A resource configuration method, device and equipment of a cloud platform and a storage medium

Abstract

The application discloses a resource configuration method and device of a cloud platform, equipment and a storage medium, relates to the technical field of cloud computing resource management, and comprises the following steps: performing data preprocessing on resource information of a node in a target cloud platform, and constructing a resource portrait based on the processed resource information; analyzing the required amount of computing resources of the node when running a storage service according to the resource portrait, and obtaining a resource demand result; evaluating the current resource pressure state of the node according to the resource portrait and the resource demand result, and obtaining a resource pressure result; generating target resource configuration parameters of the node according to the resource demand result and the resource pressure result and through a preset parameter mapping rule; performing preset constraint verification on the target resource configuration parameters, determining final resource configuration parameters based on the verification result and the target resource configuration parameters, and performing resource configuration on the node based on the final resource configuration parameters. The application can generate resource configuration parameters of a cloud platform node by comprehensively considering resource demand and resource pressure.

Description

Technical Field

[0001] This invention relates to the field of cloud computing resource management technology, and in particular to a resource configuration method, apparatus, device and storage medium for a cloud platform. Background Technology

[0002] With the development of cloud computing technology, cloud platforms have been widely used. Currently, cloud platform computing resource configuration mainly adopts manual experience or fixed configuration strategies, with some solutions dynamically adjusting configuration parameters based on the utilization of resources such as CPU and memory. However, most existing methods only optimize configuration based on the current resource operating status of nodes. When the node's storage scale, the number of OSDs (Object Storage Devices), or business load changes, the generated configuration parameters are difficult to accurately adapt to the actual resource needs of the nodes.

[0003] Furthermore, existing technologies typically handle resource demand analysis and resource pressure analysis separately, making it difficult for computing resource allocation to simultaneously consider node resource requirements and resource operating status, thus affecting resource utilization and system stability.

[0004] In summary, how to comprehensively consider resource demand and resource pressure to collaboratively generate resource configuration parameters for cloud platform nodes is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a resource allocation method, apparatus, device, and storage medium for a cloud platform, capable of comprehensively considering resource demand and resource pressure to collaboratively generate resource allocation parameters for cloud platform nodes. The specific solution is as follows: Firstly, this application provides a resource configuration method for a cloud platform, including: Resource information of nodes in the target cloud platform is collected, and the resource information is preprocessed. Based on the processed resource information, a resource profile corresponding to the node is constructed. The resource information includes computing resource information, storage resource information, and running status information. Based on the computing resource information and storage resource information in the resource profile, analyze the computing resource requirements of the node when running storage services, and obtain the corresponding resource requirement results. Based on the running status information and resource demand results in the resource profile, the current resource pressure status of the node is evaluated to obtain the corresponding resource pressure results; Based on the resource demand results and the resource pressure results, the target resource configuration parameters of the node are generated through preset parameter mapping rules; The target resource configuration parameters are subjected to preset constraint verification, and the final resource configuration parameters are determined based on the verification results and the target resource configuration parameters. The node is then configured with resources based on the final resource configuration parameters.

[0006] Optionally, the data preprocessing of the resource information includes: The format of the resource information is standardized, and the standardized resource information is then subjected to an integrity check to determine whether there is any missing information in the resource information. If the missing information exists, it is filled in, and the completed resource information is validated. The resource information that passes the validation is determined as the processed resource information.

[0007] Optionally, the step of analyzing the computing resource requirements of the node when running storage services based on the computing resource information and storage resource information in the resource profile includes: Based on the number of OSDs, storage capacity, and disk type in the storage resource information of the resource profile, as well as the computing resource information and the node role of the node, the computing resource requirements of the node when running the storage service are analyzed to obtain the resource requirement result. The number of OSDs and the storage capacity are both proportional to the computing resource requirements. The computing resource requirements are higher when the node role is a preset control node than when the node role is a preset ordinary computing node. The CPU resource requirements are higher when the disk type is a solid-state drive than when the disk type is a mechanical hard drive.

[0008] Optionally, the step of evaluating the current resource pressure status of the node based on the running status information and resource demand results in the resource profile to obtain the corresponding resource pressure results includes: Based on the CPU utilization, memory utilization, and storage utilization in the running status information of the resource profile, the current resource usage of the node is determined; The resource pressure level of the node is determined based on the resource demand results and the resource usage, and the resource pressure result is generated based on the resource pressure level.

[0009] Optionally, the step of generating the target resource configuration parameters for the node based on the resource demand results and the resource pressure results, and through a preset parameter mapping rule, includes: The resource demand level of the node is determined based on the resource demand results, and the resource pressure level of the node is determined based on the resource pressure results. Based on the resource demand level and the resource pressure level, the reserved resource amount and over-allocation ratio in the target resource configuration parameters are adjusted according to the preset parameter mapping rules.

