Cloud computing platform management method and system, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a management method and system of a cloud computing platform, and is applied to a server, the method comprises the following steps: obtaining computing resource expected data of the cloud computing platform, the computing resource expected data representing first resource data needed by the cloud computing platform to complete a computing task; screening out at least one target computing power server from a computing power server cluster according to the computing resource expected data; and accessing the target computing power server to the cloud computing platform, so that the cloud computing platform executes the computing task by using the target computing power server. According to the embodiment of the invention, through flexible resource allocation and a dynamic adjustment mechanism, the problems of low resource utilization rate, insufficient resources, poor flexibility and the like are solved, and the resource management efficiency and the overall performance of the cloud computing platform are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing, and in particular to a management method for a cloud computing platform, a management system for a cloud computing platform, an electronic device, and a computer-readable storage medium. Background Art

[0002] In the management scheme of cloud computing platform, the management of computing resources is a key link. The traditional computing resource allocation mode usually faces the problem of fixed resource allocation. That is, in this mode, computing resources (such as CPU, memory, storage, etc.) are usually fixed after being allocated to users, and users can only perform computing and storage operations within the pre-allocated resource range. This fixed resource allocation method has certain limitations, especially in application scenarios with large fluctuations in computing tasks, users may encounter the following problems:

[0003] 1. Low resource utilization: When the user's computing demand is lower than the allocated resources, the fixed allocated resources will be idle, resulting in a decrease in the utilization of computing resources. For cloud computing platforms, this means a waste of resources and increases operating costs.

[0004] 2. Insufficient resources: When computing demand suddenly increases, users may find that the allocated fixed resources are insufficient to cope with the current workload. In this case, users can only solve the problem by requesting additional resources or reallocating resources, but this usually takes a long time and affects business continuity and stability.

[0005] 3. Poor flexibility: The traditional fixed resource allocation model lacks flexibility and cannot dynamically adjust resources to adapt to changes in user needs. This is particularly disadvantageous for some application scenarios with large fluctuations in computing resource requirements (such as data analysis, streaming media processing, etc.).

[0006] In summary, the fixed allocation of computing resources is a major challenge in traditional cloud computing platform management solutions, while modern management solutions are committed to overcoming these problems through flexible resource allocation methods and providing more efficient and scalable computing resource management services. Summary of the invention

[0007] In view of the above problems, embodiments of the present invention are proposed to provide a method for managing a cloud computing platform, a management system for a cloud computing platform, an electronic device, and a computer-readable storage medium that overcome the above problems or at least partially solve the above problems.

[0008] In order to solve the above problems, an embodiment of the present invention discloses a management method for a cloud computing platform, which is applied to a server, and the method includes: obtaining expected computing resource data of the cloud computing platform, the expected computing resource data representing the first resource data required by the cloud computing platform to complete the computing task; screening out at least one target computing power server from a computing power server cluster according to the expected computing resource data; connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task.

[0009] Optionally, the obtaining of the expected computing resource data of the cloud computing platform includes: obtaining the expected computing resource data from the cloud computing platform that originates from a client corresponding to the computing task.

[0010] Optionally, screening out at least one target computing power server from the computing power server cluster based on the expected computing resource data includes: analyzing the expected computing resource data to obtain the first resource data; reading the first configuration information and energy efficiency value of each computing power server in the computing power server cluster; and screening out the target computing power server from the computing power server cluster based on the first resource data, the first configuration information and the energy efficiency value.

[0011] Optionally, the target computing power server is screened out from the computing power server cluster based on the first resource data, the first configuration information and the energy efficiency value, including: detecting from the computing power server cluster multiple candidate computing power servers whose first configuration information conforms to the first resource data; arranging the multiple candidate computing power servers in descending order according to the energy efficiency value; and selecting the candidate computing power server with the highest energy efficiency value among the multiple candidate computing power servers arranged in descending order as the target computing power server.

[0012] Optionally, after connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task, the method also includes: obtaining computing task completion notification information from the cloud computing platform; reading out the server identification information of the target computing power server from the task completion notification information; and removing the target computing power server from the cloud computing platform according to the server identification information.

[0013] Optionally, after connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task, the method also includes: obtaining result information of the computing task; reading storage requirement information from the result information; filtering out the target storage server according to the storage requirement information; connecting the target storage server to the cloud computing platform so that the cloud computing platform stores the task results of the computing task to the target storage server.

[0014] Optionally, the first resource data includes: central processing unit resource data, graphics processing unit resource data, memory resource data, and hard disk resource data.

[0015] An embodiment of the present invention also discloses a management system of a cloud computing platform, which is applied to a server, and the system includes: a resource data acquisition module, which is used to acquire the expected computing resource data of the cloud computing platform, and the expected computing resource data represents the first resource data required by the cloud computing platform to complete the computing task; a computing power server screening module, which is used to screen out at least one target computing power server from the computing power server cluster according to the expected computing resource data; and a computing power server access module, which is used to connect the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task.

[0016] Optionally, the resource data acquisition module is used to obtain the expected computing resource data from the client corresponding to the computing task from the cloud computing platform.

