Resource management method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a resource management method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining resource data for a target server type, performing resource use estimation based on historical resource data, computing resource supply data, current resource data and a current inventory resource total amount, and obtaining a target reference resource amount corresponding to the target server type at a current time, based on the target reference resource quantity and the current inventory resource total quantity, generating resource scheduling information, so that the inventory calculation resource quantity corresponding to the target server type is smaller than the target reference resource quantity after resource scheduling is carried out on resources under the target server type according to the resource scheduling information, the difference value between the target reference resource quantity corresponding to the target server type at the current time and the current actual use resource total quantity is within a preset threshold value range. By adopting the method, inventory resources can be automatically adjusted, and the resource utilization rate is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more specifically, to a resource management method, device, electronic device and storage medium. Background Art

[0002] With the continuous development of computer technology, cloud servers (CVM Cloud Virtual Machine) can provide secure and reliable elastic computing services. In order to meet customers' elastic expansion and contraction needs, service providers need to deploy and operate a large number of cloud server resources on the cloud. This often leads to two problems: on the one hand, service providers may purchase a large number of physical servers during the deployment phase, resulting in resource redundancy and energy waste; on the other hand, service providers may purchase insufficient physical server resources during the deployment phase, affecting user purchasing experience. Therefore, providing a reasonable computing resource management method is a technical problem that urgently needs to be solved. Summary of the invention

[0003] In view of this, the embodiments of the present application propose a resource management method, device, electronic device and storage medium, which can automatically adjust inventory resources to avoid the problem of excessive or insufficient resources in the current inventory resources of the target server type.

[0004] In a first aspect, an embodiment of the present application provides a resource management method, the method comprising: obtaining resource data for a target server type, the resource data comprising historical resource data, computing resource supply data, and current resource data corresponding to a current time, the current resource data comprising a current total inventory of resources and a current total actual usage of resources; performing resource usage estimation based on the historical resource data, computing resource supply data, and current resource data to obtain a target reference resource amount corresponding to the target server type at the current time; generating resource scheduling information based on the target reference resource amount and the current total inventory of resources, so that after resource scheduling is performed on the resources under the target server type according to the resource scheduling information, the inventory of computing resources corresponding to the target server type is less than the target reference resource amount, and the difference between the target reference resource amount corresponding to the target server type at the current time and the current total actual usage of resources is within a preset threshold range.

[0005] In the second aspect, the embodiment of the present application provides a resource management device, which includes: a data acquisition module, a resource quantity estimation module and a scheduling information generation module. The data acquisition module acquires resource data for the target server type, the resource data includes historical resource data, computing resource supply data and current resource data corresponding to the current time, and the current resource data includes the current total inventory resources and the current total actual use of resources; the resource quantity estimation module is used to estimate resource usage based on the historical resource data, computing resource supply data and current resource data, and obtain the target reference resource quantity corresponding to the target server type at the current time; the scheduling information generation module is used to generate resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity, so that after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource quantity corresponding to the target server type is less than the target reference resource quantity, and the difference between the target reference resource quantity corresponding to the target server type at the current time and the current total actual use of resources is within a preset threshold range.

[0006] In one possible implementation, the historical resource data includes historical actual resource usage, historical purchased resource usage, and historical reference resource usage at multiple time points within a historical time period; the resource estimation module includes a resource estimation submodule and a resource adjustment submodule. The resource estimation submodule is used to estimate based on historical purchased resource usage, target historical reference resource usage, and current resource data at multiple time points within a first preset duration to obtain an initial reference resource usage; the resource adjustment submodule is used to adjust the initial reference resource usage to obtain a target reference resource usage based on historical actual resource usage and the computing resource supply data within a second preset duration, wherein the first preset duration is greater than the second preset duration.

[0007] In one possible implementation, the resource quantity estimation submodule is also used to determine the total resource purchase quantity within the first preset time period based on the historical resource purchase quantities at multiple time points within the first preset time period; and determine the initial reference resource quantity based on the total resource purchase quantity within the first preset time period, the target historical reference resource quantity and the current total inventory resource quantity.

[0008] In one possible implementation, the resource quantity estimation submodule is further used to perform a weighted summation of the total resource purchase quantity and the target historical reference resource quantity to obtain a weighted result; and subtract the weighted result from the current total inventory resource quantity to obtain an initial reference resource quantity.

[0009] In one possible implementation, the resource quantity adjustment submodule is also used to obtain the target historical resource usage based on the historical actual resource usage within a second preset time period; add the supply resource quantity corresponding to the current time in the computing resource supply data to the target historical resource usage to obtain an intermediate reference resource quantity; determine the redundant resource ratio based on the initial reference resource quantity and the intermediate reference resource quantity; adjust the initial reference resource quantity based on the redundant resource ratio to obtain the target reference resource quantity.

[0010] In one possible implementation, the resource amount adjustment submodule is further used to determine a resource adjustment strategy corresponding to the redundant resource ratio; and adjust the initial reference resource amount according to the resource adjustment strategy corresponding to the redundant resource ratio to obtain the target reference resource amount.

[0011] In one possible implementation, the scheduling information generation module is also used to determine the target physical server and resource scheduling operation information for the target physical server from the physical servers belonging to each cloud server of the target server type when the difference between the target reference resource amount and the current actual total amount of resources in use is greater than the maximum value of a preset threshold range, based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource amount and deployment information of each cloud server belonging to the target server type, wherein the resource scheduling operation information includes at least one of online power saving operation information and balanced relocation operation information for the target physical service.

[0012] In one possible implementation, the scheduling information generation module includes a resource statistics submodule and a scheduling information generation submodule. The resource statistics submodule is used to count the total resource amount of physical servers of the target server type in the online power saving state when the difference between the target reference resource amount and the current actual total amount of used resources is less than the minimum value of the preset threshold range; the scheduling information generation submodule is used to determine the target physical server from the physical servers in the online power saving state based on the difference and the deployment information of the cloud server in the online power saving state when the total amount of resources is not less than the difference between the target reference resource amount and the current actual total amount of used resources, and generate resource scheduling information including the deployment information of the target physical server and the power-on operation of the target physical server.

[0013] In one possible implementation, the scheduling information generation submodule is also used to determine the amount of resources to be purchased based on the difference and the total amount of resources when the total amount of resources of the physical server in the online power saving state is less than the difference between the current inventory resource amount and the target reference resource amount, and to generate resource scheduling information including the amount of resources to be purchased, the deployment information of the physical server in the online power saving state, and the power-on operation of the physical server.

[0014] In one possible implementation, the scheduling information generation module is also used to determine a designated physical server and the maneuvering and regularization operation information performed on the designated physical server from the physical servers to which each of the cloud servers belongs based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity of each of the cloud servers, the current actual resource usage quantity and the deployment information, and generate resource scheduling information including the deployment information and maneuvering and regularization operation information of the designated physical server.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory; one or more programs are stored in the memory and configured to be executed by the processor to implement the above method.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a program code is stored, wherein the above method is executed when the program code is executed by a processor.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product or a computer program, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device obtains the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the above method.

