Resource supply solution determination method, apparatus and computing device cluster

Abstract

The present application relates to the technical field of cloud computing. Disclosed are a resource supply solution determination method, an apparatus and a computing device cluster. The method is executed by a cloud platform and comprises: on the basis of effect reference information of an event occurring in a first time period, by means of a resource supply prediction model, the cloud platform predicts the resource supply of the cloud platform in a second time period to obtain resource supply reference information, and thus determines a resource supply solution to the cloud platform. As the effect of the occurred event on the resource supply of the cloud platform in the future time period is taken into full account during the process, the accuracy of the resource supply reference information is high, thus improving the accuracy of the resource supply solution.

Description

Resource supply scheme determination method, device and computing device cluster

[0001] This application claims priority to Chinese patent application No. 202311525444.2, filed on November 14, 2023, with the invention name “Cloud resource allocation method, apparatus, computing device cluster and storage medium”, and Chinese patent application No. 202410382374.8, filed on March 28, 2024, with the invention name “Resource supply plan determination method, apparatus and computing device cluster”, all of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of cloud computing technology, and in particular to a method and apparatus for determining a resource provisioning solution and a computing device cluster. Background Art

[0003] With the rapid development of cloud computing technology, cloud computing services have become increasingly popular among users. Currently, cloud platforms used to provide cloud computing services can provide users with various cloud computing resources based on their needs, such as hardware resources such as computing, storage, and networking, as well as virtual resources such as load balancing, artificial intelligence (AI) enablement, and data.

[0004] In related technologies, in order to achieve accurate resource supply, cloud platforms usually determine the resource supply plan for the future time period based on the user's tag information (for example, for social events, user A's tag information is "strong", indicating that when a social event occurs, user A has a high demand for resources, and for news events, user A's tag information is "medium", indicating that when a news event occurs, user A has an average demand for resources), the user's resource request, and the resource supply prediction results obtained using the AI ​​model, etc.

[0005] However, the above method does not consider the impact of the occurrence of real-time events on resource supply in future time periods, resulting in poor accuracy of resource supply plans.

[0006] Summary of the Invention

[0007] The embodiments of the present application provide a method, apparatus, and computing device cluster for determining a resource provisioning solution, which can improve the accuracy of the resource provisioning solution.

[0008] On the first aspect, the present application provides a method for determining a resource supply solution, which can be applied to resource supply scenarios in the field of cloud computing technology, wherein resource supply refers to the process of allocating and issuing resources on a cloud platform. Schematically, the resources involved in the present application refer to cloud computing resources (referred to as cloud resources), which can be hardware resources such as computing, storage, and networking, or virtual resources such as load balancing, AI enabling, and data, without limitation. For example, virtual machines (VMs) are cloud resources that are flexibly priced for users according to usage time and are highly favored by users. By virtualizing hardware resources and providing virtual machines with corresponding computing capabilities according to user needs, one piece of hardware can provide cloud computing services to multiple users, which can effectively improve the utilization rate of hardware resources. Schematically, the method provided by the present application is executed by a cloud platform, and the method includes:

[0009] Obtaining influence reference information of an event occurring in a first time period, where the influence reference information indicates an impact of the event on resource provisioning on the cloud platform;

[0010] The influence reference information is processed by a resource supply prediction model to obtain resource supply reference information of the cloud platform. The resource supply prediction model is obtained based on training of an artificial intelligence model. The resource supply reference information indicates the predicted resource supply of the cloud platform in a second time period under the influence of the event. The second time period is a time period after the first time period.

[0011] Based on the resource supply reference information, a resource supply plan of the cloud platform in the second time period is determined.

[0012] Through this method, the cloud platform uses the resource supply prediction model to predict the cloud platform's resource supply in the second time period based on the influence reference information of events that occurred in the first time period. This information is then used to determine the cloud platform's resource supply plan. Because this process fully considers the impact of past events on the cloud platform's resource supply in future time periods, the resource supply reference information is highly accurate, thereby improving the accuracy of the resource supply plan.

[0013] In some embodiments, the method further comprises:

[0014] Based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period, supply prompt information of the cloud platform is determined, where the supply prompt information indicates a predicted supply status of the resources on the cloud platform in the second time period.

[0015] In some embodiments, the method further includes: outputting a resource management interface of the cloud platform, wherein the resource management interface includes the provisioning prompt information.

[0016] In some embodiments, determining the supply prompt information of the cloud platform based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period includes any of the following:

[0017] If the resources indicated by the resource supply plan are more than the to-be-allocated resources, determining first supply prompt information of the cloud platform, wherein the first supply prompt information indicates that there is a shortage of resources on the cloud platform during the second time period;

[0018] If the resources indicated by the resource supply plan are less than or equal to the to-be-allocated resources, second supply prompt information of the cloud platform is determined, where the second supply prompt information indicates that the resource supply on the cloud platform is sufficient in the second time period.

[0019] By determining the supply prompt information of the cloud platform, relevant personnel can promptly learn the predicted supply status of resources on the cloud platform in the second time period.

[0020] In some embodiments, the method further comprises:

[0021] If the resources indicated by the resource supply plan are more than the resources to be allocated on the cloud platform in the second time period, the resource guarantee plan of the cloud platform is determined based on the difference between the resources indicated by the resource supply plan and the resources to be allocated, and the resource guarantee plan indicates a method for expanding the resources on the cloud platform in the second time period.

[0022] In some embodiments, the method further includes: outputting a resource management interface of the cloud platform, wherein the resource management interface includes the resource guarantee plan.

[0023] The above process means that if there is a shortage of resources on the cloud platform in the second time period, the cloud platform can automatically generate a resource guarantee plan, which makes it easier for relevant personnel to adjust resources according to the resource guarantee plan and improve the efficiency of human-computer interaction.

[0024] In some embodiments, the influence reference information includes first reference information and second reference information, wherein the first reference information indicates the intensity of the impact of the event on the resource supply on the cloud platform, and the second reference information indicates the duration of the impact of the event on the resource supply on the cloud platform.

[0025] It should be understood that different events have different impacts on resource provisioning on cloud platforms. This approach allows us to measure the impact of events on resource provisioning on cloud platforms from different dimensions, improving the accuracy of impact reference information.

[0026] In some embodiments, the method further comprises:

[0027] Acquire sample events associated with the event, where the sample events refer to events occurring in a third time period, and the third time period is a time period before the first time period;

[0028] Based on the influence reference information of the sample events and the sample resource supply reference information, the artificial intelligence model is trained to obtain the resource supply prediction model.

[0029] In this way, training the model based on sample events related to events that have occurred can improve the accuracy of the model and thus improve the accuracy of resource supply reference information.

[0030] In some embodiments, the processing of the influence reference information by a resource supply prediction model to obtain the resource supply reference information of the cloud platform includes:

[0031] The resource supply prediction model is used to process the influence reference information and the event label of the event to obtain the resource supply reference information, where the event label indicates a feature of the event.

[0032] Through the above method, the resource supply reference information of the cloud platform is determined by combining the characteristics of the event and the reference information of the event's influence. The fine-grained impact of the event on the resource supply on the cloud platform is fully considered, and the accuracy of the resource supply reference information is further improved. It provides a scientific and quantitative model guarantee and explainability for the subsequent determination of the resource supply plan.

[0033] In some embodiments, determining the resource provisioning solution of the cloud platform based on the resource provisioning reference information includes:

[0034] A resource supply plan for the cloud platform is determined based on the resource supply reference information and a predicted event, where the predicted event refers to an event that may occur in the second time period.

[0035] In this way, when determining the resource supply plan, not only the impact of past events on the resource supply on the cloud platform is fully considered, but also events that may occur in the future are taken into account, thereby further improving the accuracy of the resource supply plan.

[0036] In a second aspect, the present application provides a device for determining a resource supply plan, which is configured on a cloud platform. The device includes at least one functional module for executing a method for determining a resource supply plan as provided in the first aspect or any possible implementation of the first aspect.

[0037] In a third aspect, an embodiment of the present application provides a computing device cluster comprising at least one computing device, each computing device comprising a processor and a memory; the processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes a method for determining a resource supply solution as provided in the aforementioned first aspect or any possible implementation of the first aspect.

