Enterprise resource management platform based on cloud computing

Abstract

The invention discloses an enterprise resource management platform based on cloud computing, which comprises a resource class dimension management terminal, a resource scheduling terminal and a resource configuration optimization terminal, and discloses a specific operation method of the platform, and the specific operation method comprises the following steps: step 1, carrying out multi-dimensional class management on virtual resources by utilizing a resource class management model; 2, formatting the user request, sending the formatted user request to a scheduling manager, performing calculation by the scheduling manager through a scheduling algorithm according to an existing resource use condition obtained by node state monitoring, and sending a final scheduling calculation result to a task manager; 3, the enterprise carries out resource allocation by coordinating sub-enterprises in different regions in cross-region resource allocation; 4, coordinating related departments to upgrade and optimize resource configuration according to resource configuration conditions reflected by the resource management terminals; the method has the characteristics of improving the resource utilization efficiency and optimizing the resource structure configuration.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual resource management, and in particular to an enterprise resource management platform based on cloud computing. Background Art

[0002] Virtualization technology is a comprehensive management application technology for virtual resources. It is a reasonable split and logical recombination of underlying physical resources. Resource splitting splits a single physical resource into multiple virtual resources to achieve the goal of making better use of resources, such as operating system virtualization, CPU virtualization, etc.; resource combination is to combine multiple physical resources at a logical level to form a more powerful virtual resource, such as storage virtualization, server virtualization, etc. Virtualization technology can break the physical resource isolation through this logical resource splitting and combination, aiming to maximize the use of physical hardware resources, effectively reducing the complexity of resource management, and enabling enterprises to customize services according to user needs to provide users with a better user experience. However, in the actual process of user configuration of resource pools, users have no way of knowing which algorithm to configure for the current resource pool to achieve better performance. The observation period of resource pool performance is long. If users directly configure service algorithms and make decisions after observing the performance of each algorithm in the current resource pool one by one, it will not only increase the operating costs of the enterprise cloud data center and the consumption costs of resource configuration, but also waste the enterprise's resources. Therefore, it is necessary to design an enterprise resource management platform based on cloud computing to improve resource utilization efficiency and optimize resource structure configuration. Summary of the invention

[0003] The purpose of the present invention is to provide an enterprise resource management platform based on cloud computing to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a cloud computing-based enterprise resource management platform, including a resource class dimension management terminal, a resource scheduling terminal and a resource configuration optimization terminal, wherein the resource class dimension management terminal is used to perform multi-dimensional class management of virtual resources using a resource class management model; the resource scheduling terminal is used to perform calculation scheduling based on the existing resource usage obtained by node status monitoring through the resource scheduling management model; the resource configuration optimization terminal is used to coordinate relevant departments to upgrade and optimize resource configuration through the resource configuration status reflected by each resource management terminal.

[0005] According to the above technical solution, the operation method of the platform includes the following steps: Step 1: Use the resource class management model to perform multi-dimensional class management of virtual resources; Step 2: Format the user request and send it to the scheduling manager. The scheduling manager uses the scheduling algorithm to calculate the existing resource usage obtained by node status monitoring, and sends the final scheduling calculation result to the task manager. The task manager creates a virtual machine for the server cluster and receives the virtual machine creation result. Step 3: In terms of cross-regional resource allocation, the enterprise coordinates resources among subsidiaries in different regions; Step 4: Through resource management, the resource allocation status reflected by each terminal is coordinated with relevant departments to upgrade and optimize resource allocation.