[0018] The embodiment of the present application provides a resource management method, device, electronic device and storage medium, which estimates resource usage based on the historical resource data, computing resource supply data and current resource data, and obtains the target reference resource amount corresponding to the target server type at the current time, thereby realizing real-time analysis and elastic prediction of business demand, solving the problem that traditional resource shortage and procurement redundancy are both afterthoughts. In addition, by adopting the above estimation method, it can be effectively guaranteed that the target reference resource amount obtained by the current estimate matches the actual operation of the cloud server of the current target server type. Further, by generating resource scheduling information based on the target reference resource amount and the current total inventory resource amount, the inventory computing resource amount corresponding to the target server type is less than the target reference resource amount after the resource scheduling is performed on the resources under the target server type according to the resource scheduling information, and the difference between the target reference resource amount corresponding to the target server type at the current time and the current actual total resource usage is within the preset threshold range. It avoids the problem of excessive and insufficient resources that may exist in the current inventory resources of the target server type, and improves the utilization rate of the resources of the target server type. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 An application scenario diagram of a resource management method provided by an embodiment of the present application is shown;

[0021] Figure 2 A schematic diagram of a resource management method according to an embodiment of the present application is shown;

[0022] Figure 3 Another schematic diagram of a resource management method proposed in an embodiment of the present application is shown;

[0023] Figure 4 Another schematic diagram of a resource management method provided in an embodiment of the present application is shown;

[0024] Figure 5 A block diagram of a resource management system provided by an embodiment of the present application is shown;

[0025] Figure 6 A hierarchical schematic diagram of a resource intelligent planning engine provided by an embodiment of the present application is shown;

[0026] Figure 7 A flowchart of resource management by a resource intelligent management system provided by an embodiment of the present application is shown;

[0027] Figure 8 A schematic diagram of a comparison process of a water level comparator provided in an embodiment of the present application is shown;

[0028] Fig. 9 A schematic diagram showing a resource scheduler provided in an embodiment of the present application that performs resource scheduling based on a comparison result;

[0029] Fig.10 A schematic diagram showing changes in a target reference resource amount and an actual total amount of used resources over time provided in an embodiment of the present application is shown;

[0030] Fig.11 A supply chain diagram of a cloud server provided in an embodiment of the present application is shown;

[0031] Fig.12 A connection block diagram of a resource management device provided in an embodiment of the present application is shown;

[0032] Fig.13A structural block diagram of an electronic device for executing the method of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0033] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0034] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0035] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0036] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0037] It should be noted that the “plurality” mentioned in this article refers to two or more.

[0038] Before going into detail, the terms used in this application are explained as follows:

[0039] Cloud technology refers to a hosting technology that unifies hardware, software, network and other resources within a wide area network or local area network to achieve data computing, storage, processing and sharing. Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model. It can form a resource pool that can be used on demand and is flexible and convenient. In this plan, the main consideration is to use cloud technology to deploy cloud servers in a server cluster composed of multiple physical servers and provide cloud servers to cloud server buyers.

[0040] Physical server, the functions of a physical server are very similar to those of an ordinary computer, but on a different scale. It is a hardware server that includes memory, a hard drive, a network connection, and runs an operating system and applications outside of its internal hardware resources. A physical server can be assigned to a single user (standalone server) or used by multiple users (for example, through virtualization or other technologies that establish access management and resource usage types). In an embodiment of the present application, there are multiple physical servers, and multiple physical servers constitute an integrated server, and one or more cloud servers can be deployed on each physical server.

[0041] Cloud server is a virtualization technology similar to VPS server (Virtual Private Server). + VPS uses virtual software, VZ (Virtuozzo, virtualized data center) or VM (VMware, virtual machine) to virtualize multiple parts similar to independent servers on a server. Each part can have a separate operating system and the management method is the same as that of a physical server. In other words, cloud server virtualizes multiple parts similar to independent servers on a group of server clusters. Each server in the cluster has a mirror of the cloud server, which greatly improves the security and stability of the virtual server. Unless all the servers in the cluster have problems, the cloud server will not be unusable.

[0042] The following describes an exemplary application of a device for executing the resource management method provided by an embodiment of the present invention.

[0043] See also Figure 1 , which shows a schematic diagram of an application scenario of a resource management method provided by an embodiment of the present invention, where the application scenario includes an electronic device 10 and a server cluster 20.

[0044] The server cluster 20 is a server infrastructure that uses multiple physical servers (generally referring to two or more servers working simultaneously) to provide users with access to various services and applications. The main advantage of using the server cluster 20 is that it can provide a higher level of reliability and availability than a single server environment. In this embodiment, the server cluster 20 includes multiple physical servers, each of which is used to virtualize multiple cloud servers similar to independent servers.

[0045] The electronic device 10 can be a terminal device, such as a smart device, a desktop computer, a tablet computer, a laptop computer, etc., which can communicate with the above-mentioned server cluster 20. The device can be used to obtain resource data for the target server type in the server cluster 20, and the resource data includes historical resource data, computing resource supply data, and the current total inventory resources and the current total actual resource usage at the current time. Then, based on the historical resource data, computing resource supply data and current resource data, resource usage estimation is performed to obtain the target reference resource amount corresponding to the target server type at the current time; and based on the target reference resource amount and the current total inventory resource amount, resource scheduling information is generated, and the resource scheduling information is sent to the server cluster 20, so that after the server cluster performs resource scheduling on the resources under the target server type according to the resource scheduling information, the inventory computing resource amount corresponding to the target server type is less than the target reference resource amount, and the difference between the target reference resource amount corresponding to the target server type at the current time and the current total actual resource usage is within a preset threshold range.

[0046] By adopting the above method, real-time analysis and flexible prediction of business demand are achieved, the traditional problem of insufficient resources and procurement redundancy being realized afterward is solved, and the problem of excessive and insufficient resources in the current inventory resources of the target server type is avoided, thereby improving the utilization rate of the resources of the target server type.

[0047] See also Figure 2 As shown, an embodiment of the present application provides a resource management method, which can be applied to the above-mentioned server, that is, the above-mentioned resource management method can be executed by a background service running on the above-mentioned server, and the method includes:

[0048] Step S110: Acquire resource data for the target server type.

[0049] Among them, resource data includes historical resource data, computing resource supply data and current resource data corresponding to the current time. The current resource data includes the current total inventory resources and the current total actual use of resources.

[0050] The target server type can be any cloud server type including high IO type, memory type, computing type, GPU (graphics processing unit, graphics processor) model, FPGA (Field Programmable Gate Array, programmable array logic) model, NPU (Natural Processing Unit, embedded neural network processor) model and offline computing type. It should be noted that if there are multiple cloud servers deployed on the same physical server, the types of the multiple cloud servers are the same. If the target server type corresponds to multiple cloud servers, the multiple cloud servers are deployed by at least one physical server, and each physical server corresponds to at least one cloud server, the models of the multiple physical servers can be different.

[0051] Historical resource data refers to data generated and recorded by the server cluster before the current time, and the historical resource data may include one or more of the historical purchased resource quantities, historical reference resource quantities, historical actual used resource quantities, and historical inventory resource quantities at multiple time points.

[0052] Computing resource supply data refers to that for some customers who rent or purchase cloud servers, the amount of resources corresponding to the cloud servers they rent or purchase is large. Therefore, they need to pre-order from the cloud server supplier so that the cloud server supplier can generate computing resource supply data based on the resource information of the pre-ordered cloud servers. The computing resource supply data in the embodiment of the present application may specifically include at least one of the supply location of the computing resources, the server type corresponding to the computing resources, the supply time of the computing resources, and the resource supply corresponding to each supply time. It should be understood that the server type corresponding to the computing resource is the target server type mentioned above. The supply location of the computing resources specifically refers to the cloud server required by the customer being specifically deployed in a certain city, a certain urban area of ​​a certain city, a certain computer room in a certain city, or a specific rack in a certain computer room in a certain city.

[0053] The current total inventory resources corresponding to the current time refer to the sum of the inventory resources of the cloud servers mounted on each physical server in the server cluster that is in online power-on mode at the current time. The current total actual used resources corresponding to the current time refer to the sum of the actual used resources of the cloud servers mounted on each physical server in the server cluster that is in online power-on mode at the current time.