[0010] Optionally, adjusting the reserved resource amount and over-allocation ratio in the target resource configuration parameters according to the resource demand level and the resource pressure level, and through the preset parameter mapping rules, includes: If the resource demand level is higher than the preset demand level and the resource pressure level is higher than the preset pressure level, then increase the amount of reserved resources and decrease the over-allocation ratio; If the resource demand level is higher than the preset demand level and the resource pressure level is not higher than the preset pressure level, then increase the amount of reserved resources and increase the over-allocation ratio; If the resource demand level is not higher than the preset demand level and the resource pressure level is higher than the preset pressure level, then increase the amount of reserved resources and maintain or reduce the over-allocation ratio; If the resource demand level is not higher than the preset demand level and the resource pressure level is not higher than the preset pressure level, then the amount of reserved resources is reduced and the over-allocation ratio is increased.

[0011] Optionally, the step of performing preset constraint verification on the target resource configuration parameters includes: A first verification result is obtained by verifying whether the reserved resource amount in the target resource configuration parameters exceeds the total resource capacity of the node, and a second verification result is obtained by verifying whether the reserved resource amount in the target resource configuration parameters meets the resource requirements. A third verification result is obtained by verifying whether the over-allocation ratio matches the resource pressure result, and a fourth verification result is obtained by verifying whether the correlation between the reserved resource amount and the over-allocation ratio meets the preset coordination conditions. Accordingly, determining the final resource configuration parameters based on the verification result and the target resource configuration parameters includes: If all the verification results indicate that they pass, then the target resource configuration parameters are directly determined as the final resource configuration parameters; the verification results include a first verification result, a second verification result, a third verification result, and a fourth verification result; If any of the verification results fails, the target resource configuration parameters are corrected, and the final resource configuration parameters are determined based on the corrected resource configuration parameters.

[0012] Secondly, this application provides a resource configuration device for a cloud platform, comprising: The resource information acquisition module is used to collect resource information of nodes in the target cloud platform, perform data preprocessing on the resource information, and construct a resource profile corresponding to the node based on the processed resource information; the resource information includes computing resource information, storage resource information, and operating status information; The resource requirement result determination module is used to analyze the amount of computing resources required by the node when running the storage service based on the computing resource information and the storage resource information in the resource profile, and to obtain the corresponding resource requirement result. The resource pressure result determination module is used to evaluate the current resource pressure status of the node based on the running status information and resource demand results in the resource profile, and obtain the corresponding resource pressure result. The resource configuration parameter generation module is used to generate the target resource configuration parameters of the node based on the resource demand results and the resource pressure results, and through a preset parameter mapping rule. The constraint verification module is used to perform preset constraint verification on the target resource configuration parameters, determine the final resource configuration parameters based on the verification results and the target resource configuration parameters, and configure the node based on the final resource configuration parameters.

[0013] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned resource configuration method for the cloud platform.

[0014] Fourthly, this application provides a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned cloud platform resource allocation method.

[0015] In this application, resource information of nodes in the target cloud platform is first collected, and the resource information is preprocessed. A resource profile corresponding to the node is then constructed based on the processed resource information. The resource information includes computing resource information, storage resource information, and operating status information. Next, based on the computing resource information and storage resource information in the resource profile, the computing resource requirements of the node when running storage services are analyzed to obtain corresponding resource requirement results. Then, based on the operating status information and resource requirement results in the resource profile, the current resource pressure status of the node is evaluated to obtain corresponding resource pressure results. Subsequently, based on the resource requirement results and resource pressure results, target resource configuration parameters for the node are generated through preset parameter mapping rules. Finally, preset constraint verification is performed on the target resource configuration parameters, and the final resource configuration parameters are determined based on the verification results and the target resource configuration parameters. Finally, resource configuration is performed on the node based on the final resource configuration parameters. As can be seen from the above, this application collects resource information of nodes in the target cloud platform and performs data preprocessing to construct resource profiles corresponding to the nodes. The resource profiles integrate computing resource information, storage resource information, and operational status information, providing a unified data foundation for subsequent analysis. Based on this, the computing resource requirements of the nodes when running storage services are analyzed according to the computing and storage resource information in the resource profiles, resulting in resource demand results and realizing quantitative analysis of computing resource requirements based on the scale of storage services. Next, the current resource pressure status of the nodes is evaluated based on the operational status information and resource demand results in the resource profiles, resulting in resource pressure results and realizing comprehensive pressure assessment that combines resource demand trends with the current operational status. Subsequently, target resource configuration parameters for the nodes are generated according to the resource demand results and resource pressure results and through preset parameter mapping rules, realizing the collaborative mapping of resource demand and resource pressure. Finally, the target resource configuration parameters are verified by preset constraints, and the final resource configuration parameters are determined based on the verification results before configuring the nodes. This ensures that the configuration parameters meet the node's resource capabilities and the consistency between various parameters, avoiding the problem of configuring parameters only based on resource utilization or fixed configuration strategies, and making the generated configuration parameters more consistent with the actual operation of the nodes. In this way, this application can comprehensively generate resource configuration parameters for cloud platform nodes by combining resource demand and resource pressure. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a resource configuration method for a cloud platform provided in this application; Figure 2 This application provides an architecture diagram of a specific resource allocation system; Figure 3 A specific cloud platform resource configuration flowchart provided for this application; Figure 4 A schematic diagram of a resource allocation device for a cloud platform provided in this application; Figure 5 This application provides a structural diagram of an electronic device. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] With the development of cloud computing technology, cloud platforms have been widely used. Currently, cloud platform computing resource configuration mainly adopts manual experience-based configuration or fixed configuration strategies, with some solutions dynamically adjusting configuration parameters based on the utilization of resources such as CPU and memory. However, most existing methods only optimize configuration based on the current resource operating status of nodes. When the node's storage scale, number of OSDs, or business load changes, the generated configuration parameters are difficult to accurately adapt to the actual resource needs of the nodes.