[0006] According to the above technical solution, the step of performing multi-dimensional class management includes: The resource class management model is used to classify the resource layer, and two parts are obtained: the physical machine resource layer and the virtual machine resource layer. On the underlying physical machine of the data center, the virtual machine resource monitoring terminal collects the CPU cache and memory access behavior of each virtual machine, deploys a monitoring agent through the monitoring node of each monitoring terminal, collects the monitoring information of each node and sends it to the monitoring server, and listens to the commands sent by the monitoring server and executes corresponding operations. The monitoring server receives the monitoring information sent by each monitoring agent, including the overall resource information and resource utilization information of the physical machine and virtual machine where the monitoring agent is located, analyzes and processes the sent information, and stores it in the database for classification and summary; Utilize the virtual resources of CPU, memory, network and storage on a single physical node to map the virtual resources of a single node to the physical resources, and encode the mapping relationship between the physical machine and the virtual machine. When there are m virtual machine requests, the m virtual machine requests are allocated to n physical servers, and each virtual machine request is placed on the physical machine with the corresponding serial number. The virtual machine and the server running the virtual machine are respectively used as the virtual request and the physical resource in the resource class management model are multidimensional. Each dimension represents any one of the resources including CPU, memory and network bandwidth. The dimension size of the virtual request represents its corresponding resource demand, and the dimension size of the physical resource represents its corresponding resource capacity. The dimension sizes of the virtual request and the physical resource are used as random variables. When there are N virtual machines that need to be installed in the server and the resource dimension is K, let the random variable Indicates the demand for the yth resource of the ith virtual machine. The total amount of resources in each dimension in the physical machine is greater than the sum of the resources of this dimension of all virtual machines on the physical machine.

[0007] According to the above technical solution, the step of calculating the existing resource usage obtained by the scheduling manager according to the node status monitoring by the scheduling algorithm includes: The user submits a request to use a virtual machine to the resource scheduling terminal. After receiving the virtual machine task request submitted by the user, the scheduling terminal matches the physical server according to the requirements of the task request and the resource usage of the physical server in the current data center. Based on the virtual machines and servers of given dimensions, given n dimensions of virtual machines, the dimension size of the i-th virtual machine is , an analytical scheduling model is constructed for virtual resource scheduling. The model includes a scheduling center and a data center. There is a central task cache queue in the task cache of the scheduling center. When a virtual request is made in the system to create a new virtual machine in a physical machine, each request is constructed as a resource task, and these tasks are stored in the central task cache queue. The scheduling center assigns tasks to physical machines, where all physical machines form a physical machine cluster. The resource quantity dimension on the physical machine is fixed. The number of resources in a host corresponds to n different virtual machine configuration schemes. When the operating system in the virtual machine accesses key system resources, the virtual machine resource monitoring terminal performs corresponding simulation processing according to the triggered request and returns the simulation results to the specified client system; Task management is responsible for facing users externally and server clusters internally, receiving users' virtual machine creation requests and feeding back creation results; formatting user requests and sending them to the scheduling manager, which uses the scheduling algorithm to calculate the existing resource usage based on node status monitoring, deploys the corresponding requested virtual machines to the physical server for user use, and sends the final scheduling calculation results to the task manager, which creates virtual machines for the server cluster and feeds back the virtual machine creation results.

[0008] According to the above technical solution, the step of allocating resources includes: Use virtual machines to randomly send resource allocation requests to servers. Each resource allocation request task remains independent of each other. The enterprise private cloud data center is equipped with a certain amount of physical servers. Multiple virtual machines are deployed on each physical machine. The resource allocation request tasks between virtual machines are executed separately. The resource scheduling platform of the enterprise private cloud data uses the physical machines in multiple servers to provide resource allocation services for each virtual machine. When a resource configuration task is triggered, it determines whether the virtual machine in the server can provide the resources required to run the task. If the resources do not meet the requirements, the triggered resource allocation request will be rejected and the task request will enter the queue. When the virtual machine in the server has idle resources, it will provide resource allocation services for the triggered resource allocation request in sequence. After the triggered resource request is executed, the virtual machine reclaims the allocated resources. The response process of resource allocation is as follows: when allocating resources, each resource management terminal in the enterprise checks whether the resource project exists. If the project does not exist, it will be marked for access to the resource project. If the resource project exists, it will be checked whether the resource quota is allocated under the resource project. If the resource quota exists, the quota will be pre-deducted. If it does not exist, the access mark will be configured. After the resource quota is pre-deducted, the remaining quota after deduction will be returned, and the remaining quota will be checked simultaneously to see if it is sufficient. If it is sufficient, the resource will be allocated. Otherwise, the deducted quota will be returned. If the resource allocation is successful, the quota record of this operation will be written into the database. When allocating a virtual IP, first verify the input parameters, check whether the triggered resource items, resource quantities, and resource quotas match. When these resources have corresponding matching resource registrations in the platform, create a virtual IP resource in the underlying environment, store the virtual IP data returned by the underlying environment, and associate the virtual IP with the currently triggered resource request to further determine whether the virtual IP has any sub-resources. If there are no related sub-resources, delete the virtual IP at the underlying level and recycle it.

