to claim 8, characterized in that: When the resource utilization parameter is not included in the multiple indicators, determining the target scheduling host according to the resource utilization and the comprehensive scheduling parameter value includes: Determine the host machine corresponding to the maximum comprehensive scheduling parameter value, and when the resource utilization of the host machine is greater than the preset maximum threshold, or when the resource utilization of the host machine is less than the preset minimum threshold, determine to reselect the host machine as the target scheduling host machine.

10. The host scheduling method according to claim 1, characterized in that: After determining the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator, the method further includes: Determine whether the distance between the current indicator and the center of the target is greater than the radius of the maximum selection area; When it is determined that the distance between the current indicator and the center of the target is greater than the radius of the maximum selection area, determine to filter the host machines corresponding to the indicators greater than the radius of the maximum selection area to obtain the host machines that are not filtered out; Accordingly, determining the comprehensive scheduling parameter value of the host machine according to the scheduling parameter value of each indicator corresponding to the host machine includes: The comprehensive scheduling parameter value of the host machine that has not been filtered out is determined according to the scheduling parameter value of each indicator corresponding to the host machine that has not been filtered out.

11. A host scheduling device, characterized in that: include: A circular bull's-eye model determination module is used to establish a circular bull's-eye model according to the weights of multiple indicators; wherein the circular bull's-eye model is a model that uses the angle of a sector to determine the weight of each indicator, and includes a maximum selection area and an optimal selection area, wherein the maximum selection area and the optimal selection area are two circular rings with different radii set according to requirements, and the radius of the maximum selection area is greater than the radius of the optimal selection area; the outermost circle of the circular bull's-eye model is the entire circular bull's-eye model, and the entire circular bull's-eye model includes two circular rings, the outer one is the maximum selection area, and the host machine with indicators exceeding the maximum selection area will not be scheduled; the inner one is the optimal selection area, and the scheduling parameter value of the indicator closer to the center of the circular bull's-eye model is lower, and the scheduling parameter value of the indicator closer to the optimal selection area is higher; A distance determination module, used to determine the distance of each indicator from the bull's eye of the circular bull's eye model according to the extreme value parameter corresponding to each indicator and the indicator value corresponding to each indicator; The scheduling parameter value determination module is used to determine the scheduling parameter value of each indicator using an indicator model according to the distance of each indicator from the center of the target; wherein the indicator model is a model for determining the indicator parameter according to the distance of the indicator from the center of the target, the indicator angle, the radius corresponding to the maximum selection area, and the radius corresponding to the optimal selection area; the indicator model is ;in, Indicates the scheduling parameter value of each indicator, Indicates the distance of each indicator from the center of the target. Indicates the radius of the maximum selection area. represents the radius of the optimal selection area, Indicates the indicator angle; The target scheduling host determination module is used to determine the comprehensive scheduling parameter value of the host according to the scheduling parameter value of each indicator corresponding to the host, and determine the target scheduling host from multiple hosts according to the comprehensive scheduling parameter value of the host.

12. A host scheduling device, characterized in that: include: Memory for storing computer programs; A processor, configured to execute the computer program to implement the steps of the host scheduling method as claimed in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the host scheduling method according to any one of claims 1 to 10 are implemented.

14. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the host scheduling method described in any one of claims 1 to 10 are implemented.

Images