[0017] Optionally, the computing power server screening module includes: a resource data analysis module, used to analyze the expected data of the computing resources to obtain the first resource data; a configuration energy efficiency reading module, used to read the first configuration information and energy efficiency value of each computing power server in the computing power server cluster; a target server determination module, used to screen out the target computing power server from the computing power server cluster based on the first resource data, the first configuration information and the energy efficiency value.

[0018] Optionally, the target server determination module includes: a candidate server detection module, used to detect multiple candidate computing power servers whose first configuration information conforms to the first resource data from the computing power server cluster; a candidate server sorting module, used to arrange the multiple candidate computing power servers in descending order according to the energy efficiency value; and a target server selection module, used to select the candidate computing power server with the highest energy efficiency value from the multiple candidate computing power servers arranged in descending order as the target computing power server.

[0019] Optionally, the system also includes: a completion notification acquisition module, which is used to obtain computing task completion notification information from the cloud computing platform after the computing power server access module connects the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task; a server identification reading module, which is used to read the server identification information of the target computing power server from the task completion notification information; and a target server removal module, which is used to remove the target computing power server from the cloud computing platform according to the server identification information.

[0020] Optionally, the system also includes: a result information acquisition module, used to connect the target computing power server to the cloud computing platform through the computing power server access module, so that the cloud computing platform obtains the result information of the computing task after using the target computing power server to perform the computing task; a storage requirement reading module, used to read the storage requirement information from the result information; a storage server screening module, used to screen out the target storage server according to the storage requirement information; a storage server access module, used to connect the target storage server to the cloud computing platform, so that the cloud computing platform stores the task results of the computing task to the target storage server.

[0021] Optionally, the first resource data includes: central processing unit resource data, graphics processing unit resource data, memory resource data, and hard disk resource data.

[0022] An embodiment of the present invention also discloses an electronic device, comprising: one or more processors; and one or more machine-readable media having instructions stored thereon, which, when executed by the one or more processors, enables the electronic device to execute the management method of the cloud computing platform as described above.

[0023] The embodiment of the present invention further discloses a computer-readable storage medium, which stores a computer program that enables a processor to execute the cloud computing platform management method as described above.

[0024] The embodiments of the present invention include the following advantages:

[0025] The management solution of the cloud computing platform provided in the embodiment of the present invention is applied to a server. The server obtains the computing resource expectation data of the cloud computing platform, and the computing resource expectation data represents the first resource data required by the cloud computing platform to complete the computing task. Then, at least one target computing server is selected from the computing server cluster according to the computing resource expectation data. Then, the target computing server is connected to the cloud computing platform so that the cloud computing platform uses the target computing server to perform the computing task.

[0026] In the embodiment of the present invention, by obtaining the expected data of computing resources and screening the target computing power server according to the data, the cloud computing platform can dynamically allocate and adjust computing resources. This method effectively avoids the idleness of resources, improves the utilization rate of resources, and reduces the waste caused by idle resources. For the cloud computing platform, this not only optimizes resource management, but also reduces operating costs. The traditional fixed allocation mode is prone to insufficient resources when the computing demand suddenly increases, and the embodiment of the present invention can quickly respond to changes in computing resource demand by dynamically screening and accessing the target computing power server. This means that when the amount of computing tasks suddenly increases, the computing capacity can be expanded in time to ensure the continuity and stability of the business, and avoid service interruption or performance degradation caused by insufficient resources. The traditional fixed resource allocation method is abandoned, and a demand-based dynamic resource allocation method is adopted. Allowing the cloud computing platform to flexibly adjust resource configuration according to the actual computing task requirements can better adapt to the fluctuations of computing tasks. This is particularly important for application scenarios with unstable resource requirements (such as data analysis, streaming media processing, etc.), which greatly improves the flexibility of the cloud computing platform. By dynamically selecting servers and accessing target computing servers, cloud computing platforms can continuously expand computing resources as user demand grows without the need for frequent resource reallocation or migration. This scalability not only reduces management complexity, but also ensures the continuous availability and high performance of services, meeting users' growing demand for computing resources.

[0027] In summary, compared with the traditional fixed resource allocation mode, the embodiments of the present invention solve the problems of low resource utilization, insufficient resources, poor flexibility, etc. through flexible resource allocation and dynamic adjustment mechanism, and significantly improve the resource management efficiency and overall performance of the cloud computing platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flowchart of the steps of a method for managing a cloud computing platform according to an embodiment of the present invention;

[0029] Figure 2 It is a schematic diagram of the principle of a video conference management solution based on a scheduling server according to an embodiment of the present invention;

[0030] Figure 3 It is a structural block diagram of a management system of a cloud computing platform according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] The management scheme of the cloud computing platform proposed in the embodiment of the present invention dynamically obtains the expected data of computing resources, and based on the data, selects the target computing server from the computing server cluster and connects it to the cloud computing platform, thereby realizing the flexible allocation and management of computing resources. Compared with the traditional fixed resource allocation mode, this scheme can dynamically adjust resource configuration according to the actual computing task requirements, significantly improve resource utilization, and avoid resource waste. In addition, the scheme also has the ability to quickly respond to changes in computing needs, and can expand computing power in time when resources are insufficient, ensuring business continuity and stability, enhancing the flexibility and scalability of the system, and effectively responding to application scenarios with large fluctuations in computing tasks.