[0009] According to the above technical solution, the step of optimizing resource allocation includes: Based on past experience, the maximum and minimum thresholds of a resource are set. When the resource utilization rate is less than the set minimum threshold, the physical machine is triggered to sleep. When the resource utilization rate is greater than the set maximum threshold of the resource, no new virtual machines are deployed on the physical machine. The operations for matching the application service requirements with the virtual machine resources are as follows: deploy the virtual machine on the physical machine, build a resource allocation optimization model, use the model to initialize the parameter values ​​of the physical machine and the resource values ​​configured for the deployed virtual machine, calculate the cost to be paid to the physical machine when obtaining resource allocation, and calculate the expected price of the physical machine. Within the specified resource allocation time period, the physical machine calculates the expected cost of providing unit resource configuration, and updates the resource configuration value based on the calculated payment cost and the expected cost, so as to maximize the server utility in the cloud environment within the limited physical machine resource constraints.

[0010] According to the above technical solution, the step of optimizing resource allocation also includes: Step 41: Establish a load prediction model by processing and analyzing historical host load data of the cloud platform data center including host CPU, memory, host disk and network bandwidth; Step 42: According to the currently input host load data, a prediction algorithm is used to predict the host load value of the next stage, and the predicted value is used to schedule the resources of the cloud data center; Step 43: Establish a resource scheduling model based on the load prediction result, the resource demand of the mobile terminal in the mobile cloud computing platform, and the resource ownership of the cloud server; Step 44: Based on the result of resource scheduling, the enterprise's resource management platform system optimizes the configuration of virtual machine resources according to the resource scheduling plan.

[0011] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention uses a resource class dimension management terminal, a resource scheduling terminal and a resource configuration optimization terminal to perform multi-dimensional class management of virtual resources using a resource class management model, and formats user requests and sends them to a scheduling manager, which calculates the existing resource usage obtained by node status monitoring using a scheduling algorithm, and sends the final scheduling calculation result to a task manager to coordinate resource allocation for subsidiaries in different regions; through the resource configuration status reflected by each resource management terminal, the relevant departments are coordinated to upgrade and optimize the resource configuration, thereby improving resource utilization efficiency and optimizing the configuration of virtual resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 A flow chart of an operation method of an enterprise resource management platform based on cloud computing provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the module composition of the cloud computing-based enterprise resource management platform provided in the second embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] Embodiment 1: Figure 1 This is a flow chart of an operation method of an enterprise resource management platform based on cloud computing provided in the first embodiment of the present invention. This embodiment can be applied to the scenario of enterprise virtualization resource management configuration optimization. This method can be executed by the enterprise resource management platform based on cloud computing provided in this embodiment. Figure 1 As shown, the method specifically comprises the following steps: Step 1: Use the resource class management model to perform multi-dimensional class management of virtual resources; In the embodiment of the present invention, the resource class management model is used to classify the resource layer, and two parts, the physical machine resource layer and the virtual machine resource layer, are obtained. On the underlying physical machine of the data center, the virtual machine resource monitoring terminal collects the CPU cache and memory access behavior of each virtual machine, deploys a monitoring agent through the monitoring node of each monitoring terminal, collects the monitoring information of each node and sends it to the monitoring server, and listens to the commands sent by the monitoring server and executes corresponding operations, wherein the monitoring server receives the monitoring information sent by each monitoring agent, including the overall resource information and resource utilization information of the physical machine and virtual machine where the monitoring agent is located, analyzes and processes the sent information, and stores it in the database for classification and aggregation; Exemplarily, the virtual resources of the CPU, memory, network, and storage on a single physical node are used to map the virtual resources of a single node to the physical resources, and the mapping relationship between the physical machine and the virtual machine is encoded. When there are m virtual machine requests, the m virtual machine requests are allocated to n physical servers, and each virtual machine request is placed on a physical machine with a corresponding serial number. The virtual machine and the server running the virtual machine are respectively used as virtual requests and physical resources in the resource class management model. They are multidimensional, and each dimension represents any one of the resources including CPU, memory, and network bandwidth. The dimensional size of the virtual request represents its corresponding resource demand, and the dimensional size of the physical resource represents its corresponding resource capacity. The dimensional sizes of the virtual request and the physical resource are used as random variables. When there are N virtual machines that need to be installed in the server and the resource dimension is K, let the random variable Indicates the demand for the yth resource of the ith virtual machine. The total amount of resources in each dimension in the physical machine is greater than the sum of the resources of this dimension of all virtual machines on the physical machine.