[0038] In a fourth aspect, embodiments of the present application provide a computer-readable storage medium comprising computer program instructions. When the computer program instructions are executed by a computing device cluster, the computing device cluster performs the method for determining a resource provisioning solution as provided in the first aspect or any possible implementation of the first aspect. The storage medium includes, but is not limited to, volatile memory, such as random access memory, and non-volatile memory, such as flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0039] In a fifth aspect, embodiments of the present application provide a computer program product comprising instructions that, when executed on a computing device cluster, cause the computing device cluster to implement the method for determining a resource provisioning solution provided in the aforementioned first aspect or any possible implementation of the first aspect. The computer program product may be a software installation package. To implement the aforementioned method, the computer program product may be downloaded and executed on a computing device or computing device cluster. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] FIG1 is a schematic diagram of an implementation environment provided by an embodiment of the present application;

[0041] FIG2 is a flow chart of a method for determining a resource supply solution provided in an embodiment of the present application;

[0042] FIG3 is a schematic diagram of a resource allocation rule provided in an embodiment of the present application;

[0043] FIG4 is a schematic diagram showing the principle of a method for determining a resource supply solution provided in an embodiment of the present application;

[0044] FIG5 is a schematic diagram of a framework of a method for determining a resource supply solution provided in an embodiment of the present application;

[0045] FIG6 is a flowchart of another method for determining a resource supply solution provided in an embodiment of the present application;

[0046] FIG7 is a schematic diagram of a resource management interface provided in an embodiment of the present application;

[0047] FIG8 is a schematic structural diagram of a device for determining a resource supply solution provided in an embodiment of the present application;

[0048] FIG9 is a schematic diagram of the structure of a computing device provided in an embodiment of the present application;

[0049] FIG10 is a schematic diagram of the structure of a computing device cluster provided in an embodiment of the present application;

[0050] FIG11 is a schematic diagram of a connection method of a computing device cluster provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings. It should be noted that the information involved in this application (including but not limited to user device information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of relevant countries and regions. For example, the events involved in this application, the event tags of the events, and the reference information of the influence of the events are all obtained with full authorization.

[0052] For ease of understanding, the relevant terms and concepts involved in this application are first introduced below.

[0053] Cloud computing services refer to cloud computing products that can be provided as services, including cloud hosts, cloud storage, databases, networks, software, analytics, and intelligence. They are provided to users through the cloud (Internet) and are categorized into three levels: infrastructure as a service (IaaS), platform as a service (PaaS), and software-as-a-service (SaaS).

[0054] Cloud computing resources refer to the on-demand, measurable services and products provided by cloud computing services, such as hardware resources such as computing, storage, and networking, as well as virtual resources such as load balancing, artificial intelligence (AI) enablement, and data.

[0055] A virtual central processing unit (VCPU) is a processor resource used in a virtual environment. It is a portion of a physical CPU that can be used independently by a virtual machine. Each VCPU can run instructions as an independent processor core.

[0056] Physical resources refer to devices capable of issuing virtual resources (or virtual machines), such as servers comprised of a physical central processing unit (CPU), memory, graphics processing unit (GPU), or storage device. Accordingly, virtual resources that can be issued by physical resources can include one or more resources allocated, partitioned, and virtualized based on the physical hardware configurations of the CPU, memory, GPU, or storage device. Different physical resources can be of different types, such as RTc3_2 or RTc3_4.

[0057] Artificial intelligence (AI) is based on the principle of combining massive amounts of data with powerful processing capabilities and intelligent algorithms to build an AI model that solves specific problems. This AI model automatically infers and learns underlying patterns or features from the data, thereby achieving a way of thinking similar to that of humans. AI models, also known as AI algorithms (or AI operators), are a general term for mathematical algorithms built on AI principles. They are also the foundation for using AI to solve specific problems, such as deep learning (DL) models.

[0058] The AI ​​big model is a deep neural network composed of a large number of layers and parameters. Due to its powerful predictive ability, it is widely used in various fields such as natural language processing, computer vision, speech synthesis and speech recognition, autonomous driving, etc.

[0059] Natural Language Processing (NLP), a key branch of computer science and artificial intelligence, aims to enable computers to understand, parse, generate, and process natural language. The goal of NLP is to enable natural language interaction between computers and humans, thereby solving various application problems in natural language understanding and generation, such as speech recognition, machine translation, text classification, information extraction, and question-answering systems.

[0060] A prompt is an input provided to an AI model (such as a large AI model for natural language processing) to guide and stimulate the AI ​​model to generate a corresponding response or content. For example, a prompt is a complete paragraph (or sentence) of text content, which can be used as input to a model such as a large language model (LLM), allowing the model to output the corresponding inference results based on the prompt.

[0061] The following is an introduction to the application scenarios and implementation environment of this application.

[0062] The technical solution provided by this application can be applied to resource supply scenarios in the field of cloud computing technology, wherein resource supply refers to the process of allocating and issuing resources on a cloud platform. Schematically, the resources involved in this application refer to cloud computing resources (referred to as cloud resources), which can be hardware resources such as computing, storage, and network, or virtual resources such as load balancing, AI enabling, and data, without limitation. For example, virtual machines (VMs) are cloud resources that are flexibly priced for users according to usage time and are highly favored by users. By virtualizing hardware resources and providing virtual machines with corresponding computing capabilities according to user needs, one piece of hardware can provide cloud computing services to multiple users, which can effectively improve the utilization rate of hardware resources. In order to achieve accurate supply of resources on a cloud platform, this application provides a method for determining a resource supply plan, which can reasonably predict the resource supply situation in future time periods based on the impact of events that have occurred on the resource supply on the cloud platform, and then determine the resource supply plan to improve the accuracy of the resource supply plan.

[0063] FIG1 is a schematic diagram of an implementation environment provided by an embodiment of the present application. As shown in FIG1 , the implementation environment includes a cloud platform 100 .

[0064] Cloud platform 100 is short for cloud computing platform, which refers to a platform that provides computing, network and storage capabilities based on services of hardware resources and software resources. In an embodiment of the present application, cloud platform 100 is used to supply resources to multiple users (users are also resource demanders, and the process of resource supply is also the process of supplying resources to resource demanders). In addition, cloud platform 100 provides management functions for resources on cloud platform 100, for example, configuring resources on cloud platform 100, managing resources used by each user, etc., but the present application is not limited thereto. Schematically, cloud platform 100 can output a resource management interface to be displayed on the terminal used by relevant personnel, so that relevant personnel can perform various operations on the resource management interface to trigger corresponding resource management functions. For example, the terminal installs and runs the application according to the instructions of cloud platform 100, and responds to various operations performed by relevant personnel on the application interface according to the instructions of cloud platform 100, that is, when the method provided in the present application is executed by cloud platform 100, for the terminal used by the relevant personnel, the actions performed by the terminal are supported by cloud platform 100. The terminal may be at least one of a smartphone, a desktop computer, an augmented reality terminal, a tablet computer, an e-book reader, and a laptop computer, but the present application is not limited thereto. Furthermore, the relevant personnel involved in the present application are, for example, resource management personnel, who may be personnel who manage resources on the cloud platform 100 or personnel who manage the required resources on the resource demander's side, but the present application does not limit this.

[0065] Schematically, the cloud platform 100 processes and analyzes huge amounts of data remotely through the network "cloud" and returns it to the user. It has the characteristics of large-scale, distributed, virtualized, high availability, scalability, on-demand service and security. The cloud platform 100 can achieve rapid issuance and release of configurable computing resources at a relatively low management cost or with low interaction complexity between users and service providers. Schematically, the cloud platform 100 is a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. In an embodiment of the present application, the cloud platform 100 provides cloud services to users. In addition, the cloud platform 100 can be a computing device cluster, a virtual machine or a container engine, etc. The embodiment of the present application does not limit the number and type of each device in the implementation environment.

[0066] In some embodiments, the wireless network or wired network described above uses standard communication technologies and / or protocols. The network includes, but is not limited to, any combination of a data center network, a storage area network (SAN), a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some implementations, technologies and / or formats including hypertext markup language (HTML), extensible markup language (XML), etc. are used to represent data exchanged over the network. In addition, conventional encryption technologies such as secure sockets layer (SSL), transport layer security (TLS), virtual private network (VPN), and internet protocol security (IPsec) can be used to encrypt all or part of the links. In other embodiments, customized and / or dedicated data communication technologies can be used to replace or supplement the above-mentioned data communication technologies.