[0033] Reference Figure 1 , shows a flow chart of the steps of a cloud computing platform management method according to an embodiment of the present invention. The cloud computing platform management method can be applied to a server (such as a management server, a scheduling server, etc.), and the cloud computing platform management method can specifically include the following steps:

[0034] Step 101, obtaining the expected data of computing resources of the cloud computing platform.

[0035] In an embodiment of the present invention, the computing resource expectation data represents the first resource data that the cloud computing platform needs to use to complete the computing task. The process of obtaining the computing resource expectation data can be carried out in a variety of ways, mainly including task analysis, historical data analysis, and user demand prediction. Among them, task analysis refers to determining the required first resource by analyzing the characteristics of the current task, the computational complexity, and the amount of data processing. For example, in a video conferencing system, if the upcoming meeting is expected to have a large amount of video streaming data transmission and a large number of participants, then a higher computing resource requirement will be obtained when analyzing the task. Historical data analysis is based on the data of similar tasks in the past, and infers the amount of resources that may be required for the current task. For example, if a video conferencing system has held meetings of similar scales many times, the resource requirements required for the current meeting can be estimated through historical data. User demand prediction is to predict the required first resource based on the meeting information provided by the user, such as the expected number of participants, video quality requirements, meeting time, etc.

[0036] In actual operation, the acquisition of computing resource expectation data is a dynamic process. The cloud computing platform not only needs to obtain resource expectation data before the task starts, but also continuously updates this data during the task execution to cope with sudden changes in demand. For example, during a video conference, if more participants suddenly join, or a higher-resolution video stream needs to be processed, the system will update the resource requirement data in real time to ensure timely adjustment of computing resources.

[0037] To further illustrate, assume that in a video conferencing system, a meeting is expected to have 100 participants, half of whom will use HD video and the other half will use standard definition video. When analyzing the task, based on the number of participants, video resolution, and possible interaction requirements, it is concluded that this meeting requires computing resources of 4 CPU cores, 8GB of memory, and 500GB of storage space. This data is the expected computing resource data, which will serve as the basis for subsequent resource allocation.

[0038] Step 102: select at least one target computing server from the computing server cluster according to the computing resource expectation data.

[0039] In an embodiment of the present invention, a computing power server cluster is usually composed of multiple computing power servers, which may differ in performance, configuration, and current load conditions. When screening the target computing power server, these factors will be comprehensively considered in order to select the most suitable computing resources for the computing task. According to the hardware requirements contained in the expected data of the computing resources, such as the number of CPU cores, memory size, storage capacity, etc., the target computing power servers that meet these hardware requirements are screened out in the computing power server cluster. Secondly, the current load conditions of these computing power servers will also be evaluated to avoid selecting computing power servers with excessive loads, thereby avoiding performance bottlenecks or delays during task execution.

[0040] The screening process also involves matching the configuration of the computing servers to ensure that the special needs of the task are met. For example, in a video conferencing system, if the meeting requires high-definition real-time transcoding operations, computing servers equipped with high-performance graphics processors will be given priority to ensure the efficiency and quality of video processing. In addition, if the meeting needs to store a large amount of recorded video data, computing servers with sufficient storage space and fast read and write speeds will be selected.

[0041] To further illustrate, assume that a video conferencing system needs to conduct an important meeting with an expected 200 participants, which includes the requirements of multiple HD videos and real-time subtitle generation. After obtaining the expected computing resource data for the meeting, the management server selects the computing servers in the computing server cluster that meet these requirements. In the end, a computing server with 8 CPU cores, 32GB of memory, 1TB of high-speed storage space, and a high-performance GPU may be selected as the target computing server. This server can not only meet the computing and storage needs of the meeting, but also efficiently handle real-time transcoding and subtitle generation tasks.

[0042] Step 103, connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform computing tasks.

[0043] In an embodiment of the present invention, connecting the target computing server to the cloud computing platform generally involves multiple technical steps. It is necessary to ensure that the network connection and configuration of the target computing server are compatible with the overall architecture of the cloud computing platform. This may include the allocation of network addresses, security authentication, and the setting of connection protocols. The computing tasks and resource allocation strategies need to be loaded onto the target computing server so that it can perform tasks according to predetermined requirements. This is usually achieved through a scheduler or resource manager, which is responsible for dividing the tasks into executable units and scheduling them according to the resource allocation strategy.

[0044] During the access process, it is also necessary to monitor the operating status of the target computing server to ensure that it can handle tasks normally. For example, key indicators such as CPU utilization, memory usage, and network traffic of the target computing server can be tracked in real time through monitoring tools to promptly identify and resolve possible problems. In addition, in order to ensure the continuity and security of task execution, a failover mechanism is usually set up. If the target computing server fails during task execution, the task can be automatically transferred to other backup servers to ensure the smooth completion of the task.

[0045] To further illustrate, assume that in a video conferencing system, the management server selects a high-performance server to handle an important international conference. First, an IP address is assigned to the server through the configuration management tool, and access permissions and security authentication are set. Subsequently, the relevant tasks of the conference (such as video transcoding, real-time subtitle generation, etc.) are assigned to the server, and the monitoring tool is started to track the operating status of the server. If the server is overloaded or a hardware failure occurs during the conference, the backup server can be immediately started to transfer the tasks to ensure the smooth progress of the conference.