[0015] Step 2: Format the user request and send it to the scheduling manager. The scheduling manager uses the scheduling algorithm to calculate the existing resource usage obtained by node status monitoring, and sends the final scheduling calculation result to the task manager. The task manager creates a virtual machine for the server cluster and receives the virtual machine creation result. In the embodiment of the present invention, a user submits a request to use a virtual machine to a resource scheduling terminal. After receiving the virtual machine task request submitted by the user, the scheduling terminal matches the physical server according to the requirements of the task request and the resource usage of the physical server in the current data center. Based on the virtual machines and servers of given dimensions, given n dimensions of virtual machines, the dimension size of the i-th virtual machine is , an analytical scheduling model is constructed for virtual resource scheduling. The model includes a scheduling center and a data center. There is a central task cache queue in the task cache of the scheduling center. When a virtual request is made in the system to create a new virtual machine in a physical machine, each request is constructed as a resource task, and these tasks are stored in the central task cache queue. The scheduling center assigns tasks to physical machines, where all physical machines form a physical machine cluster. The resource quantity dimension on the physical machine is fixed. The number of resources in a host corresponds to n different virtual machine configuration schemes. When the operating system in the virtual machine accesses key system resources, the virtual machine resource monitoring terminal performs corresponding simulation processing according to the triggered request and returns the simulation results to the specified client system; Exemplarily, task management is responsible for facing users externally and server clusters internally, receiving users' virtual machine creation requests and feeding back creation results; formatting user requests and sending them to the scheduling manager, which uses the scheduling algorithm to calculate the existing resource usage based on node status monitoring, deploys the virtual machines corresponding to the requests to the physical server for users to use, sends the final scheduling calculation results to the task manager, and the task manager creates virtual machines for the server cluster and feeds back the virtual machine creation results.