[0067] The following is an introduction to the method for determining the resource supply plan provided by this application.

[0068] FIG2 is a flow chart of a method for determining a resource provisioning solution provided in an embodiment of the present application. As shown in FIG2 , the method is described by taking the method executed by a cloud platform as an example. Schematically, the method includes the following steps 201 to 205.

[0069] 201. The cloud platform obtains influence reference information of an event occurring in a first time period, where the influence reference information indicates an impact of the event on resource supply on the cloud platform.

[0070] In the embodiment of the present application, the first time period is a preset time period and can be set as needed. For example, the first time period is the 24 hours before the current moment, and for another example, the first time period is the last month, the last week, and so on, and the present application does not limit this. In addition, the present application does not limit the number and type of events that occur in the first time period. That is, the event that occurs in the first time period can be one or more. For example, the event can be a natural disaster, an infectious disease, a scientific discovery, an academic research, a sporting event, and so on, and the present application is not limited thereto.

[0071] In some embodiments, the cloud platform obtains multiple candidate events that occurred in the first time period, determines at least one target event that meets the conditions from the multiple candidate events, and obtains reference information on the influence of at least one target event. The content of the conditions can be set according to business needs. For example, the condition is set to the event type as a natural disaster or a sports event, or for another example, the condition is set to the AA area where the event occurred, and so on. This application is not limited to this. This process is that the cloud platform selects a part of the target events that meet the conditions from the events that occurred in the first time period to execute subsequent steps. In this way, the events that occurred in the first time period can be analyzed in a targeted manner according to business needs, saving the computing resources of the cloud platform.

[0072] Taking any event as an example, the event's influence reference information indicates the impact of the event on resource supply on the cloud platform. This means that after the event occurs, due to the impact of the event, user demand for resources will change, thereby affecting resource supply on the cloud platform. For example, if the event is an earthquake in Region AA and the user is a social platform, after the earthquake, due to concerns about post-earthquake reconstruction and material donations, people will continue to follow the event on the social platform and post information related to the event on the social platform. This may increase the demand for resources on the social platform, and accordingly, the cloud platform will need to allocate more resources to the social platform, thus affecting resource supply on the cloud platform. It should be understood that different events often have different impacts on resource supply on the cloud platform. For example, if Event 1 is an earthquake in Region AA and Event 2 is a promotion at BB Supermarket, Event 1 will often have a greater impact on resource supply on the cloud platform than Event 2. Therefore, by obtaining event influence reference information, it is possible to accurately measure the impact of events on resource supply on the cloud platform, providing technical support for the subsequent determination of resource supply plans for the cloud platform.

[0073] In some embodiments, the influence reference information of an event includes first reference information and second reference information, wherein the first reference information indicates the intensity of the impact of the event on the supply of resources on the cloud platform. For example, the first reference information is numerical information, such as a structured score (1-100 points). The larger the numerical value, the greater the impact of the event on the supply of resources on the cloud platform. The second reference information indicates the duration of the impact of the event on the supply of resources on the cloud platform. For example, the second reference information is a time range in days, which is not limited in this application. It should be understood that different events have different impacts on the supply of resources on the cloud platform. In this way, the impact of events on the supply of resources on the cloud platform can be measured from different dimensions, thereby improving the accuracy of the influence reference information.

[0074] Schematically, the cloud platform obtains event information of an event that occurred in a first time period, processes the event information, and obtains reference information on the influence of the event. The event information of an event includes at least one of the following: time of occurrence, location of occurrence, event content, event type, event level, etc., but the present application is not limited thereto. For example, taking the event of an earthquake in region AA as an example, the event information of the event includes: an earthquake occurred in region AA at time Z on day X of month Y, with a magnitude of W and a focal depth of K. It should be understood that this is merely an example and does not constitute a limitation of the present application.

[0075] In some embodiments, taking any event as an example, the cloud platform processes the event information of the event based on the influence prediction model and the first prompt word to obtain the influence reference information of the event. Schematically, the influence prediction model refers to an AI generation model based on natural language processing (such as a large language model LLM, etc.), which is obtained based on historical event training. This application does not limit the training process of the influence prediction model. It should be understood that the process of training an artificial intelligence model based on historical events to obtain an influence prediction model is also the process of summarizing rules from historical events, so that the trained influence prediction model can accurately evaluate the influence of the event on the resource supply on the cloud platform based on the event information of the event that occurred in the first time period, and generate influence reference information to provide technical support for the subsequent determination of the resource supply of the cloud platform. In addition, the first prompt word is a preset prompt word and can be set according to needs. For example, taking the influence reference information including the first reference information and the second reference information as an example, the first prompt word is "Please evaluate the impact of the event on the resource supply on the cloud platform based on the event information of the following event {an earthquake occurred in the AA region at Z time on X month Y day, the earthquake level was W, and the focal depth was K}, and give a structured score (1-100 points) to reflect the intensity of the impact, and give the expected impact time range (in days) to reflect the duration of the impact". The cloud platform processes the event information of the event based on the influence prediction model and the first prompt word, and obtains the influence reference information of the event as "70 points, 14 days". It should be understood that this is only an example and does not constitute a limitation of this application.

[0076] It should be noted that the influence reference information of an event is not limited to being generated through the above-mentioned influence prediction model. Any other network based on machine learning or deep learning and intended to obtain influence reference information can be used to obtain the influence reference information of an event. For example, the cloud platform can extract the event tag of the event based on the event information of the event (for the extraction process of the event tag, please refer to step 202 below), and then determine the influence reference information of the event based on the event tag. This application does not limit this.

[0077] 202. The cloud platform processes the influence reference information through a resource supply prediction model to obtain resource supply reference information of the cloud platform. The resource supply reference information indicates the predicted resource supply of the cloud platform in the second time period under the influence of the event.

[0078] In an embodiment of the present application, a resource supply prediction model is obtained based on artificial intelligence model training, and is used to predict the resource supply of the cloud platform in the second time period based on the influence reference information of the event. In addition, the present application does not limit the type of resource supply prediction model. For example, the resource supply prediction model is a deep neural network. For another example, the resource prediction model is a multi-linear regression fitting prediction model y=f(K(Ev)) (wherein, Ev represents the event, and K(Ev) represents the feature value obtained by feature extraction of event-related information (such as event information and / or event influence reference information)). The present application is not limited to this. Any other network based on machine learning or deep learning and for the purpose of obtaining resource supply reference information can be used as the resource supply prediction model provided by the present application.

[0079] The second time period is the time period after the first time period. The second time period is a preset time period and can be set as needed. For example, if the first time period is 24 hours before the current time, the second time period can be 24 hours after the current time. For another example, if the first time period is the past month, the second time period can be the next three months, and so on. This application does not limit this.

[0080] Taking any event as an example, the resource supply reference information of the event indicates the predicted resource supply of the cloud platform in the second time period under the influence of the event, which means the resource supply that the cloud platform needs to allocate in the second time period due to the influence of the event in the second time period after the occurrence of the event. For example, the resource supply reference information is numerical information, such as the resource supply reference information of event 1 is VCPU=10 (unit is k), which means that the supply of VCPU is 10. In addition, the resource supply reference information can indicate a predicted resource supply in the second time period. For example, if the second time period includes the next three months, such as November and December of XX and January of YY, the resource supply reference information indicates that the predicted resource supply from November of XX to January of YY is VCPU=20. The resource supply reference information can also indicate the predicted resource supply corresponding to each sub-time period in the second time period. For example, the second time period includes the next three months, such as November and December of XX and January of YY. The resource supply reference information indicates that the predicted resource supply in November of XX is VCPU=10, the predicted resource supply in December of XX is VCPU=20, and the predicted resource supply in January of YY is VCPU=30. It should be understood that this is only an example and does not constitute a limitation of this application. The resource supply reference information of the cloud platform is obtained by processing the influence reference information through the resource supply prediction model, which fully considers the impact of the events that have occurred on the resource supply on the cloud platform, can improve the accuracy of the resource supply reference information, and thus provide technical support for the subsequent determination of the resource supply plan of the cloud platform.