[0046] The management solution of the cloud computing platform provided in the embodiment of the present invention is applied to a server. The server obtains the computing resource expectation data of the cloud computing platform, and the computing resource expectation data represents the first resource data required by the cloud computing platform to complete the computing task. Then, at least one target computing server is selected from the computing server cluster according to the computing resource expectation data. Then, the target computing server is connected to the cloud computing platform so that the cloud computing platform uses the target computing server to perform the computing task.

[0047] In the embodiment of the present invention, by obtaining the expected data of computing resources and screening the target computing power server according to the data, the cloud computing platform can dynamically allocate and adjust computing resources. This method effectively avoids the idleness of resources, improves the utilization rate of resources, and reduces the waste caused by idle resources. For the cloud computing platform, this not only optimizes resource management, but also reduces operating costs. The traditional fixed allocation mode is prone to insufficient resources when the computing demand suddenly increases, and the embodiment of the present invention can quickly respond to changes in computing resource demand by dynamically screening and accessing the target computing power server. This means that when the amount of computing tasks suddenly increases, the computing capacity can be expanded in time to ensure the continuity and stability of the business, and avoid service interruption or performance degradation caused by insufficient resources. The traditional fixed resource allocation method is abandoned, and a demand-based dynamic resource allocation method is adopted. Allowing the cloud computing platform to flexibly adjust resource configuration according to the actual computing task requirements can better adapt to the fluctuations of computing tasks. This is particularly important for application scenarios with unstable resource requirements (such as data analysis, streaming media processing, etc.), which greatly improves the flexibility of the cloud computing platform. By dynamically selecting servers and accessing target computing servers, cloud computing platforms can continuously expand computing resources as user demand grows without the need for frequent resource reallocation or migration. This scalability not only reduces management complexity, but also ensures the continuous availability and high performance of services, meeting users' growing demand for computing resources.

[0048] In summary, compared with the traditional fixed resource allocation mode, the embodiments of the present invention solve the problems of low resource utilization, insufficient resources, poor flexibility, etc. through flexible resource allocation and dynamic adjustment mechanism, and significantly improve the resource management efficiency and overall performance of the cloud computing platform.

[0049] In an exemplary embodiment of the present invention, an implementation method of obtaining the expected computing resource data of the cloud computing platform is to obtain the expected computing resource data from the client corresponding to the computing task from the cloud computing platform. The expected computing resource data can be directly derived from the client corresponding to the computing task, that is, the client submits its resource requirements to the cloud computing platform. Specifically, when a client initiates a computing task, it will submit the expected computing resource data to the cloud computing platform through an interface according to its own task characteristics and requirements. The data generally includes the following: the number of central processing unit (CPU) cores required, memory size, storage capacity, network bandwidth requirements, and other hardware resources that may be involved. For example, in a video conferencing system, the client may submit the number of participants in the meeting, video quality requirements (such as high definition or standard definition), meeting duration, and other functions that may be required (such as real-time subtitles, recording functions, etc.). This information will be used by the cloud computing platform as a basis for computing resource allocation.

[0050] In this implementation, the client directly participates in the process of proposing resource requirements, ensuring that the submitted computing resource expectation data can accurately reflect the actual requirements of the current task. Compared with the cloud computing platform's self-prediction or estimation of resource requirements, this implementation is obviously more flexible and accurate. Especially in application scenarios with variable task requirements or high resource requirements (such as real-time meetings that require large amounts of data processing or video encoding), the computing resource expectation data submitted by the client can help the platform better allocate resources and avoid task execution problems caused by insufficient or excessive resources.

[0051] For example, suppose that in a multinational enterprise video conference, there are expected to be 500 participants, including multiple high-definition conference rooms and the need for real-time subtitle generation. In the preparation stage, the client that initiates the meeting will select the corresponding configuration through the system interface. The system will generate the expected computing resource data based on these configurations and send it to the cloud computing platform. After the cloud computing platform receives this data, the management server connected to the cloud computing platform will screen out the target computing servers that meet the conditions according to the requirements submitted by the client, and allocate them to the video conference. This method not only ensures the smooth progress of the meeting, but also improves the resource utilization efficiency of the platform.

[0052] In the embodiment of the present invention, since the computing resource expectation data comes directly from the client, the management server can accurately match the actual task requirements, thereby avoiding idle or insufficient resources and improving overall resource utilization. The client can adjust the resource requirements at any time according to actual needs, and the management server can flexibly respond to such adjustments to ensure the continuity and stability of the task, especially in scenarios with large fluctuations in demand (such as real-time video processing, live broadcast, etc.). By obtaining resource requirements directly from the client, the management server can respond quickly and allocate appropriate resources, reduce task preparation time and execution delays, and improve user experience.