[0054] Among them, the current total amount of inventory resources at the current time can be the total amount of inventory resources corresponding to the starting time of the time period within the current time, or it can be the total amount of inventory resources at the current time. Similarly, the current total amount of actually used resources at the current time can be the current total amount of actually used resources corresponding to the starting time of the time period within the current time, or it can be the current total amount of actually used resources at the current time.

[0055] It should be noted that the starting time of the time period in which the current time is located can be zero o'clock on the day in which the current time is located, or zero minutes in the hour in which the current time is located, or zero o'clock on Monday in the week in which the current time is located. The above-mentioned time period can be determined according to the cycle of resource management. If the cycle of resource management is executed every other day, then the above-mentioned time period is also one day. Accordingly, the starting time of the time period in which the current time is located can be zero o'clock on the day in which the current time is located. If the cycle of resource management is executed every other week, then the above-mentioned time period is also one week. Accordingly, the starting time of the time period in which the current time is located can be zero o'clock on Monday in the week in which the current time is located.

[0056] The resource data for the target server type may be obtained by obtaining the resource data for the target server type from a data warehouse corresponding to the server cluster.

[0057] Specifically, the type identifier corresponding to the target server type can be used to obtain resource data corresponding to the type identifier from a database or storage device associated with the server cluster. Resource data for the target server type can also be obtained from a database or storage device associated with the server cluster using data warehouse technology.

[0058] It should be noted that data warehouse technology (Extract-Transform-Load, ETL) is used to describe the process of extracting, transforming, and loading data from the source end to the destination end. The purpose of data warehouse technology is to integrate the scattered, messy, and non-standard data in the enterprise to provide an analytical basis for the enterprise's decision-making. In other words, ETL is an important part of the BI (business intelligence) project. That is, the above step S110 can specifically be to use ETL technology to extract and transform the data corresponding to the target server type in the database or storage device associated with the server cluster and load it into the electronic device to obtain the resource data of the target server type.

[0059] Step S120: perform resource usage estimation based on historical resource data, computing resource supply data, and current resource data to obtain a target reference resource amount corresponding to the target server type at the current time.

[0060] By obtaining the target reference resource quantity so that it can be used for resource scheduling later, it can be ensured that the inventory resources after scheduling will not have insufficient supply or large redundancy problems, thereby avoiding the traditional problems of insufficient resources and procurement redundancy which are realized belatedly.

[0061] In one possible implementation, the historical resource data may include historical actual resource usage and historical reference resource usage. The above step S120 may be based on the historical actual resource usage, historical reference resource usage, computing resource supply data and current actual resource usage at multiple time points within a preset duration to estimate and obtain the target reference resource usage. The above preset duration may be one week, one month or two months, etc. Among them, the historical actual resource usage at a certain point in time refers to the sum of the actual resource usage of the cloud servers mounted on each physical server in the server cluster that is in online power-on mode at that time. It should be understood that the historical reference resource usage at a certain point in time refers to the reference resource usage at a certain point in time obtained in a manner similar to the above step S120.

[0062] Exemplarily, in this implementation mode, the historical average of redundant resources can be obtained based on the historical actual resource usage and historical reference resource usage at multiple time points. Based on the supply resource amount in the computing resource supply data that belongs to the same time period as the current moment, the current actual resource usage and the historical average of redundant resources are superimposed to obtain the target reference resource amount corresponding to the current moment. Among them, belonging to the same time period as the current moment means a certain time period of the same day as the current moment, or the same day as the current moment data, or the same week as the current moment, etc.

[0063] In another possible implementation, the historical resource data includes historical actual resource usage, historical reference resource usage, and historical purchased resource usage at multiple time points. The above step S120 can also be estimated based on the historical actual resource usage, historical reference resource usage, and the current actual total resource usage at multiple time points within a preset duration to obtain an initial reference resource amount, and the initial reference resource amount is adjusted based on the historical purchased resource amounts at multiple time points within a specified time period and the target resource supply in the calculated resource supply data to obtain the target reference resource amount. The preset duration can be one week, one month, or two months, etc. The specified time period can be a time period with the same attributes as the current moment. For example, if the current moment is a holiday, the specified time period can also be a holiday. For another example, if the current time is a weekday, the specified time period is also a weekday.

[0064] The above-mentioned historical purchased resource volume refers to the resource volume provided by the purchased physical server when the resource data provided by the cloud server deployed in the server cluster at the time point corresponding to the historical resource purchase volume did not reach the historical reference resource volume corresponding to the time point. That is, the purchased physical server is used to deploy an instruction server type cloud server so that the deployed cloud server can provide the historical purchased resource volume.

[0065] Exemplarily, in this implementation, the historical redundant resource amount average can be obtained based on the historical actual resource usage and historical reference resource amounts at multiple time points within a preset time period, and the redundant resource amount average is added to the actual total resource usage to obtain the initial reference resource amount. The resource purchase average is obtained based on the historical resource purchase amount at multiple time points within a specified time period, and the resource purchase average is added to the initial reference resource amount and the computing resource supply amount in the preset market from the computing resource supply data at the current moment to obtain the target reference resource amount.

[0066] In another possible implementation, the historical resource data includes historical actual resource usage and historical reference resource usage at multiple time points. The above step S120 can also be based on the historical resource usage, historical reference resource usage and current actual total resource usage at multiple time points within a first preset duration to obtain an initial reference resource amount; based on the historical actual resource usage and computing resource supply data within a second preset duration, the initial reference resource amount is adjusted to obtain a target reference resource amount, and the first preset duration is greater than the second preset duration.

[0067] It should be understood that the above method of determining the target reference resource quantity at the current moment based on historical resource data, calculated resource supply data and the current actual total amount of resources used is merely illustrative, and this scheme may also have more confirmation methods, which will not be elaborated here one by one.

[0068] Step S130: Generate resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity, so that after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource quantity corresponding to the target server type is less than the target reference resource quantity, and the difference between the target reference resource quantity corresponding to the target server type at the current time and the current actual total resource usage is within a preset threshold range.

[0069] By ensuring that the difference between the target reference resource amount and the current actual resource usage is within a preset threshold range, the current actual total amount of resource usage after resource scheduling can be made within a safe range, which can facilitate resource scheduling and ensure that the inventory resources after resource scheduling can supply new resource demands in a timely manner without a large amount of resource redundancy.

[0070] It should be understood that if the difference between the target reference resource amount and the current actual total amount of resources in use is greater than the maximum value of the preset threshold range, it means that the total amount of inventory resources is too much and the inventory resources need to be reduced. At this time, resource scheduling information can be generated, including balanced scheduling of a specified number of inventory resources and online power saving of physical servers in the server cluster that do not provide resources to customers after balanced scheduling.

[0071] When the difference between the safety resource reference amount and the current actual total amount of resources in use is less than the minimum value of the preset threshold range, it means that the total amount of inventory resources is small and the inventory resources need to be increased. At this time, resource scheduling information can be generated, including executing online power-on for physical servers in the server cluster that are in online power saving, and / or purchasing physical servers and deploying cloud servers of specified physical server types to the purchased physical servers.

[0072] Considering the convenience and operability when adjusting resources, the above-mentioned preset threshold range may include a first preset threshold (that is, the first preset threshold is the maximum value in the preset threshold range) and a second preset threshold (that is, the second preset threshold is the minimum value in the preset threshold range), and the difference between the first preset threshold and the second preset threshold may be set according to the maximum amount of resources that a physical server in the server cluster can provide when deploying a cloud server of the target server type. Exemplarily, the difference between the above-mentioned first threshold and the second threshold may be a specified multiple of the maximum amount of resources that a physical server in the server cluster can provide when deploying a cloud server of the target server type.