[0081] In some embodiments, the cloud platform combines the event tag and influence reference information of an event in the process of obtaining resource supply reference information through the resource supply prediction model. Schematically, the cloud platform processes the influence reference information and the event tag of the event through the resource supply prediction model to obtain the resource supply reference information. The event tag is a semantic tag that can indicate the characteristics of the event. By combining the event tag and influence reference information of an event to obtain the resource supply reference information, the fine-grained impact of the event on the resource supply on the cloud platform in the future time period is fully considered, which can further improve the accuracy of the resource supply reference information. For example, if the influence reference information of event 1 and event 2 is the same (e.g., the influence reference information is "70 points, 14 days"), due to the characteristics of the events themselves, the impact of event 1 and event 2 on the amount of resources required by the user is different, which in turn leads to differences in the amount of resources required to be allocated on the cloud platform. If event 1 has a greater impact on user 1 and a smaller impact on user 2 after it occurs, and event 2 has a greater impact on both user 1 and user 2 after it occurs, then overall, due to the impact of event 2, the amount of resources required to be allocated by the cloud platform is greater than that of event 1. Therefore, through the above method, the resource supply reference information of the cloud platform is determined by combining the characteristics of the event and the reference information of the event's influence, which fully considers the fine-grained impact of the event on the resource supply on the cloud platform in the future time period, further improves the accuracy of the resource supply reference information, and provides a scientific and quantitative model guarantee and explainability for the subsequent determination of the resource supply plan.

[0082] In some embodiments, taking any event as an example, the cloud platform processes the event information of the event based on the label generation model and the second prompt word to obtain the event label of the event. Schematically, the label generation model refers to an AI generation model based on natural language processing (such as a large language model LLM), etc., which is obtained based on historical event training, wherein the historical event refers to an event that occurred before the first time period. This application does not limit the training process of the label generation model. It should be understood that the process of training an artificial intelligence model based on historical events to obtain a label generation model is also the process of summarizing rules from historical events, so that the trained label generation model can accurately generate the event label of the event based on the event information of the event that occurred in the first time period, and provide technical support for the subsequent determination of the resource supply of the cloud platform. In addition, the second prompt word is a preset prompt word and can be set according to needs. For example, the second prompt word is "Please generate an event label for the event based on the event information of the following event {an earthquake occurred in the AA region at Z time on X month Y day, the earthquake level was W, and the focal depth was K}". The cloud platform processes the event information of the event based on the label generation model and the second prompt word, and obtains the event label of the event as "AA region, X month Y day, earthquake, negative" (where "negative" is used to indicate that the event has a negative impact on the resource supply on the cloud platform). It should be understood that this is only an example and does not constitute a limitation of the present application.

[0083] In some embodiments, the resource supply prediction model is an artificial intelligence model that is pre-trained based on historical events and configured on the cloud platform. In this way, the efficiency of the cloud platform in obtaining resource supply reference information can be improved. In other embodiments, the resource supply prediction model is an artificial intelligence model trained based on sample events associated with the event. In this way, training the model based on sample events related to the event that has occurred can improve the accuracy of the model, thereby improving the accuracy of the resource supply reference information. This method is introduced below. Schematically, taking the cloud platform using the resource supply prediction model to process the influence reference information to obtain the resource supply reference information of the cloud platform as an example, the training process of the resource supply prediction model includes: obtaining sample events associated with the event, where the sample event refers to an event that occurs in a third time period, which is a time period before the first time period; based on the influence reference information of the sample event and the sample resource supply reference information, the artificial intelligence model is trained to obtain the resource supply prediction model. In other embodiments, taking the cloud platform using the resource supply prediction model to process the influence reference information and the event label of the event to obtain the resource supply reference information of the cloud platform as an example, the training process of the resource supply prediction model includes the following steps:

[0084] Step 1: Based on the event tag of the event, obtain sample events associated with the event, where the sample events refer to events that occur in the third time period.

[0085] Among them, the third time period is the time period before the first time period, and the third time period is a preset time period, which can be set according to demand. For example, taking the first time period as the most recent month as an example, such as October of XX, the third time period can be from January to September of XX, and so on. This application does not limit this. Schematically, a resource record file is configured on the cloud platform to store historical events that have occurred, historical event tags and historical influence reference information of historical events, and historical resource supply reference information (in this application, it refers to the actual resource supply of the cloud platform in the historical time period), etc. This application does not limit the content stored in the resource record file. Based on the event tag of the event, the cloud platform obtains historical events associated with the event from the resource record file, and uses this part of the historical events as sample events for training the resource supply prediction model. For example, the event tag of event 1 that occurred in the first time period is "AA region, X month Y day, earthquake". Based on the event tag, the cloud platform searches in the resource record file to determine the historical events A and B associated with the event tag, and uses historical events A and B as sample events for training the resource supply prediction model, hereinafter referred to as sample events A and sample events B, wherein, sample events A and B are the sample events for training the resource supply prediction model. The event label of event A is "AA region, H month L day, football match", the influence reference information is "80 points, 12 days", and the sample resource supply information is "the resource supply for the four months is VCPU=10, VCPU=20, VCPU=20, and VCPU=20 respectively". The event label of sample event B is "BB region, X month Z day, earthquake", the influence reference information is "70 points, 14 days", and the sample resource supply information is "the resource supply for the four months is VCPU=15, VCPU=20, VCPU=20, and VCPU=10 respectively".

[0086] Step 2: Based on the event labels of the sample events, the influence reference information of the sample events, and the sample resource supply reference information, the artificial intelligence model is trained to obtain a resource supply prediction model.

[0087] The number of sample events is usually multiple, and the artificial intelligence model is trained through multiple sample events, so that the resource supply prediction model finally trained has good universality and robustness. It should be understood that the process of training the artificial intelligence model based on sample events is also the process of summarizing patterns from historical events, so that the trained resource supply prediction model can accurately predict the resource supply of the cloud platform in the second time period based on the event labels and influence reference information of the events that occurred in the first time period, providing technical support for determining the resource supply of the cloud platform.

[0088] In addition, this application does not limit the training process of the artificial intelligence model. For example, the loss value of the model training process can be determined by constructing an absolute value loss function, a cosine similarity loss function, a square loss function, or a cross entropy loss function, and the model parameters can be continuously adjusted according to the loss value until the training is stopped when the training end conditions are met.

[0089] In some embodiments, the cloud platform is configured with a code language generation model (a large-scale pre-trained programming language model based on Transformers), which can automatically generate a resource supply prediction model based on the event label of the sample event, the influence reference information of the sample event, and the sample resource supply reference information. Schematically, the cloud platform generates a resource supply prediction model based on the code language generation model and the third prompt word. Among them, the third prompt word is a preset prompt word, which can be set according to demand. For example, the third prompt word is "generate {based on the event label of the event, obtain the sample event from the resource record file, with the date of the month, day and week as the date, with the event label of the sample event, the influence reference information of the sample event and the sample resource supply information as features, use matplotlib (a Python library for drawing charts and visualizing data) to draw the resource supply information and save it to a specified file, and predict the resource supply on the cloud platform in the next 4 months} ​​a multi-linear regression prediction model". It should be understood that this is only an example and does not constitute a limitation of this application. In this way, the automatic generation of the resource supply prediction model is achieved, which can improve the processing efficiency of the cloud platform.

[0090] It should be noted that the training process shown in the above steps one and two is introduced by taking the resource supply prediction model as an example in which the resource supply prediction model processes the event labels and influence reference information to obtain the resource supply reference information. In some embodiments, if the resource supply prediction model processes the influence reference information to obtain the resource supply reference information, the process of obtaining sample events during the training of the resource supply prediction model is the same as the above steps. Moreover, the resource supply prediction model can also be obtained by generating a model through a code language, which will not be repeated in this application.

[0091] After the above steps 201 and 202, the cloud platform predicts the resource supply of the cloud platform in the second time period based on the influence reference information of the event that occurred in the first time period through the resource supply prediction model. Since this method fully considers the impact of the event that has occurred on the resource supply of the cloud platform in the future time period, it can improve the accuracy of the predicted resource supply, and provide a scientific and quantitative model guarantee and explainability for the subsequent determination of the resource supply plan. In addition, in the above embodiment, an event is used as an example to illustrate the process of obtaining resource supply reference information through the resource supply prediction model. In the case that multiple events occur in the first time period, the resource supply prediction model can be used to output the resource supply reference information corresponding to each event based on the influence reference information of each event in the multiple events that occurred in the first time period. The cloud platform can use the resource supply reference information corresponding to each event as the resource reference information of the cloud platform, or it can fuse the resource supply reference information corresponding to each event (for example, weighted summation, arithmetic average, etc.) to obtain the resource supply reference information of the cloud platform. This application does not limit this.