[0053] In an exemplary embodiment of the present invention, one implementation method of selecting at least one target computing server from a computing server cluster according to computing resource expectation data is to analyze the computing resource expectation data to obtain first resource data; read the first configuration information and energy efficiency value of each computing server in the computing server cluster; and select the target computing server from the computing server cluster based on the first resource data, the first configuration information and the energy efficiency value. In this implementation, when the cloud computing platform receives the computing resource expectation data from the client, the management server connected to the cloud computing platform will analyze these data in detail. These data generally include specific requirements for the number of CPU cores, memory size, storage capacity, and network bandwidth. By analyzing these data, the first resource data required for the task can be extracted. The first resource data not only includes direct resource requirements, but may also involve estimates of task performance requirements, such as processing speed, response time, etc. Next, the management server will read the first configuration information and energy efficiency value of each computing server in the computing server cluster. The first configuration information covers the CPU type, memory capacity, storage space, network interface, and other related hardware parameters of each computing server. The energy efficiency value reflects the energy utilization efficiency of each computing server during operation, which is usually determined by the power consumption and performance ratio of the computing server under a specific workload. With this information, the management server can fully understand the performance and operating cost of each computing server, providing data support for resource allocation decisions.

[0054] Then, based on the first resource data, the first configuration information and the energy efficiency value, the management server starts to screen the target computing servers. During the screening process, priority is given to those computing servers that can meet the requirements of the first resource data, and the most energy-efficient and efficient computing servers are further selected based on the energy efficiency value. For example, if a task requires higher computing power and larger memory, and there are several computing servers in the cluster that can meet this requirement, the computing server with a higher energy efficiency value will be selected to perform the task. This not only ensures the smooth completion of the task, but also effectively reduces energy consumption.

[0055] For example, in a large-scale enterprise's multinational video conference, there may be hundreds of participants, and the conference system needs to process high-definition video streams and generate subtitles in real time. For such a complex task, the management server of the cloud computing platform will analyze the expected computing resource data submitted by the client and obtain the required first resource data. Then, each server in the server cluster is checked, its first configuration information and energy efficiency value are evaluated, and several servers that best meet the conditions are selected. Among these servers, the server with the highest energy efficiency value is selected to undertake the task, thereby meeting the task requirements and optimizing resource utilization.

[0056] The embodiment of the present invention can select the most suitable server to perform the task through comprehensive analysis of the first resource data, the first configuration information and the energy efficiency value, thereby avoiding the problem of over-configuration or under-configuration of resources and improving the overall resource utilization. The introduction of energy efficiency values ​​can give priority to energy-saving and efficient computing servers, which not only helps to reduce energy consumption, but also reduces the operating costs of the platform, especially in large-scale clusters. This saving effect is particularly significant. By accurately matching task requirements with server performance, it can ensure that each computing task is executed in the most suitable environment, thereby improving the execution speed and stability of computing tasks.

[0057] In an exemplary embodiment of the present invention, an implementation method for selecting a target computing power server from a computing power server cluster based on first resource data, first configuration information and energy efficiency value is to detect multiple candidate computing power servers whose first configuration information conforms to the first resource data from the computing power server cluster; arrange the multiple candidate computing power servers in descending order according to the energy efficiency value; and select the candidate computing power server with the highest energy efficiency value from the multiple candidate computing power servers arranged in descending order as the target computing power server.

[0058] In this embodiment, multiple candidate computing servers that meet the requirements of the first resource data are detected from the computing server cluster. The first resource data is usually generated by the cloud computing platform after receiving the task requirements submitted by the client, and reflects the hardware conditions required for task execution, such as the number of cores of the central processing unit, the capacity of the memory, the size of the storage, etc. At the same time, each computing server in the computing server cluster has its own first configuration information, which describes the actual performance and capabilities of the computing server. By comparing the first resource data and the first configuration information, the management server can screen out multiple candidate servers that can meet the task requirements. Next, these candidate computing servers are arranged in descending order according to the energy efficiency value. The energy efficiency value is an indicator of the energy utilization efficiency of the computing server when performing a specific task, and is usually determined by the ratio of the power consumption of the computing server under a certain load to its computing performance. By arranging the energy efficiency values ​​in descending order, those computing servers with higher energy utilization efficiency can be given priority to ensure that energy consumption is reduced as much as possible while meeting the computing requirements. Finally, among the multiple candidate computing servers arranged in descending order, the candidate computing server with the highest energy efficiency value is selected as the final target computing server.

[0059] For example, suppose that in a video conference of a large enterprise, there are multiple conference rooms that need to broadcast high-definition live at the same time. After receiving the meeting requirements, the cloud computing platform generates the corresponding first resource data, including the CPU core requirements, memory capacity, and network bandwidth requirements of each conference room. The management server then selects multiple candidate servers that meet these requirements in the computing power server cluster and arranges them in descending order according to their energy efficiency values. In the end, the server with the highest energy efficiency value is selected to handle the meeting task, thereby ensuring the smooth progress of the meeting while minimizing energy consumption.

[0060] The embodiment of the present invention can select the server with the highest energy utilization efficiency by sorting the energy efficiency values ​​of the candidate servers, thereby maximizing resource utilization while meeting the task requirements. Prioritizing the selection of servers with the highest energy efficiency values ​​helps to reduce the energy consumption of computing tasks, reduce the operating costs of cloud computing platforms, and reduce the impact on the environment, which is particularly important in large-scale data center operations. By selecting a server with a high degree of match between the first resource data and the first configuration information and a high energy efficiency value, the platform can ensure that the task is executed in the most suitable environment, thereby improving the execution efficiency and stability of the task. The introduction of energy efficiency values ​​allows the platform to maintain low energy consumption and operating costs while providing high-performance computing services, which is of great significance for improving the market competitiveness of the platform.