[0073] The present application provides a resource management method, which estimates resource usage based on historical resource data, computing resource supply data and the current actual total amount of resources used for the resource data of the target server type, and obtains the target reference resource amount corresponding to the target server type at the current time, thereby realizing real-time analysis and elastic prediction of business demand conditions, and solving the problem that traditional resource shortage and procurement redundancy are both hindsight problems. In addition, by adopting the above estimation method, it can effectively ensure that the target reference resource amount obtained by the current estimate matches the actual operation of the cloud server of the current target server type. After obtaining the target reference resource amount, based on the target reference resource amount and the current total amount of inventory resources, resource scheduling information is generated, so that the server cluster can perform resource scheduling according to the above resource scheduling information, so that the difference between the target reference resource amount corresponding to the server provider at each time point and the current actual total amount of resources used at each time point is kept within a reasonable range, that is, even if the inventory computing resource amount corresponding to the target server type is less than the target reference resource amount after the resources under the target server type are scheduled according to the resource scheduling information, and the difference between the target reference resource amount corresponding to the target server type at the current time and the current actual total amount of resources used is within the preset threshold range. This avoids the problem of excessive or insufficient resources in the current inventory resources of the target server type, thereby improving the utilization rate of the resources of the target server type.

[0074] See also Figure 3 As shown, a resource management method provided by another embodiment of the present application includes:

[0075] Step S210: Acquire resource data for the target server type.

[0076] Resource data includes historical resource data, computing resource supply data, and current resource data corresponding to the current time. Current resource data includes the current total inventory resources and the current total actual usage resources.

[0077] Step S220: An initial reference resource quantity is obtained by estimating based on the historical purchased resource quantity at multiple time points within the first preset time period, the target historical reference resource quantity, and the current resource data.

[0078] Among them, the above-mentioned first preset time period can be one week, two weeks, one month or two months, and can be set according to actual needs.

[0079] In one possible implementation, the above step S220 may be to obtain an average of the historical purchased resource quantities based on the historical purchased resource quantities at multiple time points, and determine a target historical reference resource quantity based on the historical reference resource quantities at multiple time points, and the target historical reference resource quantity may be the resource quantity average of the historical reference resource quantities at multiple time points, or may be the maximum historical reference resource quantity among the historical reference resource quantities at multiple time points. The total amount of actually used resources in the current resource data is added or weighted to the average of the historical purchased resource quantities and the target historical reference resource quantity to obtain the initial reference resource quantity.

[0080] In another possible implementation, the above-mentioned step S220 may also be to determine the total resource purchase quantity within the first preset time period based on the historical resource purchase quantity at multiple time points within the first preset time period; and determine the initial reference resource quantity according to the total resource purchase quantity within the first preset time period, the target historical reference resource quantity and the current total inventory resource quantity.

[0081] In this implementation, an optimistic estimation algorithm formula with purchase success priority can be used to determine the initial reference resource quantity based on the total resource purchase quantity within the first preset time period, the target historical reference resource quantity, and the current total inventory resource quantity. The initial reference resource quantity calculated based on the optimistic estimation formula can effectively ensure that the calculated initial reference resource quantity can meet the customer's resource purchase needs.

[0082] Specifically, the initial reference resource quantity can be determined based on the total resource purchase quantity within the first preset time period, the target historical reference resource quantity and the current total inventory resource quantity. Specifically, the method can be: weighted summing the total resource purchase quantity and the target historical reference resource quantity to obtain a weighted result; subtracting the weighted result from the current total inventory resource quantity to obtain the initial reference resource quantity.

[0083] Exemplarily, the weight coefficient of the total resource purchase amount within the first preset time period can be set to 1.5, and the weight coefficient of the target historical reference resource amount can be set to 1, and the total resource purchase amount within the first preset time period and the target historical reference resource amount are weighted and summed according to the set weight coefficients to obtain a weighted result; the weighted result is subtracted from the current total inventory of resources to obtain an initial reference resource amount.

[0084] It should be understood that the above-mentioned method of obtaining the initial reference resource amount is merely illustrative, and there may be more methods of obtaining it, which will not be described one by one in the embodiments of the present application.

[0085] Step S230: Based on the historical actual resource usage and computing resource supply data within the second preset time period, the initial reference resource amount is adjusted to obtain the target reference resource amount.

[0086] Among them, the first preset time length is greater than the second preset time length.

[0087] Since the initial reference resource quantity obtained by using the historical resource data within the first preset time period reflects the supply of reference resources over a longer period of time, it may not be consistent with the reference resources required at the current time. Therefore, in order to make the estimated reference resource quantity more accurate, the initial reference resource quantity can be adjusted using the resource usage situation in a shorter period of time that is closer to the current time to obtain a more accurate target reference resource quantity.

[0088] The second preset duration can be 2 days, 3 days, 5 days or a week, etc., and can be set according to actual needs. Among them, the end point of the first preset duration and the second preset duration can be the current time, or the start time of the time period corresponding to the current time. For example, before midnight of the day when the current time is located.

[0089] In one possible implementation, the above step S230 may be, based on the actual use resources corresponding to multiple time points within the second preset time length, the average historical actual use resources are obtained, and an intermediate reference resource amount is obtained by superimposing the supply resource amount in the calculated resource supply data that belongs to the same time period as the current moment and the average historical actual use resources. Among them, belonging to the same time period as the current moment means a certain time period of the same day as the current moment, or the same day as the current moment data, or the same week as the current moment, etc. The intermediate reference resource amount and the initial reference resource amount are weighted and summed to obtain the target reference resource amount.

[0090] In another possible implementation, the above step S220 can also be, based on the historical actual use resources corresponding to multiple time points within the second preset time length, the target historical actual use resources are obtained, and the supply resource amount in the computing resource supply data belonging to the same time period as the current moment is superimposed with the target historical actual use resources to obtain an intermediate reference resource amount. Among them, belonging to the same time period as the current moment refers to a certain time period belonging to the same day as the current moment, or the same day as the current moment data, or it can also be the same week as the current moment, etc. According to the initial reference resource amount and the intermediate reference resource amount, the redundant resource ratio is determined, and the initial reference resource amount is adjusted according to the redundant resource ratio to obtain the target reference resource amount.

[0091] The difference between the initial reference resource amount and the intermediate reference resource amount is compared with the intermediate reference resource amount or the initial reference resource amount to obtain a redundant resource ratio. The redundant resource ratio reflects the ratio of the difference between the initial reference resource amount and the intermediate reference resource amount to the initial reference resource amount or the intermediate reference resource amount, and the larger the value of the redundant resource ratio is, the larger the difference between the initial reference resource amount and the intermediate reference resource amount is. The method of adjusting the initial reference resource amount to obtain the target reference resource amount based on the redundant resource ratio can be specifically to determine the resource adjustment strategy corresponding to the redundant resource ratio, and adjust the initial reference resource amount according to the resource adjustment strategy to obtain the target reference resource amount.

[0092] Different redundant resource ratios may correspond to different adjustment strategies. For example, if the redundant resource ratio is between 20% and 30%, the adjustment strategy is to reduce the initial reference resource amount by 10% to 20%; if the redundant resource ratio is between 10% and 20%, the adjustment strategy is to reduce the initial reference resource amount by 0% to 10%.

[0093] Step S240: Generate resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity, so that after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource quantity corresponding to the target server type is less than the target reference resource quantity, and the difference between the target reference resource quantity corresponding to the target server type at the current time and the current actual total resource usage is within a preset threshold range.

[0094] For the detailed description of the above step S240, please refer to the detailed description of step S130 above, which will not be described in detail in this embodiment.