[0092] 203. The cloud platform determines a resource supply plan for the cloud platform in the second time period based on the resource supply reference information.

[0093] In an embodiment of the present application, a resource supply plan refers to a plan for the cloud platform to supply resources to at least one user in the second time period. In some embodiments, the cloud platform determines the resource supply plan of the cloud platform in the second time period based on the resource supply reference information and the resource allocation rules. Schematically, the cloud platform calls the resource scheduling program, and based on the resource supply reference information and the resource allocation rules, previews the allocation method of resources on the cloud platform for the second time period to obtain the resource supply plan of the cloud platform. The present application does not limit the number of resource supply plans, which can be one or more. Among them, the resource allocation rules are information pre-configured on the cloud platform, which can be manually configured by relevant personnel or automatically generated by the cloud platform. The present application does not limit this. Schematically, the resource allocation rules indicate the priority of resource allocation on the cloud platform, that is, resources with higher priorities are allocated first. For example, referring to Figure 3, Figure 3 is a schematic diagram of a resource allocation rule provided in an embodiment of the present application. As shown in Figure 3, the virtual machine is a resource to be allocated on the cloud platform. Taking the virtual machine type c3.xlarge.2 as an example, RTc3_2, RTc3_4 and RTgpu_7 in the physical resource pool are all used to issue c3.xlarge.2. If the cloud platform needs to allocate x cores (i.e., x VCPUs) for c3.xlarge.2, then the physical resource pool RTc3_2 will be used to issue c3.xlarge.2 in descending order of priority until x cores are cumulatively issued or the idle server resources in the physical resource pool RTc3_2 are exhausted. If x cores are not met, the physical resource pool RTc_4 will be used to issue c3.xlarge.2 until x cores are cumulatively issued, and so on. It should be noted that the resource allocation rules shown in Figure 3 are for illustration only and do not constitute a limitation of this application.

[0094] In some embodiments, the resource provisioning plan indicates a method for allocating resources on the cloud platform during the second time period. For example, the method for allocating resources on the cloud platform is as follows: for availability zone (AZ) 1, 40 virtual machines of type c3.xlarge.4 are allocated, with a total of 80 VCPUs; 30 virtual machines of type c3.xlarge.2 are allocated, with a total of 60 VCPUs; 40 VCPUs are allocated to user A, and 100 VCPUs are allocated to user B. It should be understood that a cloud platform typically has multiple users, and users can send resource requests to the cloud platform on demand, triggering the cloud platform to release resources on the cloud platform to the users. After determining the resource provisioning plan for the cloud platform in step 203, relevant personnel (such as resource management personnel of the cloud platform) can adjust the resources on the cloud platform in a timely manner according to the resource provisioning plan, for example, by expanding scarce resources, thereby avoiding a situation where resources on the cloud platform are insufficient when users send resource requests to the cloud platform during the second time period. In addition, relevant personnel (such as resource management personnel of resource demanders) can adjust their resource requests in the second time period in a timely manner according to the resource supply plan, such as requesting more or less resources from the cloud platform, etc. For example, the second time period is the next three months, such as November and December of XX and January of YY. The resource supply reference information indicates that the predicted resource supply in November of XX is VCPU=10, the predicted resource supply in December of XX is VCPU=20, and the predicted resource supply in January of YY is VCPU=30, with a total of 60. The cloud platform determines the resource supply plan based on the resource supply reference information in the second time period. For AZ1, 10 virtual machines of type c3.xlarge.4 are allocated, with a total of 20 VCPUs, and 20 virtual machines of type c3.xlarge.2 are allocated, with a total of 40 VCPUs. However, the cloud platform has 10 virtual machines of type c3.xlarge.2 and 20 VCPUs in the resources to be allocated in the second time period, and the resource supply is insufficient. Based on this, the cloud platform can output a resource management interface, which includes a resource supply plan, and relevant personnel can expand resources in a timely manner according to the resource supply plan. For example, the resource management personnel of the cloud platform can expand the amount of resources on the cloud platform in a timely manner according to the resource supply plan, and the resource management personnel of the resource demander can adjust their own resource requests in a timely manner to increase the requested amount of resources. This application does not limit this.

[0095] In some embodiments, the cloud platform determines multiple resource supply schemes of the cloud platform and the evaluation results of each resource supply scheme based on resource supply reference information and resource allocation rules. Among them, the evaluation result of the resource supply scheme indicates the effect of resource supply according to the resource supply scheme. For example, the evaluation result of the resource supply scheme is reflected by a numerical value. The larger the numerical value, the better the effect of resource supply according to the resource supply scheme. In other words, the evaluation result can also be understood as the score obtained by evaluating the resource supply scheme. Of course, the evaluation result of the resource supply scheme can also be reflected in other ways, which is not limited in this application. Schematically, based on the above introduction, it can be seen that the resource allocation rule indicates the priority of resource allocation on the cloud platform, that is, resources with higher priorities are allocated first. When multiple resource supply schemes are determined based on the resource supply reference information and resource allocation rules, the cloud platform can determine the evaluation result of each resource supply scheme based on the priority of resource allocation corresponding to each resource supply scheme. That is, the higher the priority of resource allocation corresponding to the resource supply scheme, the better the evaluation result of the scheme. For example, continuing to refer to Figure 3, if the resource supply reference information indicates that the predicted resource supply for the second time period is VCPU=60, resource supply plan 1 can be "For AZ1, 30 virtual machines of type c3.xlarge.4 are allocated, with a total of 60 VCPUs, that is, virtual machines of type c3.xlarge.4 are preferentially issued through RTc3_4"; resource supply plan 2 can be "For AZ1, 30 virtual machines of type c3.xlarge.2 are allocated, with a total of 60 VCPUs, that is, virtual machines of type c3.xlarge.2 are preferentially issued through RTc3_2". Since the priority corresponding to resource supply plan 1 is 2 and the priority corresponding to resource supply plan 2 is 3, the evaluation result of resource supply plan 1 is 80 points, and the evaluation result of resource supply plan 2 is 90 points. It should be understood that this is only an example and does not constitute a limitation of the present application. In this way, associating the evaluation results of resource provisioning plans with resource allocation rules can improve the rationality of resource provisioning. Furthermore, by determining the evaluation results of each resource provisioning plan, it can provide reference information for relevant personnel, facilitating the formulation of more reasonable resource provisioning plans. For example, the cloud platform outputs a resource management interface that includes multiple resource provisioning plans and the evaluation results of each resource provisioning plan, so that relevant personnel can obtain this information.

[0096] In some embodiments, the cloud platform determines a resource supply plan for the cloud platform based on resource supply reference information and predicted events, where the predicted event refers to an event that may occur in the second time period. For example, if the second time period includes a holiday, then a holiday-related event may occur in the second time period, thereby affecting the resource supply on the cloud platform. Based on the predicted event, the cloud platform can determine the predicted resource supply of the cloud platform in the second time period under the influence of the predicted event in a manner similar to steps 201 to 202, and determine the resource supply plan in combination with the resource supply reference information determined in step 202. This is not limited in this application. In this way, when determining the resource supply plan, not only the impact of the event that has occurred on the resource supply on the cloud platform is fully considered, but also the events that may occur in the future are considered, thereby further improving the accuracy of the resource supply plan. In addition, when determining the resource supply plan, the cloud platform can also combine the user's resource request, that is, the preset resource amount requested by the user, or the user's declared amount. This is not limited in this application.

[0097] After step 203, the cloud platform determines its resource supply plan based on the predicted resource supply reference information. In some embodiments, the cloud platform can also determine supply reminder information based on the resource supply plan and the resources to be allocated by the cloud platform in the second time period, thereby promptly notifying relevant personnel. This process is described below with reference to steps 204 and 205.

[0098] 204. The cloud platform determines supply prompt information of the cloud platform based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period. The supply prompt information indicates a predicted supply status of the resources on the cloud platform in the second time period.