[0061] In an exemplary embodiment of the present invention, after the target computing power server is connected to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform computing tasks, an implementation method is to obtain computing task completion notification information from the cloud computing platform; read the server identification information of the target computing power server from the task completion notification information; and remove the target computing power server from the cloud computing platform according to the server identification information.

[0062] In this implementation, after the target computing server completes the specified computing task, the cloud computing platform will generate a notification message that the computing task is completed. This notification message usually contains all necessary data related to the task, including but not limited to the status of task execution, result data, and identification information of the computing server used by the task. Next, the management server will read the server identification information of the target computing server from this task completion notification message. The server identification information is usually a unique identifier used to distinguish each server in the computing server cluster. By reading this information, the platform can accurately locate the specific computing server that executed the task. Finally, based on the read server identification information, the management server will remove the target computing server from the cloud computing platform. This operation means that the server is no longer involved in any ongoing computing tasks, and the resources it occupies will be released and returned to the computing server cluster, thereby preparing for new computing tasks.

[0063] For example, suppose in a video conferencing system, the recording and processing tasks of a video conference are assigned to a specific computing server. After the conference is over, the cloud computing platform receives a notification that the task is completed. After the management server reads the identification information of the computing server responsible for the task, it immediately removes it from the computing task and reallocates its resources to other video conferences or tasks, thereby improving the efficiency of resource use.

[0064] By releasing the computing power server resources that have completed the task in a timely manner, the embodiment of the present invention enables the platform to quickly allocate its resources to new tasks, avoid idle resources, and thus improve overall resource utilization. The problem of computing power servers continuing to occupy resources after the task is completed is avoided, thereby reducing unnecessary energy consumption and operating costs. The platform is able to dynamically adjust and manage computing power resources, quickly allocate and recycle according to real-time needs, and enhance the flexibility and adaptability of the platform, especially when the task volume fluctuates greatly. Through automated resource recovery and reallocation, the platform can better cope with the challenges brought about by scale expansion, support more users and tasks, and thus improve its scalability.

[0065] In an exemplary embodiment of the present invention, after connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform computing tasks, an implementation method is to obtain result information of the computing task; read storage requirement information from the result information; filter out the target storage server according to the storage requirement information; and connect the target storage server to the cloud computing platform so that the cloud computing platform stores the task results of the computing task to the target storage server.

[0066] In this implementation, when the target computing server completes the specified computing task, the cloud computing platform will generate the result information of the task. This result information contains the output data of the task, as well as various requirements related to the data, including but not limited to the size, type, access frequency and security level of the data. This information will serve as the basis for selecting the storage server in the next step. Next, the management server will read the storage requirement information from the result information. The storage requirement information usually includes the storage space size required for the task result, the data type (such as structured data or unstructured data), the storage access frequency (such as hot data or cold data), and the security requirements (such as whether encrypted storage is required). Through these detailed demand data, the platform can select the most suitable storage server in a targeted manner. Then, based on the storage requirement information, the management server will select the target storage server that meets the conditions in the storage server cluster. For example, if the result data of the computing task is large and needs to be accessed frequently, the management server may select a storage server with high-speed access capability and large storage capacity; for those sensitive data that require high security, the management server will give priority to storage servers with data encryption and access control functions. Finally, after the target storage server is selected, the management server will connect it to the cloud computing platform so that the result data of the computing task can be stored in the target storage server. This operation ensures that the results of the computing task can be properly preserved and can be efficiently accessed and managed according to needs in the future.

[0067] Let's take a specific example: In a video conferencing system, the recorded files of a meeting usually involve multimedia files with large data volumes. Suppose the recorded files of a meeting reach hundreds of GB and may be accessed multiple times in the future for playback or analysis. In this case, the management server may select a storage server with large capacity, high-speed access capability, and support for streaming media processing to store these files by reading the storage demand information.

[0068] The embodiments of the present invention can ensure the high efficiency of data storage and avoid waste of resources and unnecessary storage delays by selecting appropriate storage servers according to actual storage needs. Storage resources are dynamically allocated according to the specific needs of the task, avoiding the problem of low resource utilization caused by the fixed allocation method of storage resources. When sensitive data is involved, by selecting storage servers with encryption functions and strict access control, data security can be improved to prevent unauthorized access and data leakage. Through the automated storage server selection and access process, the storage needs of task results can be responded to quickly, reducing manual intervention, thereby improving business continuity and stability. When processing computing tasks involving large-scale data, storage servers with large capacity and high throughput can be flexibly selected to ensure stable operation even when the amount of data increases dramatically.