[0095] A resource scheduling method provided in an embodiment of the present application obtains an initial reference resource amount by making an estimate based on historical purchased resource amounts, target historical reference resource amounts, and current resource data at multiple time points within a first preset time period, and adjusts the initial reference resource amount based on historical actual resource usage and computing resource supply data within a second preset time period to obtain a target reference resource amount. The first preset time period is greater than the second preset time period, thereby achieving real-time analysis and elastic prediction of business demand based on resource usage in the past period and resource supply demand of major customers, and solving the problem of traditional resource shortage and procurement redundancy being realized afterward. In addition, by adopting the above-mentioned method of estimating the target reference resource amount in steps, and the first step is to estimate the initial reference resource amount based on multiple historical resource data within a long time period, and the second step is to adjust the aforementioned initial reference resource amount based on the historical resource usage in a time period close to the current time to obtain the target reference resource amount, it can effectively ensure that the estimated target reference resource amount matches the actual operating conditions of the cloud server of the current target server type.

[0096] Furthermore, by generating resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity, after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource quantity corresponding to the target server type is less than the target reference resource quantity, and the difference between the target reference resource quantity corresponding to the target server type at the current time and the current actual total resource quantity is within a preset threshold range. This avoids the problem of excessive or insufficient resources that may exist in the current inventory resources of the target server type, and improves the utilization rate of the resources of the target server type.

[0097] See also Figure 4 The embodiment of the present application also provides a resource scheduling method, the method comprising:

[0098] Step S310: Acquire resource data for the target server type.

[0099] Resource data includes historical resource data, computing resource supply data, and current resource data corresponding to the current time. Current resource data includes the current total inventory resources and the current total actual usage resources.

[0100] Step S320: Estimate resource usage based on historical resource data, computing resource supply data, and current resource data to obtain a target reference resource amount corresponding to the target server type at the current time.

[0101] Step S330: Compare the difference between the target reference resource amount and the current actual total amount of used resources with a preset threshold range.

[0102] Among them, the above-mentioned comparison of the difference with the preset threshold range can specifically be to compare the difference with the maximum value and the minimum value of the preset threshold range respectively, and the possible comparison results include that the difference is within the preset threshold range, the difference is greater than the maximum value of the preset threshold range, or the difference is less than the minimum value of the preset threshold range.

[0103] If the difference between the target reference resource quantity and the current actual total amount of resources used is greater than the maximum value of the preset threshold range, execute step S340: based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity and deployment information of each cloud server belonging to the target server type, determine the target physical server and the resource scheduling operation information for the target physical server from the physical servers belonging to each cloud server of the target server type.

[0104] The resource scheduling operation information includes at least one of online power saving operation information and balanced migration operation information for the target physical service.

[0105] It should be understood that if the difference between the target reference resource quantity and the current actual total resource usage is greater than the maximum value of the preset threshold range, it means that the current inventory resource quantity is too much. At this time, the current inventory resource quantity needs to be reduced to improve resource utilization.

[0106] Among them, based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity and deployment information of each cloud server belonging to the target server type, the method of determining the target physical server and the resource scheduling operation information of the target physical server from the physical servers belonging to each cloud server of the target server type can specifically be: based on the resource supply position corresponding to the current time, the resource supply quantity corresponding to each resource supply position, and the current inventory resource quantity of each cloud server, the current actual resource usage quantity and deployment information, determine the target physical server and at least one of the information such as the relocation and regularization information or the balanced relocation information or the online power saving information of each target physical server from the physical servers belonging to multiple cloud servers.

[0107] Among them, balanced relocation means that for multiple physical servers belonging to the same area, according to the current use resources and current inventory resources of each physical server, the current use resources in the target physical server are completely moved to other servers, and then the target physical server is transported to other areas by express delivery or logistics to go online and deploy a cloud server of the target server type in the target physical server. Online power saving means that when the cloud server deployed by the physical server is not rented or sold, a physical server adopts a working state to save electricity. In this working state, the power consumed is much lower than the power consumed when the cloud server deployed by the physical server is not rented or sold but is powered on.

[0108] If the difference between the target reference resource amount and the current actual total resource usage is smaller than the minimum value of the preset threshold range, step S350: counting the total resource amount of the physical server of the target server type in the online power saving state.

[0109] It should be understood that if the difference between the target reference resource amount and the current actual total amount of used resources is less than the minimum value of the preset threshold range, it means that the current inventory of resources is small, and there may be a situation where the cloud server purchasing customer cannot purchase the resource data they need in time. In this case, it is necessary to power on the physical server of the target server type that is online to save power in order to increase the current inventory of the target server type. It may even be necessary for the cloud server supplier to purchase physical servers and put the purchased physical servers or the cloud server supplier's spare and not online physical servers online to deploy the cloud server type of the target server type in the online processed physical servers. This will increase the current inventory of the target server type.

[0110] Step S360: Determine whether the total resource amount is greater than the difference between the target reference resource amount and the current actual total resource amount used.

[0111] If the total amount of resources is not less than the difference between the target reference amount of resources and the total amount of resources currently actually used, execute step S370: based on the difference and the deployment information of the cloud server in the online power saving state, determine the target physical server from the physical servers in the online power saving state, and generate resource scheduling information including the deployment information of the target physical server and the power-on operation of the target physical server.

[0112] It should be understood that if the total amount of resources is not less than the difference between the target reference amount of resources and the total amount of resources currently actually used, it means that the sum of the total amount of resources of the target server type that can be provided by the physical server in the online power-saving state and the current total amount of inventory resources is greater than the target reference amount of resources. Here, there is no need for the cloud server provider to purchase physical servers or install and deploy spare physical servers.

[0113] The above-mentioned step S370 can specifically be, based on the difference, the deployment information of the cloud server in the online power saving state, and the resource supply amount and resource supply position corresponding to the current moment in the computing resource supply data, determine the physical server corresponding to the deployment information matching the resource supply position from the physical servers in the online power saving state as the target physical server.

[0114] For example, if the difference is 500G of memory resources, the cloud server deployment information in the online power saving state includes 5 physical servers in area A, each of which can provide 100G of memory resources, and 5 physical servers in area B, each of which can provide 100G of memory resources, and the corresponding resource supply at the current moment is 200G, and the supply location is area B. Then, according to the above method, the target physical servers determined from the physical servers in the online power saving state are five, and at least two of them can include physical servers in area B. It should be understood that the remaining 3 target physical servers can be randomly selected from the set of physical servers in area A and area B that are in the online power saving state.

[0115] If the total amount of resources is less than the difference between the current inventory amount of resources and the target reference amount of resources, execute step S380: determine the amount of resources to be purchased based on the difference and the total amount of resources, and generate resource scheduling information including the amount of resources to be purchased, deployment information of the physical server in the online power saving state, and power-on operation of the physical server.

[0116] It should be understood that if the total amount of resources is less than the difference between the current inventory amount of resources and the target reference amount of resources, it means that the total amount of resources of the target server type that can be provided by the physical server in the online power saving state and the total amount of resources in the current inventory are less than the target reference amount of resources. Here, the cloud server provider needs to purchase physical servers or install and deploy spare physical servers. If purchase is required, the physical server model to be purchased and the purchase quantity corresponding to each physical server model can be determined based on the amount of resources to be purchased and the preset physical server model, as well as the amount of resources that can be provided by the physical server corresponding to the model.

[0117] For example, if the amount of resources to be purchased is 500G, the amount of resources that can be provided by the physical server corresponding to model 1 is 100G, and the amount of resources that can be provided by the physical server corresponding to model 2 is 200G, then you can generate the purchase of 5 physical servers corresponding to model 1, or purchase 3 physical servers corresponding to model 1 and 1 physical server corresponding to model 2, or purchase 1 physical server corresponding to model 1 and 2 physical servers corresponding to model 2.