[0099] In the embodiment of the present application, the resources to be allocated on the cloud platform in the second time period are the resources that can be allocated to the user by the cloud platform in the second time period. Schematically, this step includes the following two situations:

[0100] Case 1: If the resources indicated by the resource provisioning plan are greater than the resources to be allocated, the cloud platform's first provisioning prompt is determined. The first provisioning prompt indicates that there is a resource shortage on the cloud platform in the second time period. For example, the resource provisioning plan allocates 10 c3.xlarge.4 VMs with 20 VCPUs for AZ1 in the second time period. The cloud platform's resources to be allocated in the second time period include 5 c3.xlarge.4 VMs with 10 VCPUs. This indicates that the resources indicated by the resource provisioning plan are greater than the resources to be allocated, and there is a resource shortage on the cloud platform in the second time period.

[0101] Case 2: If the resources indicated by the resource provisioning plan are less than or equal to the resources to be allocated, the cloud platform's second provisioning prompt information is determined. The second provisioning prompt information indicates that the cloud platform has sufficient resource supply in the second time period. For example, the resource provisioning plan allocates 10 c3.xlarge.4 VMs with 20 VCPUs for AZ1 in the second time period. The cloud platform's resources to be allocated in the second time period include 20 c3.xlarge.4 VMs with 40 VCPUs. This indicates that the resources indicated by the resource provisioning plan are less than the resources to be allocated, and the cloud platform has sufficient resource supply in the second time period.

[0102] 205. The cloud platform outputs a resource management interface of the cloud platform, which includes provisioning prompt information.

[0103] Among them, by outputting the resource management interface and displaying the resource management interface on the terminal used by the relevant personnel, it is convenient for the relevant personnel to obtain the supply reminder information in a timely manner through the resource management interface, that is, to obtain the predicted supply status of the resources on the cloud platform in the second time period, and provide the relevant personnel with reference information for resource management. In some embodiments, if the supply reminder information is the first supply reminder information, the cloud platform outputs the resource management interface including the supply reminder information, that is, when the cloud platform determines that there is a shortage of resource supply in the second time period, it reminds the relevant personnel by outputting the resource management interface. If the supply reminder information is the second supply reminder information, the cloud platform may not output the resource management interface including the supply reminder information, and this application does not limit this.

[0104] In some embodiments, if the resources indicated by the resource provisioning plan are greater than the resources to be allocated on the cloud platform during the second time period, the cloud platform determines a resource assurance plan for the cloud platform based on the difference between the resources indicated by the resource provisioning plan and the resources to be allocated. The resource assurance plan indicates how to expand the resources on the cloud platform during the second time period. That is, if there is a shortage of resources on the cloud platform during the second time period, the cloud platform can automatically generate a resource assurance plan, making it easier for relevant personnel to adjust resources on the cloud platform based on the resource assurance plan, thereby improving human-computer interaction efficiency. Illustratively, the cloud platform outputs a resource management interface, which includes the resource assurance plan.

[0105] It should be understood that the resource management interface output by the cloud platform may include at least one of a resource supply plan, supply prompt information, and a resource guarantee plan. For example, if the cloud platform determines that there is a shortage of resource supply on the cloud platform in the second time period, the cloud platform may output a resource management interface including the first supply prompt information and a resource guarantee plan, or may output a resource management interface including the first supply prompt information, or may output a resource management interface including a resource guarantee plan, or may output a resource management interface including a resource supply plan, the first supply prompt information, and a resource guarantee plan. That is, the cloud platform can output a resource management interface according to business needs, and this application does not limit the content included in the resource management interface.

[0106] With reference to Figures 4 and 5 below, the method for determining the resource supply plan shown in Figure 2 above is illustrated. Figure 4 is a schematic diagram of the principle of a method for determining a resource supply plan provided by an embodiment of the present application. As shown in Figure 4, the cloud platform obtains reference information on the influence of events that occurred in the first time period based on the events, and based on the resource supply reference information, resource allocation rules, resource status, etc., calls the resource scheduling program to preview the allocation method of resources on the cloud platform in the second time period, and obtains the resource supply plan, supply prompt information, and resource guarantee plan of the cloud platform.

[0107] The following is an example of any event that occurs in the first time period, combined with Figure 5, to illustrate the above process. Figure 5 is a schematic diagram of the framework of a method for determining a resource supply solution provided by an embodiment of the present application. As shown in Figure 5, taking event 1 that occurs in the first time period (taking real-time occurrence as an example) as an earthquake in the AA region, the cloud platform processes the event information of event 1 and obtains the event label of event 1 as "AA region, X month Y day, earthquake, negative", and the influence reference information is "70 points, 14 days". Based on this, sample events A and B associated with event 1 are obtained. The sample event label of sample event A is "AA region, H month L day, football match, negative," the sample influence reference information is "80 points, 12 days," and the sample resource supply information is "the resource supply for the four months is VCPU=10, VCPU=20, VCPU=20, and VCPU=20, respectively." The sample event label of sample event B is "BB region, X month Z day, earthquake, negative," the sample influence reference information is "70 points, 14 days," and the sample resource supply information is "the resource supply for the four months is VCPU=15, VCPU=20, VCPU=20, and VCPU=10, respectively." Based on the relevant information of the sample events, an artificial intelligence model is trained to obtain a resource supply prediction model, such as a multi-linear regression fitting prediction model y=f(K(Ev)). Furthermore, the event label and influence reference information of event 1 are processed through the resource supply prediction model to obtain the resource supply reference information of the cloud platform, and combined with the user's resource request (i.e., the user's declared amount) to determine the resource supply plan for reference by relevant personnel.

[0108] In summary, in the resource supply plan determination method provided in this application, the cloud platform uses a resource supply prediction model to predict the cloud platform's resource supply in the second time period based on the influence reference information of events that occurred in the first time period, thereby obtaining resource supply reference information and determining the cloud platform's resource supply plan. Because this process fully considers the impact of past events on the cloud platform's resource supply in future time periods, the resource supply reference information is highly accurate, thereby improving the accuracy of the resource supply plan.

[0109] Referring to Figure 6 below, the method for determining the resource supply solution provided in this application is illustrated by taking the cloud platform processing the event label and influence reference information of the event through the resource supply prediction model to obtain resource supply reference information as an example.

[0110] Figure 6 is a flow chart of another method for determining a resource provisioning solution provided by an embodiment of the present application. As shown in Figure 6 , the method is executed by a cloud platform and includes the following steps 601 to 606 .

[0111] 601. The cloud platform obtains an event tag of an event occurring in a first time period, where the event tag indicates a feature of the event.

[0112] The process of the cloud platform obtaining the event tag refers to the aforementioned step 202 and will not be repeated here. In some embodiments, the cloud platform outputs a resource management interface to provide management functions for the resources on the cloud platform, and in response to the resource supply prediction operation implemented on the resource management interface, step 601 and the steps after step 601 are executed. The present application does not limit the implementation method of the resource supply prediction operation, for example, it can be implemented through voice or text input, or by triggering the controls on the resource management interface, etc.

[0113] 602. The cloud platform obtains influence reference information of the event, where the influence reference information indicates the impact of the event on resource supply on the cloud platform.

[0114] 603. The cloud platform processes the event label and influence reference information of the event through a resource supply prediction model to obtain resource supply reference information of the cloud platform. The resource supply reference information indicates the predicted resource supply of the cloud platform in the second time period under the influence of the event.

[0115] The implementation of steps 602 and 603 refers to steps 201 to 202 in FIG. 2 , and will not be described in detail here.

[0116] 604. The cloud platform determines a resource supply plan for the cloud platform in the second time period based on the resource supply reference information.

[0117] 605. The cloud platform determines supply prompt information of the cloud platform based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period. The supply prompt information indicates the predicted supply status of the resources on the cloud platform in the second time period.

[0118] 606. The cloud platform outputs a resource management interface including a resource supply plan and supply prompt information. If the supply prompt information indicates that there is a shortage of resources on the cloud platform in the second time period, the resource management interface also includes a resource guarantee plan.

[0119] The implementation of steps 604 to 606 is similar to steps 204 and 205 in FIG. 2 , and will not be described in detail here. In other embodiments, the resource management interface output by the cloud platform further includes at least one of the following: event tags, influence reference information, resource supply reference information, etc., but the present application is not limited thereto.