[0016] Step 3: In terms of cross-regional resource allocation, the enterprise coordinates resources among subsidiaries in different regions; In the embodiment of the present invention, a virtual machine is used to randomly send a resource allocation request to a server, and each resource allocation request task remains independent of each other. The enterprise private cloud data center is equipped with a certain amount of physical servers, and multiple virtual machines are deployed on each physical machine. The resource allocation request tasks between virtual machines are executed separately. The resource scheduling platform of the enterprise private cloud data performs resource allocation services for each virtual machine through the physical machines in multiple servers. When a resource configuration task is triggered, it is determined whether the virtual machine in the server can provide the resources required to run the task. When the resources do not meet the requirements, the triggered resource allocation request will be rejected, and the task request will enter the queue. When the virtual machine in the server has idle resources, it will provide resource allocation services for the triggered resource allocation request in sequence. After the triggered resource request is executed, the virtual machine reclaims the allocated resources. Exemplarily, the reaction process of resource configuration is as follows: when allocating resources, each resource management terminal in the enterprise checks whether the resource project exists. If the project does not exist, the access mark of the resource project is performed. If the resource project exists, the quota of the resource is pre-deducted. If it does not exist, the access mark is configured. After the resource quota is pre-deducted, the remaining quota after deduction is returned, and the remaining quota is checked simultaneously to see whether it is sufficient. If it is sufficient, the resource is allocated. Otherwise, the deducted quota is returned. If the resource allocation is successful, the quota record of this operation is written into the database. Exemplarily, when allocating a virtual IP, the input parameters are first checked to see whether the triggered resource items, resource quantities and resource quotas match. When these resources have corresponding matching resource registrations in the platform, a virtual IP resource is created in the underlying environment, the virtual IP data returned by the underlying environment is stored in the database, and the virtual IP and the currently triggered resource request are associated to further determine whether the virtual IP has any sub-resources. If there are no related sub-resources, the virtual IP is deleted and recycled at the underlying environment.

[0017] Step 4: Through resource management, the resource allocation of each terminal is reflected, and relevant departments are coordinated to upgrade and optimize resource allocation; In an embodiment of the present invention, the maximum and minimum thresholds of a resource are set based on past experience. When the resource usage rate is less than the set minimum threshold, the dormant state of the physical machine is triggered. When the resource usage rate is greater than the set maximum threshold of the resource, no new virtual machine is deployed on the physical machine. The operation of matching the application service demand with the virtual machine resources is as follows: deploying the virtual machine on the physical machine, building a resource allocation optimization model, using the model to initialize the parameter values ​​of the physical machine and the resource values ​​configured for the deployed virtual machine, calculating the cost to be paid to the physical machine when obtaining resource allocation, and calculating the expected price of the physical machine. Within the specified resource allocation time period, the physical machine calculates the estimated cost of providing unit resource configuration, and updates the resource configuration value based on the calculated payment cost and the estimated cost. The method for optimizing the allocation of physical machine resources within the limited physical machine resource limit to maximize the server utility in the cloud environment is as follows: Step 41: Establish a load prediction model by processing and analyzing historical host load data of the cloud platform data center including host CPU, memory, host disk and network bandwidth; Step 42: According to the currently input host load data, a prediction algorithm is used to predict the host load value of the next stage, and the predicted value is used to schedule the resources of the cloud data center; Step 43: Establish a resource scheduling model based on the load prediction result, the resource demand of the mobile terminal in the mobile cloud computing platform, and the resource ownership of the cloud server; Step 44: Based on the result of resource scheduling, the enterprise's resource management platform system optimizes the configuration of virtual machine resources according to the resource scheduling plan.

[0018] Embodiment 2: Embodiment 2 of the present invention provides an enterprise resource management platform based on cloud computing. Figure 2 A schematic diagram of the module composition of the cloud computing-based enterprise resource management platform provided in the second embodiment of the present invention is shown in FIG. Figure 2 As shown, the platform includes: Resource class dimension management terminal, used to perform multi-dimensional class management of virtual resources using resource class management model; The resource scheduling terminal is used to perform computational scheduling based on the existing resource usage obtained through node status monitoring through the resource scheduling management model; The resource allocation optimization terminal is used to coordinate relevant departments to upgrade and optimize resource allocation through the resource management terminals to reflect the resource allocation status.

[0019] In some embodiments of the present invention, the resource class dimension management terminal includes: Resource class management model module, used to classify resource layers using resource class management model; A virtual resource mapping module is used to map the virtual resources of a single node to physical resources; The resource dimension module is used to treat the virtual machine and the server running the virtual machine as multi-dimensional virtual requests and physical resources in the resource class management model.