[0118] To further improve resource utilization, in one possible implementation, the method further includes:

[0119] Based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity of each cloud server, the current actual resource usage quantity and deployment information, the designated physical server and the maneuvering and regularization operation information performed on the designated physical server are determined from the physical servers belonging to each cloud server, and resource scheduling information including the deployment information of the designated physical server and the maneuvering and regularization operation information is generated.

[0120] Among them, regular relocation means that for multiple physical servers belonging to the same area, according to the current usage resources and current inventory resources of each physical server, the cloud servers provided by the target physical server to the cloud server purchaser will be transferred to be provided to the cloud server purchaser by other physical servers, so that the cloud servers deployed in the target physical server are no longer provided to the cloud server purchaser, that is, the current usage resources of each cloud server deployed on the target physical server are zero, so that online power saving can be performed on the target physical server, or the current inventory resources are made greater than a certain threshold, so that cloud servers with larger required resources can be directly provided to the cloud server purchaser.

[0121] The above-mentioned method of determining the designated physical server can specifically be: based on the resource supply position corresponding to the current time in the computing resource supply data, the resource supply quantity corresponding to each supply position, the current inventory resource quantity of each cloud server, the current actual resource usage quantity and deployment information, determine the designated physical server and the relocation and regularization operation information performed on the designated physical server from the physical servers belonging to each cloud server.

[0122] Specifically, a designated physical server can be determined from multiple physical servers based on the resource supply position corresponding to the current time, the current inventory resource quantity of each cloud server, the current actual resource usage and deployment information, and a relocation and regularization operation can be performed on the designated physical server, wherein the deployment position included in the deployment information of each cloud server deployed in the designated physical server matches the resource supply position.

[0123] It is also possible to determine a designated physical server from multiple physical servers and perform a relocation and regularization operation on the designated physical server based on the resource supply position corresponding to the current time in the computing resource supply data, the amount of resources required for each of at least one cloud server to be supplied that corresponds to each resource supply position, the current inventory of resources of each cloud server, the current actual amount of resources used and the deployment information, wherein the corresponding deployment position of each cloud server deployed in the designated physical server should match the resource supply position, and the cloud server deployed in the at least one designated physical server after the relocation and regularization can provide the required amount of resources to at least one cloud server to be supplied.

[0124] Please refer to Figure 5-9 , taking the resource management method provided in the embodiment of the present application as an example, the specific application in the resource management system is as follows: Figure 5 As shown, the system includes a resource intelligent planning engine, a water level comparator and a resource scheduler. It should be understood that the resource intelligent planning engine, the water level comparator and the resource scheduler can be software program modules. When the resource management method is executed by using each software program module in the resource management system, the specific structure and processing process of each software program module are as follows:

[0125] 1) The intelligent resource planning engine can be used to dynamically update the elastic safety stock water level (target reference resource quantity) for each server type at regular intervals (e.g., one day, two days, or one week) based on the resource data of the specified server type, and the elastic safety stock water level (target reference resource quantity) is calculated based on business data (resource data) such as current resource data at the current moment, historical resource data, and computing resource supply data.

[0126] like Figure 6The figure shows a hierarchical diagram of the resource intelligent planning engine. Since the intelligent resource planning engine is designed to solve the data analysis of the resource supply process and the prediction of the target reference resource quantity, the entire intelligent planning engine is divided into three parts: the data layer, the intelligent model layer, and the strategy adjustment layer.

[0127] Please refer to Figure 7 The data layer mainly classifies and stores the business data corresponding to the cloud servers of different server types in the server cluster, providing data support for the subsequent resource management analysis. The intelligent model layer mainly uses data warehouse technology to obtain the resource data of the target server type (specifically, the business data of the same server type in the server cluster can be processed into multi-dimensional metadata through ETL technology, and then stored in the data warehouse. The multi-dimensional metadata mainly includes the deployment information of the cloud server of the target server type (such as the rack information of the computer room park, the inventory of instance specifications of each park), the computing resource supply data of the target server type (such as the metadata of the instance purchased by the major customer, the metadata of the supply chain operation process), and the resource data of the target server type, and makes an optimistic estimate to obtain the initial reference resource volume. The policy adjustment layer mainly adjusts the initial reference resource volume according to the current sales situation of major customers, seasons, holidays and sudden sales to obtain the target reference resource volume, ensuring that the target reference resource volume matches the current actual operation situation.

[0128] When using the above-mentioned intelligent planning engine to predict the target reference resource quantity, for the target server type, firstly make an optimistic estimate of the resources required for the target server type based on the optimistic estimation algorithm with purchase success priority, and the calculation formula is: (total resource purchase quantity * 1.5 + historical reference resource quantity) - the current total inventory resource quantity, and the initial reference resource quantity is calculated by this formula. Then, according to the actual resource consumption every two days (target historical resource usage) and the resource supply quantity corresponding to the current time in the calculation resource supply data, an intermediate reference resource quantity is determined, and the initial reference resource quantity is subtracted from the intermediate reference resource quantity to obtain the redundant resource quantity, and the initial reference resource quantity is gradient adjusted using the redundant resource quantity. For example, if the ratio between the redundant resource quantity and the initial reference resource quantity is between 20% and 30%, the initial reference resource quantity can be reduced by 20%; if the ratio between the redundant resource quantity and the initial reference resource quantity is between 10% and 20%, the initial reference resource quantity is reduced by 10%. The above-mentioned adjustment method is used to perform gradient adjustment to obtain an accurate target reference resource quantity. So that when the difference between the target reference resource quantity and the current actual total amount of resources in use is less than the minimum value of the preset threshold range, the inventory resources can be replenished immediately; and when the difference between the current inventory resource quantity and the target reference resource quantity is greater than the maximum value of the preset threshold range, the inventory can be reduced. At the same time, since the above adjustment of the initial reference resource quantity takes into account the supply quantity corresponding to the current time in the computing resource supply data, it effectively ensures that large customers can purchase resources.

[0129] 2) Water level comparator

[0130] Please refer to Figure 8 The water level comparator is used to periodically and automatically compare the current actual total amount of resources used with the target reference amount of resources to obtain inventory comparison results. This allows the subsequent resource scheduler to provide targeted power saving, relocation, relocation and return strategies through network-wide resource data analysis when the inventory is sufficient; when the inventory is insufficient, it takes into account business growth and current power-saving machine data to provide resource power-on, resource regulation, balance and procurement strategies.

[0131] 3) Resource Scheduler:

[0132] Please refer to Fig. 9 After the water level comparator obtains the comparison result, the resource scheduler is used to implement automatic resource scheduling according to the resource scheduling strategies given for the current scenarios of sufficient and insufficient total inventory resources, such as automatically submitting the plan for purchasing physical servers, power-on processing, power-saving processing, balanced relocation and resource regularization, so that the current total inventory resources that can be provided by the physical servers in the existing network operation (in the online power-on state) match the target reference resource amount, so as to achieve full utilization of resources.

[0133] The resource scheduler is used to relocate machines to other regions with lower water levels and adopt power-saving strategies to reduce the current inventory water level when the target reference resource volume is greater than the current total inventory resource volume; when the target reference resource volume is less than the current total inventory resource volume, it is necessary to save power and organize resources for some physical servers, and prepare the procurement needs of physical servers in advance. Fig.10 , which is a schematic diagram showing the total amount of resources actually used, the target reference amount of resources, and the total amount of resources provided by the physical servers of the online target server type over time. It should be understood that Fig.10 The online physical servers include physical servers for deploying cloud servers of the target server type that are in power-saving and power-on states. If the target reference resources currently required are small, and the total resources provided by the physical servers in the online power-on state are large, in order to avoid energy waste and low resource usage, the resources can be reduced by reorganizing the physical servers and saving power online, and even the physical servers can be shut down to adjust the current total inventory resources. If the target reference resources currently required are large, and the total resources provided by the physical servers in the online power-on state are small, the current total inventory resources can be adjusted by powering on the online power-saving physical servers, or even relocating and purchasing physical servers, so that the adjusted total inventory resources match the current actual operating conditions.