[0120] The resource management interface output by the cloud platform is introduced below with reference to Figure 7. Figure 7 is a schematic diagram of a resource management interface provided by an embodiment of the present application. As shown in Figure 7, the resource management interface includes a second time period, a resource supply plan, an event label, influence reference information, resource supply reference information, and a resource guarantee plan, etc. (the specific meaning of this part of the content is referred to the aforementioned method embodiment and will not be repeated here). For example, taking the sub-time period of October XX in the second time period as an example, its predicted resource supply is VCPU=10 (that is, the resource supply reference information). The prediction is based on event 1 that occurred in the first time period (such as September XX). Based on this, 10 VCPUs can be allocated to user A, and relevant personnel can adjust it by performing adjustment operations on the resource management interface. In addition, the resource supply reference information can be displayed in the form of a chart. In this way, the predicted resource supply of the cloud platform in the second time period can be clearly and intuitively displayed, thereby improving the efficiency of human-computer interaction. It should be noted that the display style and deployment method of each part of the content on the resource management interface shown in Figure 7 are only for example purposes and do not constitute a limitation of this application. In actual application, the resource management interface can be configured according to needs.

[0121] This embodiment of the present application also provides a device for determining a resource supply solution. Referring to Figure 8 , which is a schematic diagram of the structure of a device for determining a resource supply solution provided in this embodiment of the present application, the device is configured on a cloud platform and includes an acquisition module 801 , a processing module 802 , and a supply solution determination module 803 .

[0122] An acquisition module 801 is configured to acquire influence reference information of an event occurring in a first time period, where the influence reference information indicates the impact of the event on resource provisioning on the cloud platform;

[0123] Processing module 802 is configured to process the influence reference information using a resource supply prediction model to obtain resource supply reference information for the cloud platform, wherein the resource supply prediction model is trained based on an artificial intelligence model, and the resource supply reference information indicates a predicted resource supply amount for the cloud platform in a second time period under the influence of the event, where the second time period is a time period after the first time period.

[0124] The supply plan determination module 803 is used to determine the resource supply plan of the cloud platform in the second time period based on the resource supply reference information.

[0125] In some embodiments, the apparatus further comprises:

[0126] The prompt information determination module is used to determine the supply prompt information of the cloud platform based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period, and the supply prompt information indicates the predicted supply status of the resources on the cloud platform in the second time period.

[0127] In some embodiments, the device further includes a first output module for outputting a resource management interface of the cloud platform, where the resource management interface includes provisioning prompt information.

[0128] In some embodiments, the prompt information determination module is used for any of the following:

[0129] If the resources indicated by the resource supply plan are more than the resources to be allocated, determining first supply prompt information of the cloud platform, the first supply prompt information indicating that there is a shortage of resources on the cloud platform in the second time period;

[0130] If the resources indicated by the resource supply plan are less than or equal to the resources to be allocated, second supply prompt information of the cloud platform is determined, and the second supply prompt information indicates that the resource supply of the cloud platform is sufficient in the second time period.

[0131] In some embodiments, the apparatus further comprises:

[0132] The guarantee plan determination module is used to determine the resource guarantee plan of the cloud platform based on the difference between the resources indicated by the resource supply plan and the resources to be allocated if the resources indicated by the resource supply plan are more than the resources to be allocated. The resource guarantee plan indicates the method of expanding the resources on the cloud platform in the second time period.

[0133] In some embodiments, the device further includes: a second output module for outputting a resource management interface of the cloud platform, where the resource management interface includes a resource guarantee plan.

[0134] In some embodiments, the influence reference information includes first reference information and second reference information, wherein the first reference information indicates the intensity of the impact of the event on the resource supply on the cloud platform, and the second reference information indicates the duration of the impact of the event on the resource supply on the cloud platform.

[0135] In some embodiments, the apparatus further comprises a training module for:

[0136] Obtaining sample events associated with the event, where the sample events refer to events that occurred in a third time period, where the third time period is a time period before the first time period;

[0137] Based on the influence reference information of sample events and the sample resource supply reference information, the artificial intelligence model is trained to obtain a resource supply prediction model.

[0138] In some embodiments, the processing module 802 is configured to process the influence reference information and the event label of the event through a resource supply prediction model to obtain the resource supply reference information, where the event label indicates a feature of the event.

[0139] Acquisition module 801, processing module 802, and supply solution determination module 803 can all be implemented via software or hardware. For example, the following describes the implementation of acquisition module 801, taking acquisition module 801 as an example. Similarly, the implementation of processing module 802 and supply solution determination module 803 can refer to the implementation of acquisition module 801.

[0140] As an example of a software functional unit, the acquisition module 801 may include code running on a computing instance. The computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Furthermore, the computing instance may be one or more. For example, the acquisition module 801 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone (AZ) or in different AZs, each AZ including one data center or multiple geographically close data centers. Typically, a region may include multiple AZs.

[0141] Similarly, multiple hosts / virtual machines / containers running the code can be distributed within the same virtual private cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Cross-region communication between two VPCs within the same region, or between VPCs in different regions, requires a communication gateway within each VPC to interconnect the VPCs.

[0142] As an example of a hardware functional unit, the acquisition module 801 may include at least one computing device, such as a server. Alternatively, the acquisition module 801 may be implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0143] The multiple computing devices included in acquisition module 801 can be distributed in the same region or in different regions. The multiple computing devices included in acquisition module 801 can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in acquisition module 801 can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of servers, ASICs, PLDs, CPLDs, FPGAs, GALs, and other computing devices.

[0144] It should be noted that, in other embodiments, the acquisition module 801 can be used to execute any step in the method for determining the resource supply plan, the processing module 802 can be used to execute any step in the method for determining the resource supply plan, and the supply plan determination module 803 can be used to execute any step in the method for determining the resource supply plan. The steps that the acquisition module 801, the processing module 802 and the supply plan determination module 803 are responsible for implementing can be specified as needed. The full functions of the resource supply plan determination device are realized by respectively implementing different steps in the method for determining the resource supply plan through the acquisition module 801, the processing module 802 and the supply plan determination module 803.

[0145] The present application also provides a computing device 900. Referring to FIG. 9 , FIG. 9 is a schematic diagram of the structure of a computing device provided in an embodiment of the present application. As shown in FIG. 9 , computing device 900 includes a bus 902, a processor 904, a memory 906, and a communication interface 908. Processor 904, memory 906, and communication interface 908 communicate with each other via bus 902. Computing device 900 may be a server or a terminal device. It should be understood that the present application does not limit the number of processors and memories in computing device 900.

[0146] Bus 902 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, among others. Buses may be classified as address buses, data buses, control buses, and the like. For ease of illustration, FIG9 illustrates a single bus line, but this does not imply a single bus or type of bus. Bus 902 may include a path for transmitting information between various components of computing device 900 (e.g., memory 906, processor 904, and communication interface 908).

[0147] The processor 904 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0148] The memory 906 may include volatile memory, such as random access memory (RAM). The processor 904 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0149] Memory 906 stores executable program code. Processor 904 executes the executable program code to implement the functions of the aforementioned acquisition module 801, processing module 802, and supply plan determination module 803, thereby implementing the method for determining a resource supply plan. In other words, memory 906 stores instructions for executing the method for determining a resource supply plan.

[0150] Alternatively, the memory 906 stores executable code, and the processor 904 executes the executable code to implement the functions of the apparatus shown in Figures 8 and 9, respectively, thereby implementing the method for determining a resource provisioning solution. In other words, the memory 906 stores instructions for executing the method for determining a resource provisioning solution.

[0151] The communication interface 908 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 900 and other devices or a communication network.

[0152] Embodiments of the present application also provide a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0153] Figure 10 is a schematic diagram of the structure of a computing device cluster provided in an embodiment of the present application. As shown in Figure 10, the computing device cluster includes at least one computing device 900. The memory 906 of one or more computing devices 900 in the computing device cluster may store the same instructions for executing the method for determining a resource provisioning solution.

[0154] In some possible implementations, the memory 906 of one or more computing devices 900 in the computing device cluster may also store partial instructions for executing the method for determining a resource provisioning solution. In other words, the combination of one or more computing devices 900 may jointly execute the instructions for executing the method for determining a resource provisioning solution.

[0155] It should be noted that the memory 906 in different computing devices 900 in the computing device cluster can store different instructions, each for executing part of the functions of the apparatus shown in FIG8 . In other words, the instructions stored in the memory 906 in different computing devices 900 can implement the functions of one or more of the acquisition module 801, the processing module 802, and the supply solution determination module 803.