[0069] In an exemplary embodiment of the present invention, the first resource data includes: central processing unit resource data, graphics processing unit resource data, memory resource data, and hard disk resource data. Among them, the central processing unit resource data refers to various resource information related to the CPU, including the number of CPU cores, main frequency, load, processing power and other performance indicators. The CPU is the core component for executing computing tasks, and its performance directly affects the execution efficiency and response time of the task. By monitoring and obtaining CPU resource data, the management server can understand the computing power and current load status of each server, so as to effectively optimize when allocating resources. For example, in a video conferencing system, video encoding and decoding are CPU-intensive tasks. The management server needs to obtain the CPU resource data of each server in order to allocate processing tasks to those servers with sufficient computing power and low load, so as to ensure the smoothness of video processing and the stability of the conference. The graphics processing unit resource data involves the relevant resource information of the GPU, such as the number of GPU cores, video memory size, processing power and current load. The GPU is mainly used for graphics rendering and parallel computing, especially when processing complex computing tasks such as image processing, video encoding and decoding, and machine learning. Obtaining GPU resource data can help the platform select a suitable server to process graphics-intensive tasks. For example, in an application that requires real-time video rendering, the management server needs to select a server with a high-performance GPU from the server cluster to meet the needs of high-quality video output. Memory resource data includes information such as the total capacity of the server memory, the amount of memory used, the amount of memory remaining, and the access speed of the memory. Memory is a key component for storing and processing data and has a direct impact on the overall performance of the system. Obtaining memory resource data can help the management server evaluate the memory status of each computing server and make reasonable task allocation. For example, when performing large-scale data analysis, the management server needs to ensure that the selected server has enough memory to store and process a large number of data sets to avoid performance bottlenecks or task failures caused by insufficient memory. Hard disk resource data includes information such as the total capacity, used space, remaining space, and read and write speed of the hard disk. Hard disk resources are crucial for the persistent storage and access speed of data. Obtaining hard disk resource data can help the platform make the right choice in storage-intensive tasks. For example, in a cloud storage service that needs to process a large number of video files, the management server needs to select a hard disk with sufficient storage capacity and high-speed read and write capabilities to ensure fast access and stable storage of video data.

[0070] Based on the above description of the management method embodiment of a cloud computing platform, a video conference management solution based on a scheduling server is introduced below. Figure 2, showing a schematic diagram of the principle of a video conference management solution based on a scheduling server according to an embodiment of the present invention. The video conference management solution based on a scheduling server may involve: a cloud computing platform, a scheduling server (i.e., the above-mentioned management server), a computing server, a disk array, an access switch, and a storage network switch. The cloud computing platform is connected to the visual network. The cloud computing platform and the computing server can communicate through an IP network.

[0071] A video conference management solution based on a scheduling server may specifically include the following steps:

[0072] Step 1: Establish communication links and determine resource requirements

[0073] 1. Establish a computing network cloud computing communication link: Each cloud computing platform is connected to a scheduling server as a container for performance and resource requirements. These scheduling servers are responsible for processing the performance and resource demand data of each computing server.

[0074] 2. Determine the energy efficiency value of the computing server: Calculate the energy efficiency value of each computing server in the computing server cluster to understand its efficiency in processing tasks. The energy efficiency value calculation model will comprehensively consider the performance and power consumption of the server to evaluate its overall effectiveness.

[0075] 3. Determine resource requirements: Determine the resources required by the cloud computing platform, including the resources required by the computing server, such as GPU resources. Resource requirements are used to guide the subsequent resource allocation process.

[0076] Step 2: Determine resource allocation principles

[0077] 1. The scheduling server determines the resource demand and energy efficiency value: The scheduling server will formulate a resource allocation strategy based on the resource demand of the cloud computing platform and the energy efficiency value of each computing server.

[0078] 2. Energy efficiency-first allocation principle: Sort the energy efficiency values ​​of computing servers from high to low, and give priority to computing servers with high energy efficiency values. This can improve overall performance and reduce overall power consumption.

[0079] Step 3: Allocate resources based on energy efficiency priority

[0080] 1. The scheduling server allocates resources: The scheduling server sorts each computing server by energy efficiency value according to the energy efficiency priority allocation principle, and calculates from high to low. The cloud computing platform is connected to computing servers with equivalent performance to meet computing needs.

[0081] Step 4: Addressing Compute and Storage Requirements

[0082] 1. Processing client computing needs: After the cloud computing platform receives the client's computing needs, it calculates the required computing resources based on the demand.

[0083] 2. Scheduling server access to storage server: Based on the computing demand, the scheduling server also needs to access at least one storage server based on the principle of equivalent distribution to meet data storage needs.

[0084] 3. Allocate computing servers by resource demand: Sort the computing servers in the set of computing servers to be allocated from high to low according to the resource demand of the cloud computing platform. Prioritize the allocation of computing servers to the customer with the highest resource demand, followed by the customer with the second highest resource demand, and so on.

[0085] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0086] Reference Figure 3 , shows a structural block diagram of a management system of a cloud computing platform according to an embodiment of the present invention. The management system of the cloud computing platform is applied to a server (such as a management server, a scheduling server, etc.), and the management system of the cloud computing platform may specifically include the following modules.

[0087] The resource data acquisition module 31 is used to acquire the expected computing resource data of the cloud computing platform, wherein the expected computing resource data represents the first resource data required by the cloud computing platform to complete the computing task;

[0088] A computing power server screening module 32, configured to screen out at least one target computing power server from the computing power server cluster according to the computing resource expected data;

[0089] The computing power server access module 33 is used to connect the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task.

[0090] In an exemplary embodiment of the present invention, the resource data acquisition module 31 is used to acquire the expected computing resource data from the client corresponding to the computing task from the cloud computing platform.