[0134] like Fig.11 Shown is the supply chain roadmap of the cloud server: the supply chain specifically includes data analysis, forecasting, procurement and sales, etc., wherein the above-mentioned data analysis and forecasting processes can be performed by the above-mentioned resource management system, and the subsequent procurement and sales processes can be performed by the server cluster.

[0135] Among them, data analysis refers to the use of data warehouse related technologies to extract, convert and load business data related to the target server type. In fact, it is to classify and divide the daily sales data, analyze and clean it in multiple dimensions, and other operations. Among them, business data may include resource data related to the target server type and the deployment information of each cloud server corresponding to the target server type. Forecasting refers to the deduction of resource usage in the future based on resource data over a long period of time in the past to obtain an initial reference resource volume. At the same time, it is also necessary to consider the impact of holidays, activities and emergencies as well as resource data over a short period of time in the past to adjust and correct the initial reference resource volume to obtain a target reference resource volume. Procurement means that cloud service buyers can submit purchase requests to cloud server providers. Sales means that cloud server providers provide cloud server buyers with the purchased cloud servers based on purchase requests.

[0136] By utilizing the resource management system described above for resource management, it is possible to utilize the resource planning engine to perform real-time analysis and flexible prediction of business demand situations, thereby solving the problem of traditional resource shortages and procurement redundancy being realized only afterthoughts. By utilizing the scheduler for resource inventory and resource scheduling, the entire process can be completed in a short period of time, thereby avoiding the tedious manual operation of power saving, power on, procurement, relocation, and relocation of physical servers in the traditional resource scheduling process, thereby greatly improving work efficiency. Therefore, by adopting the resource management method described above, it is also possible to solve the two problems of sufficient and insufficient resources in the existing network while improving the utilization rate of idle resources.

[0137] Fig.12 The resource management device 400 is shown according to an embodiment. The resource management device 400 includes: a data acquisition module 410, a resource quantity estimation module 420 and a scheduling information generation module 430. The data acquisition module 410 acquires resource data for the target server type, the resource data includes historical resource data, computing resource supply data and current resource data corresponding to the current time, and the current resource data includes the current total inventory resource amount and the current total actual use resource amount; the resource quantity estimation module 420 is used to estimate resource usage based on the historical resource data, computing resource supply data and current resource data, and obtain the target reference resource amount corresponding to the target server type at the current time; the scheduling information generation module 430 is used to generate resource scheduling information based on the target reference resource amount and the current total inventory resource amount, so that after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource amount corresponding to the target server type is less than the target reference resource amount, and the difference between the target reference resource amount corresponding to the target server type at the current time and the current total actual use resource amount is within a preset threshold range.

[0138] In one possible implementation, the historical resource data includes the historical actual resource usage, historical purchased resource usage, and historical reference resource usage at multiple time points within a historical time period; the resource estimation module 420 includes a resource estimation submodule and a resource adjustment submodule. The resource estimation submodule is used to estimate based on the historical purchased resource usage, the target historical reference resource usage, and the current resource data at multiple time points within a first preset duration to obtain an initial reference resource usage; the resource adjustment submodule is used to adjust the initial reference resource usage based on the historical actual resource usage and computing resource supply data within a second preset duration to obtain a target reference resource usage, wherein the first preset duration is greater than the second preset duration.

[0139] In one possible implementation, the resource quantity estimation submodule is also used to determine the total resource purchase quantity within the first preset time period based on the historical resource purchase quantities at multiple time points within the first preset time period; and determine the initial reference resource quantity based on the total resource purchase quantity within the first preset time period, the target historical reference resource quantity and the current total inventory resource quantity.

[0140] In one possible implementation, the resource quantity estimation submodule is further used to perform weighted summation of the total resource purchase quantity and the target historical reference resource quantity to obtain a weighted result; and subtract the weighted result from the current total inventory resource quantity to obtain an initial reference resource quantity.

[0141] In one possible implementation, the resource quantity adjustment submodule is also used to obtain the target historical resource usage based on the historical actual resource usage within a second preset time period; add the supply resource quantity corresponding to the current time in the computing resource supply data to the target historical resource usage to obtain the intermediate reference resource quantity; determine the redundant resource ratio based on the initial reference resource quantity and the intermediate reference resource quantity; adjust the initial reference resource quantity based on the redundant resource ratio to obtain the target reference resource quantity.

[0142] In one possible implementation, the resource amount adjustment submodule is further used to determine a resource adjustment strategy corresponding to the redundant resource ratio; and adjust the initial reference resource amount according to the resource adjustment strategy corresponding to the redundant resource ratio to obtain a target reference resource amount.

[0143] In one embodiment, the scheduling information generation module 430 is also used to determine the target physical server and resource scheduling operation information for the target physical server from the physical servers belonging to each cloud server of the target server type when the difference between the target reference resource quantity and the current actual total resource usage is greater than the maximum value of a preset threshold range, based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity and deployment information of each cloud server belonging to the target server type, the resource scheduling operation information includes at least one of online power saving operation information and balanced relocation operation information for the target physical service.

[0144] In one possible implementation, the scheduling information generation module 430 includes a resource statistics submodule and a scheduling information generation submodule. The resource statistics submodule is used to count the total resource amount of the physical servers of the target server type in the online power saving state when the difference between the target reference resource amount and the current actual total amount of used resources is less than the minimum value of the preset threshold range; the scheduling information generation submodule is used to determine the target physical server from the physical servers in the online power saving state based on the difference and the deployment information of the cloud server in the online power saving state when the total amount of resources is not less than the difference between the target reference resource amount and the current actual total amount of used resources, and generate the deployment information of the target physical server and the resource scheduling information for powering on the target physical server.

[0145] In one possible implementation, the scheduling information generation submodule is also used to determine the amount of resources to be purchased based on the difference and the total amount of resources when the total amount of resources of the physical servers in the online power-saving state is less than the difference between the current inventory amount of resources and the target reference amount of resources, and to generate resource scheduling information including the amount of resources to be purchased, the deployment information of the physical servers in the online power-saving state, and the power-on operation of the physical servers.

[0146] In one possible implementation, the scheduling information generation module 430 is also used to determine the designated physical server and the maneuvering and regularization operation information performed on the designated physical server from the physical servers to which each cloud server belongs based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity of each cloud server, the current actual resource usage quantity and deployment information, and generate resource scheduling information including the deployment information of the designated physical server and the maneuvering and regularization operation information.

[0147] It should be noted that the device 400 embodiment in the present application corresponds to the aforementioned method embodiment. The specific principles of the device 400 embodiment can be found in the aforementioned method embodiment, and will not be repeated here.

[0148] The following will be combined Fig.13 An electronic device provided by the present application is described.

[0149] See also Fig.13 Based on the resource management method provided in the above embodiment, the embodiment of the present application also provides another electronic device 100 including a processor 102 that can execute the above method. The electronic device 100 can be a server or a terminal device, and the terminal device can be a smart phone, tablet computer, computer, portable computer or other device.

[0150] The electronic device 100 further includes a memory 104 . The memory 104 stores a program that can execute the contents of the aforementioned embodiments, and the processor 102 can execute the program stored in the memory 104 .