[0156] In some possible implementations, one or more computing devices in a computing device cluster may be connected via a network. The network may be a wide area network (WAN) or a local area network (LAN), etc. FIG11 is a schematic diagram of a connection method for a computing device cluster provided in an embodiment of the present application. As shown in FIG11 , two computing devices 900 are connected via a network. Specifically, the connection to the network is achieved via a communication interface 908 in each computing device 900. It should be understood that the functions of the computing device 900 shown in FIG11 may also be performed by multiple computing devices 900.

[0157] Embodiments of the present application also provide a computer program product containing instructions. The computer program product may be software or a program product containing instructions that can be run on a computing device or stored on any available medium. When the computer program product is run on at least one computing device, it causes the at least one computing device to execute the aforementioned method for determining a resource provisioning solution.

[0158] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute a method for determining a resource provisioning solution.

[0159] In this application, the terms "first", "second", etc. are used to distinguish between identical or similar items having substantially the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there a limit on quantity and execution order. It should also be understood that although the following description uses the terms first, second, etc. to describe various elements, these elements should not be limited by the terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the various described examples, the first reference information can be referred to as the second reference information, and similarly, the second reference information can be referred to as the first reference information. The first reference information and the second reference information can both be reference information, and in some cases, can be separate and different reference information.

[0160] In this application, the term "at least one" means one or more, and the term "plurality" means two or more. For example, a plurality of references means two or more references.

[0161] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0162] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of program structure information. The program structure information includes one or more program instructions. When the program instructions are loaded and executed on a computing device, all or part of the processes or functions described in the embodiments of the present application are generated.

[0163] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0164] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above 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 scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining a resource supply plan, characterized in that: Executed by a cloud platform, the method includes: Acquire influence reference information of an event occurring in a first time period, wherein the influence reference information indicates an influence of the event on resource supply on the cloud platform; The influence reference information is processed by a resource supply prediction model to obtain resource supply reference information of the cloud platform, wherein the resource supply prediction model is obtained based on training of an artificial intelligence model, and the resource supply reference information indicates a predicted resource supply of the cloud platform in a second time period under the influence of the event, and the second time period is a time period after the first time period; Based on the resource supply reference information, a resource supply plan for the cloud platform in the second time period is determined.

2. The method according to claim 1, characterized in that The method further comprises: Based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period, supply prompt information of the cloud platform is determined, and the supply prompt information indicates a predicted supply status of the resources on the cloud platform in the second time period.

3. The method according to claim 2, characterized in that The method further comprises: Outputting a resource management interface of the cloud platform, wherein the resource management interface includes the supply prompt information.

4. The method according to claim 2 or 3, characterized in that: The determining, based on the resource supply scheme and the resources to be allocated on the cloud platform in the second time period, the supply prompt information of the cloud platform includes any of the following: If the resources indicated by the resource supply plan are more than the to-be-allocated resources, determining first supply prompt information of the cloud platform, the first supply prompt information indicating that there is a shortage of resources on the cloud platform in the second time period; If the resources indicated by the resource supply plan are less than or equal to the to-be-allocated resources, second supply prompt information of the cloud platform is determined, where the second supply prompt information indicates that the resource supply of the cloud platform is sufficient in the second time period.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: If the resources indicated by the resource supply plan are more than the resources to be allocated on the cloud platform in the second time period, the resource guarantee plan of the cloud platform is determined based on the difference between the resources indicated by the resource supply plan and the resources to be allocated, and the resource guarantee plan indicates a method for expanding the resources on the cloud platform in the second time period.

6. The method according to claim 5, characterized in that The method further comprises: Outputting the resource management interface of the cloud platform, wherein the resource management interface includes the resource guarantee solution.

7. The method according to any one of claims 1 to 6, characterized in that The influence reference information includes first reference information and second reference information, wherein the first reference information indicates the intensity of the influence of the event on the resource supply on the cloud platform, and the second reference information indicates the duration of the influence of the event on the resource supply on the cloud platform.

8. The method according to any one of claims 1 to 7, characterized in that The method further comprises: Acquire a sample event associated with the event, where the sample event refers to an event occurring in a third time period, and the third time period is a time period before the first time period; Based on the influence reference information of the sample event and the sample resource supply reference information, the artificial intelligence model is trained to obtain the resource supply prediction model.

9. The method according to any one of claims 1 to 8, characterized in that The resource supply prediction model is used to process the influence reference information to obtain the resource supply reference information of the cloud platform, including: The influence reference information and the event label of the event are processed by the resource supply prediction model to obtain the resource supply reference information, wherein the event label indicates the characteristics of the event.

10. A device for determining a resource supply plan, characterized in that: Configured on a cloud platform, the device includes: An acquisition module, configured to acquire influence reference information of an event occurring in a first time period, wherein the influence reference information indicates an influence of the event on resource supply on the cloud platform; a processing module, configured to process the influence reference information through a resource supply prediction model to obtain resource supply reference information of the cloud platform, wherein the resource supply prediction model is obtained based on training of an artificial intelligence model, and the resource supply reference information indicates a predicted resource supply of the cloud platform in a second time period under the influence of the event, wherein the second time period is a time period after the first time period; A supply plan determination module is used to determine the resource supply plan of the cloud platform in the second time period based on the resource supply reference information.

11. The device according to claim 10, characterized in that The device also includes: A prompt information determination module is used to determine the supply prompt information of the cloud platform based on the resource supply plan and the resources to be allocated on the cloud platform in the second time period, and the supply prompt information indicates the predicted supply status of the resources on the cloud platform in the second time period.

12. The device according to claim 11, characterized in that The device also includes: The first output module is used to output the resource management interface of the cloud platform, and the resource management interface includes the supply prompt information.

13. The device according to claim 11 or 12, characterized in that The prompt information determination module is used for any of the following: If the resources indicated by the resource supply plan are more than the to-be-allocated resources, determining first supply prompt information of the cloud platform, the first supply prompt information indicating that there is a shortage of resources on the cloud platform in the second time period; If the resources indicated by the resource supply plan are less than or equal to the to-be-allocated resources, second supply prompt information of the cloud platform is determined, where the second supply prompt information indicates that the resource supply of the cloud platform is sufficient in the second time period.

14. The device according to any one of claims 10 to 13, characterized in that The device also includes: A guarantee plan determination module is used to determine the resource guarantee plan of the cloud platform based on the difference between the resources indicated by the resource supply plan and the resources to be allocated if the resources indicated by the resource supply plan are more than the resources to be allocated on the cloud platform in the second time period. The resource guarantee plan indicates a method for expanding the resources on the cloud platform in the second time period.

15. The device according to claim 14, characterized in that The device also includes: The second output module is used to output the resource management interface of the cloud platform, and the resource management interface includes the resource guarantee plan.

16. The device according to any one of claims 10 to 15, characterized in that The influence reference information includes first reference information and second reference information, wherein the first reference information indicates the intensity of the influence of the event on the resource supply on the cloud platform, and the second reference information indicates the duration of the influence of the event on the resource supply on the cloud platform.

17. The device according to any one of claims 1 to 16, characterized in that The device also includes a training module for: Acquire a sample event associated with the event, where the sample event refers to an event occurring in a third time period, and the third time period is a time period before the first time period; Based on the influence reference information of the sample event and the sample resource supply reference information, the artificial intelligence model is trained to obtain the resource supply prediction model.

18. The device according to any one of claims 1 to 17, characterized in that The processing module is used to: The influence reference information and the event label of the event are processed by the resource supply prediction model to obtain the resource supply reference information, wherein the event label indicates the characteristics of the event.

19. A computing device cluster, characterized in that: comprising at least one computing device, each computing device comprising a processor and a memory; The processor of the at least one computing device is configured to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the method for determining a resource provisioning solution according to any one of claims 1 to 9.

20. A computer program product comprising instructions, characterized in that When the instructions are executed by a computing device cluster, the computing device cluster executes the method for determining a resource provisioning solution according to any one of claims 1 to 9.

21. A computer-readable storage medium, characterized in that: The method comprises computer program instructions. When the computer program instructions are executed by a computing device cluster, the computing device cluster executes the method for determining a resource provisioning solution according to any one of claims 1 to 9.

Images