[0020] In some embodiments of the present invention, the resource scheduling terminal includes: A virtual machine task request module is used to schedule a terminal to receive a virtual machine task request submitted by a user; A resource task building module is used to build each allocation request for virtual resources into a resource task; A resource allocation service module, used to utilize a virtual machine to randomly allocate resource requests to a server; The resource configuration response module is used to respond to resource configuration status.

[0021] In some embodiments of the present invention, the resource configuration optimization terminal includes: Resource configuration threshold module, used to set the maximum and minimum thresholds of resources based on empirical data; A resource value initialization module is used to initialize the parameter values ​​of the physical machine and the resource values ​​configured for the deployed virtual machine using the model; The resource configuration value updating module is used to update the resource configuration value according to the calculated payment cost and the estimated cost.

[0022] 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 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 device.

[0023] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A cloud computing-based enterprise resource management platform, comprising a resource dimension management terminal, a resource scheduling terminal and a resource configuration optimization terminal, characterized in that: The resource class dimension management terminal is used to perform multi-dimensional class management of virtual resources using a resource class management model; the resource scheduling terminal is used to perform calculation scheduling based on the existing resource usage obtained by node status monitoring through the resource scheduling management model; the resource configuration optimization terminal is used to coordinate relevant departments to upgrade and optimize resource configuration through the resource configuration reflected by each resource management terminal; The resource class dimension management terminal includes: a resource class management model module, a virtual resource mapping module, and a resource dimension module. The resource class management model module is used to classify resource layers using the resource class management model; the virtual resource mapping module is used to map the virtual resources of a single node to physical resources; the resource dimension module is used to use the virtual machine and the server running the virtual machine as multi-dimensional virtual requests and physical resources in the resource class management model respectively; The resource scheduling terminal includes: a virtual machine task request module, a resource task construction module, a resource allocation service module and a resource configuration response module. The virtual machine task request module is used for the scheduling terminal to receive virtual machine task requests submitted by users; the resource task construction module is used to construct each allocation request for virtual resources into a resource task; the resource allocation service module is used to use virtual machines to randomly execute resource allocation requests to servers; the resource configuration response module is used to respond to resource configuration conditions; The resource configuration optimization terminal includes: a resource configuration threshold module, a resource value initialization module and a resource configuration value update module. The resource configuration threshold module is used to set the maximum and minimum thresholds of resources according to empirical data; the resource value initialization module is used to initialize the resource values ​​configured for the parameter values ​​of the physical machine and the deployed virtual machine using the model; the resource configuration value update module is used to update the resource configuration value according to the calculated payment cost and the estimated cost; The operation method of the enterprise resource management platform based on cloud computing includes the following steps: Step 1: Use the resource class management model to perform multi-dimensional class management of virtual resources; Step 2: Format the user request and send it to the scheduling manager. The scheduling manager uses the scheduling algorithm to calculate the existing resource usage obtained by node status monitoring, and sends the final scheduling calculation result to the task manager. The task manager creates a virtual machine for the server cluster and receives the virtual machine creation result. Step 3: In terms of cross-regional resource allocation, the enterprise coordinates resources among subsidiaries in different regions; Step 4: Through resource management, the resource allocation of each terminal is reflected, and relevant departments are coordinated to upgrade and optimize resource allocation; The steps of performing multi-dimensional class management include: The resource class management model is used to classify the resource layer, and two parts are obtained: the physical machine resource layer and the virtual machine resource layer. On the underlying physical machine of the data center, the virtual machine resource monitoring terminal collects the CPU cache and memory access behavior of each virtual machine, deploys a monitoring agent through the monitoring node of each monitoring terminal, collects the monitoring information of each node and sends it to the monitoring server, and listens to the commands sent by the monitoring server and executes corresponding operations. The monitoring server receives the monitoring information sent by each monitoring agent, including the overall resource information and resource utilization information of the physical machine and virtual machine where the monitoring agent is located, analyzes and processes