[0151] Among them, the processor 102 may include one or more cores for processing data and a message matrix unit. The processor 102 uses various interfaces and lines to connect various parts of the entire electronic device 100, and executes various functions and processes data of the electronic device 100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 104, and calling data stored in the memory 104. Optionally, the processor 102 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 102 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 102, but may be implemented separately through a communication chip.

[0152] The memory 104 may include a random access memory (RAM) or a read-only memory (ROM). The memory 104 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 104 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the following various method embodiments, etc. The data storage area may also store data obtained by the electronic device 100 during use, etc.

[0153] In some embodiments, the electronic device 100 may further include: a peripheral interface 106 and at least one peripheral device. The processor 102, the memory 104 and the peripheral interface 106 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral interface via a bus, a signal line or a circuit board. Exemplarily, in this embodiment, the peripheral device may include a network module 108, which is used to receive and send electromagnetic waves, realize the mutual conversion of electromagnetic waves and electrical signals, and thus communicate with a communication network or other devices, such as communicating with an audio playback device. The network module 108 may include various existing circuit elements for performing these functions, such as an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a user identity module (SIM) card, a memory, etc. The network module 108 may communicate with various networks such as the Internet, an intranet, a wireless network, or communicate with other devices via a wireless network. The above-mentioned wireless network may include a cellular telephone network, a wireless local area network or a metropolitan area network.

[0154] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable medium stores program codes, which can be called by a processor to execute the method described in the above method embodiment.

[0155] The computer-readable storage medium may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program codes that execute any of the method steps in the above method. These program codes can be read from or written to one or more computer program products. The program code can be compressed, for example, in an appropriate form.

[0156] The embodiment of the present application also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device performs the method described in the above various optional implementations.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A resource management method, characterized in that: The method comprises: Obtain resource data for the target server type, the resource data including historical resource data, computing resource supply data, and current resource data corresponding to the current time, the current resource data including the current total inventory resources and the current total actual use of resources; Based on the historical resource data, computing resource supply data and current resource data, resource usage estimation is performed to obtain a target reference resource amount corresponding to the target server type at the current time; Based on the target reference resource quantity and the current total inventory resource quantity, resource scheduling information is generated so that after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource quantity corresponding to the target server type is less than the target reference resource quantity, and the difference between the target reference resource quantity corresponding to the target server type at the current time and the current actual total amount of resources in use is within a preset threshold range.

2. The method according to claim 1, characterized in that The historical resource data includes historical actual resource usage, historical purchased resource volume and historical reference resource volume at multiple time points within a historical period; The performing resource usage estimation based on the historical resource data, the computing resource supply data and the current resource data to obtain the target reference resource amount corresponding to the target server type at the current time includes: An initial reference resource volume is obtained by estimating based on the historical purchased resource volume, the target historical reference resource volume and the current resource data at multiple time points within the first preset time period; Based on the historical actual usage resources within a second preset time period and the computing resource supply data, the initial reference resource amount is adjusted to obtain a target reference resource amount, and the first preset time period is greater than the second preset time period.

3. The method according to claim 2, characterized in that Based on the historical purchased resource quantities at multiple time points within the first preset time period, the target historical reference resource quantities and the current resource data, an initial reference resource quantity is obtained, including: Determine the total resource purchase amount within the first preset time period based on the historical purchase amount of resources at multiple time points within the first preset time period; The initial reference resource quantity is determined according to the total resource purchase quantity within the first preset time period, the target historical reference resource quantity and the current total inventory resource quantity.

4. The method according to claim 3, characterized in that The determining the initial reference resource amount according to the total resource purchase amount within the first preset time period, the target historical reference resource amount and the current total inventory resource amount includes: Performing a weighted summation of the total resource purchase amount and the target historical reference resource amount to obtain a weighted result; Subtract the weighted result from the current total amount of inventory resources to obtain an initial reference resource amount.

5. The method according to claim 2, characterized in that: The adjusting the initial reference resource amount based on the historical actual resource usage within the second preset time period and the computing resource supply data to obtain the target reference resource amount includes: Based on the historical actual resource usage within the second preset time period, the target historical resource usage is obtained; Adding the supply resource amount corresponding to the current time in the computing resource supply data to the target historical usage resource to obtain an intermediate reference resource amount; Determining a redundant resource ratio according to the initial reference resource amount and the intermediate reference resource amount; According to the redundant resource ratio, the initial reference resource amount is adjusted to obtain a target reference resource amount.

6. The method according to claim 5, characterized in that The adjusting the initial reference resource amount according to the redundant resource ratio to obtain a target reference resource amount includes: Determine a resource adjustment strategy corresponding to the redundant resource ratio; According to the resource adjustment strategy corresponding to the redundant resource ratio, the initial reference resource amount is adjusted to obtain the target reference resource amount.

7. The method according to claim 1, characterized in that The generating resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity includes: If the difference between the target reference resource amount and the current actual total amount of resources in use is greater than the maximum value of a preset threshold range, based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource amount and deployment information of each cloud server belonging to the target server type, the target physical server and resource scheduling operation information for the target physical server are determined from the physical servers belonging to each cloud server belonging to the target server type, and the resource scheduling operation information includes at least one of online power saving operation information and balanced relocation operation information for the target physical service.

8. The method according to claim 7, characterized in that The generating resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity also includes: If the difference between the target reference resource amount and the current actual total amount of used resources is less than the minimum value of the preset threshold range, counting the total resource amount of physical servers of the target server type in the online power saving state; If the total amount of resources is not less than the difference between the target reference amount of resources and the current total amount of actually used resources, based on the difference and the deployment information of the cloud server in the online power saving state, the target physical server is determined from the physical servers in the online power saving state, and resource scheduling information including the deployment information of the target physical server and the power-on operation of the target physical server is generated.

9. The method according to claim 8, characterized in that The generating resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity also includes: If the total amount of resources is less than the difference between the current inventory amount of resources and the target reference amount of resources, the amount of resources to be purchased is determined based on the difference and the total amount of resources, and resource scheduling information including the amount of resources to be purchased, the deployment information of the physical server in the online power saving state, and the power-on operation of the physical server is generated.

10. The method according to any one of claims 7 to 9, characterized in that: The generating resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity also includes: Based on the resource supply position corresponding to the current time in the computing resource supply data, the current inventory resource quantity of each cloud server, the current actual resource usage quantity and deployment information, the designated physical server and the maneuvering and regularization operation information performed on the designated physical server are determined from the physical servers to which each cloud server belongs, and resource scheduling information including the deployment information and maneuvering and regularization operation information of the designated physical server is generated.

11. A resource management device, characterized in that: The device comprises: A data acquisition module is used to acquire resource data for a target server type, wherein the resource data includes historical resource data, computing resource supply data, and current resource data corresponding to the current time, wherein the current resource data includes the current total inventory resource amount and the current total actual usage resource amount; A resource quantity estimation module, used to estimate resource usage based on the historical resource data, computing resource supply data and current resource data, and obtain the target reference resource quantity corresponding to the target server type at the current time; A scheduling information generation module is used to generate resource scheduling information based on the target reference resource quantity and the current total inventory resource quantity, so that after the resources under the target server type are scheduled according to the resource scheduling information, the inventory computing resource quantity corresponding to the target server type is less than the target reference resource quantity, and the difference between the target reference resource quantity corresponding to the target server type at the current time and the current actual total amount of resources used is within a preset threshold range.

12. An electronic device, characterized in that: include: one or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method according to any one of claims 1-10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, and the program codes can be called by a processor to execute the method according to any one of claims 1 to 10.

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

Images