[0091] In an exemplary embodiment of the present invention, the computing power server screening module 32 includes:

[0092] A resource data analysis module, configured to analyze the expected data of the computing resources to obtain the first resource data;

[0093] A configuration energy efficiency reading module is used to read the first configuration information and energy efficiency value of each computing power server in the computing power server cluster;

[0094] A target server determination module is used to select the target computing server from the computing server cluster based on the first resource data, the first configuration information and the energy efficiency value.

[0095] In an exemplary embodiment of the present invention, the target server determination module includes:

[0096] A candidate server detection module, used to detect multiple candidate computing servers whose first configuration information conforms to the first resource data from the computing server cluster;

[0097] A candidate server sorting module, used to sort the plurality of candidate computing power servers in descending order according to the energy efficiency values;

[0098] The target server selection module is used to select the candidate computing power server with the highest energy efficiency value as the target computing power server from the multiple candidate computing power servers arranged in descending order.

[0099] In an exemplary embodiment of the present invention, the system further comprises:

[0100] A completion notification acquisition module, used for acquiring computing task completion notification information from the cloud computing platform after the computing power server access module 33 connects the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task;

[0101] A server identification reading module, used to read the server identification information of the target computing power server from the task completion notification information;

[0102] The target server removal module is used to remove the target computing power server from the cloud computing platform according to the server identification information.

[0103] In an exemplary embodiment of the present invention, the system further comprises:

[0104] A result information acquisition module, used for connecting the target computing power server to the cloud computing platform by the computing power server access module 33, so that the cloud computing platform can obtain the result information of the computing task after using the target computing power server to perform the computing task;

[0105] A storage requirement reading module, used to read the storage requirement information from the result information;

[0106] A storage server screening module, used to screen out target storage servers according to the storage demand information;

[0107] The storage server access module is used to connect the target storage server to the cloud computing platform so that the cloud computing platform stores the task results of the computing task in the target storage server.

[0108] In an exemplary embodiment of the present invention, the first resource data includes: central processing unit resource data, graphics processing unit resource data, memory resource data, and hard disk resource data.

[0109] As for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0110] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0111] It will be appreciated by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0112] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0113] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0114] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0115] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0116] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0117] The above is a detailed introduction to a management method for a cloud computing platform and a management system for a cloud computing platform provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for managing a cloud computing platform, characterized in that: Applied to a server, the method comprises: Acquire computing resource expectation data of the cloud computing platform, where the computing resource expectation data represents first resource data required by the cloud computing platform to complete a computing task; Filtering at least one target computing server from the computing server cluster according to the computing resource expectation data; The target computing power server is connected to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task.

2. The method according to claim 1, characterized in that The obtaining of the expected data of computing resources of the cloud computing platform includes: The computing resource expectation data originating from the client corresponding to the computing task is obtained from the cloud computing platform.

3. The method according to claim 1, characterized in that The step of selecting at least one target computing server from the computing server cluster according to the computing resource expectation data includes: Analyzing the computing resource expected data to obtain the first resource data; Reading first configuration information and energy efficiency value of each computing power server in the computing power server cluster; The target computing server is selected from the computing server cluster based on the first resource data, the first configuration information and the energy efficiency value.

4. The method according to claim 3, characterized in that The step of selecting the target computing server from the computing server cluster according to the first resource data, the first configuration information and the energy efficiency value includes: Detecting from the computing power server cluster a plurality of candidate computing power servers whose first configuration information conforms to the first resource data; Arrange the plurality of candidate computing power servers in descending order according to the energy efficiency values; Among the multiple candidate computing power servers arranged in descending order, the candidate computing power server with the highest energy efficiency value is selected as the target computing power server.

5. The method according to claim 1, characterized in that After connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task, the method further includes: Obtain computing task completion notification information from the cloud computing platform; Reading the server identification information of the target computing power server from the task completion notification information; The target computing power server is removed from the cloud computing platform according to the server identification information.

6. The method according to claim 1, characterized in that After connecting the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task, the method further includes: Obtaining result information of the computing task; Reading storage requirement information from the result information; Filter out a target storage server according to the storage requirement information; The target storage server is connected to the cloud computing platform so that the cloud computing platform stores the task result of the computing task in the target storage server.

7. The method according to claim 1, characterized in that The first resource data includes: central processing unit resource data, graphics processing unit resource data, memory resource data, and hard disk resource data.

8. A management system for a cloud computing platform, characterized in that: Applied to a server, the system comprises: A resource data acquisition module, used to acquire the expected computing resource data of the cloud computing platform, wherein the expected computing resource data represents the first resource data required by the cloud computing platform to complete the computing task; A computing power server screening module, used to screen out at least one target computing power server from the computing power server cluster according to the computing resource expected data; The computing power server access module is used to connect the target computing power server to the cloud computing platform so that the cloud computing platform uses the target computing power server to perform the computing task.

9. An electronic device, characterized in that: include: one or more processors; and One or more machine-readable media having instructions stored thereon, when executed by the one or more processors, enable the electronic device to execute the cloud computing platform management method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer program stored therein enables the processor to execute the management method of the cloud computing platform as described in any one of claims 1 to 7.

Images