the sent information, and stores it in the database for classification and summary; Utilize the virtual resources of CPU, memory, network and storage on a single physical node to map the virtual resources of a single node to the physical resources, encode the mapping relationship between the physical machine and the virtual machine, and when there are m virtual machine requests, allocate the m virtual machine requests to n physical servers, and place each virtual machine request on the physical machine with the corresponding serial number. The virtual machine and the server running the virtual machine are respectively used as the virtual request and the physical resource in the resource class management model are multidimensional, and each dimension represents any one of the resources including CPU, memory and network bandwidth. The dimension size of the virtual request represents its corresponding resource demand, and the dimension size of the physical resource represents its corresponding resource capacity. The dimension sizes of the virtual request and the physical resource are used as random variables. When there are N virtual machines that need to be installed in the server and the resource dimension is K, let the random variable Indicates the demand for the yth resource of the ith virtual machine. The total amount of resources in each dimension in the physical machine is greater than the sum of the resources of this dimension of all virtual machines on the physical machine. The step of calculating the existing resource usage obtained by the scheduling manager according to the node status monitoring by the scheduling algorithm includes: The user submits a request to use a virtual machine to the resource scheduling terminal. After receiving the virtual machine task request submitted by the user, the scheduling terminal matches the physical server according to the requirements of the task request and the resource usage of the physical server in the current data center. Based on the virtual machines and servers of given dimensions, given n dimensions of virtual machines, the dimension size of the i-th virtual machine is , an analytical scheduling model is constructed for virtual resource scheduling. The model includes a scheduling center and a data center. There is a central task cache queue in the task cache of the scheduling center. When a virtual request is made in the system to create a new virtual machine in a physical machine, each request is constructed as a resource task, and these tasks are stored in the central task cache queue. The scheduling center assigns tasks to physical machines, where all physical machines form a physical machine cluster. The resource quantity dimension on the physical machine is fixed. The number of resources in a host corresponds to n different virtual machine configuration schemes. When the operating system in the virtual machine accesses key system resources, the virtual machine resource monitoring terminal performs corresponding simulation processing according to the triggered request and returns the simulation results to the specified client system; Task management is responsible for facing users externally and server clusters internally, receiving users' virtual machine creation requests and feeding back creation results; formatting user requests and sending them to the scheduling manager, which uses the scheduling algorithm to calculate the existing resource usage obtained by node status monitoring, deploys the corresponding virtual machine to the physical server for user use, and sends the final scheduling calculation results to the task manager, which creates virtual machines for the server cluster and feeds back the virtual machine creation results; The step of optimizing resource configuration includes: Based on past experience, the maximum and minimum thresholds of a resource are set. When the resource utilization rate is less than the set minimum threshold, the physical machine is triggered to sleep. When the resource utilization rate is greater than the set maximum threshold of the resource, no new virtual machines are deployed on the physical machine. The operations for matching the application service requirements with the virtual machine resources are as follows: deploy the virtual machine on the physical machine, build a resource allocation optimization model, use the model to initialize the parameter values ​​of the physical machine and the resource values ​​configured for the deployed virtual machine, calculate the cost to be paid to the physical machine when obtaining resource allocation, and calculate the expected price of the physical machine. Within the specified resource allocation time period, the physical machine calculates the expected cost of providing unit resource configuration, and updates the resource configuration value based on the calculated payment cost and the expected cost, so as to maximize the server utility in the cloud environment within the limited physical machine resource constraints.

2. The cloud computing-based enterprise resource management platform according to claim 1, characterized in that: The step of optimizing resource allocation also includes: Step 41: Establish a load prediction model by processing and analyzing historical host load data of the cloud platform data center including host CPU, memory, host disk and network bandwidth; Step 42: According to the currently input host load data, a prediction algorithm is used to predict the host load value of the next stage, and the host load value is used to schedule the resources of the cloud data center; Step 43: Establish a resource scheduling model based on the load prediction result, the resource demand of the mobile terminal in the mobile cloud computing platform, and the resource ownership of the cloud server; Step 44: Based on the result of resource scheduling, the enterprise's resource management platform system optimizes the configuration of virtual machine resources according to the resource scheduling plan.

Images