[0020] Furthermore, existing technologies typically handle resource demand analysis and resource pressure analysis separately, making it difficult to simultaneously consider node resource requirements and resource operating status in computing resource allocation, thus affecting resource utilization and system stability. To address this, this application provides a resource allocation scheme for a cloud platform that can collaboratively generate resource configuration parameters for cloud platform nodes by comprehensively considering resource demand and resource pressure.

[0021] See Figure 1 As shown in the figure, an embodiment of the present invention discloses a resource configuration method for a cloud platform, which may include: Step S11: Collect resource information of nodes in the target cloud platform, perform data preprocessing on the resource information, and construct a resource profile corresponding to the node based on the processed resource information; the resource information includes computing resource information, storage resource information, and running status information.

[0022] In this embodiment, the target cloud platform is a hyperconverged cloud platform built on OpenStack and Ceph.

[0023] It should be noted that, since the computing resource information, storage resource information, and node operating status information in the hyperconverged cloud platform originate from different components, the acquisition methods and data formats for these various resource information differ. If subsequent resource demand analysis and resource pressure analysis collect resource information separately, it will not only increase resource collection overhead but also easily lead to inconsistent analysis results due to different collection times. Therefore, in this embodiment, node resource information is first collected uniformly, and a resource profile is constructed to provide a consistent data foundation for subsequent resource demand analysis and resource pressure analysis.

[0024] In this embodiment, resource information in the target cloud platform is collected on a node-by-node basis. The computing resource information mainly comes from the OpenStack platform, including information such as the number of CPU cores, physical memory capacity, and virtualization resource configuration of the nodes. The storage resource information mainly comes from the Ceph cluster, including information such as the number of OSDs, storage capacity, disk type, and storage utilization. The running status information mainly comes from the Linux operating system, including running status data such as CPU utilization, memory utilization, disk utilization, and system load.

[0025] Next, the resource information is preprocessed. The specific process may include: first, standardizing the format of the resource information, and then performing a completeness check on the standardized resource information to determine whether there is any missing information; if so, the missing information is filled in, and the completed resource information is validated for validity, and the resource information that passes the validity check is determined as the processed resource information.

[0026] Specifically, after resource information collection is completed, all types of resource information are uniformly organized. Since data from different sources may differ in name, unit, and data format, the resource information is first standardized, converting all resource information into a unified data format. For example, CPU resources are uniformly converted to the number of cores, memory resources are uniformly converted to MB or GB as the unit of measurement, and the utilization rates of various resources are uniformly converted to percentages, enabling data from different sources to be analyzed in a unified manner. Next, to ensure the accuracy of resource information, the standardized resource information is also checked for completeness and validity during the data organization process. When a piece of resource information is found to be missing, the corresponding resource information is reacquired; when the collected data is found to be outside the normal range or contains outliers, it can be supplemented using the most recent valid collection result or preset default values ​​to ensure the completeness and reliability of the resource information.

[0027] After data preprocessing, resource profiles are created at the node level. Resource profiles record the current resource configuration and operating status of each node, including at least node identifier, number of CPU cores, physical memory capacity, storage capacity, number of OSDs, disk type, node role, CPU utilization, memory utilization, and storage utilization. Through resource profiles, resource data that was originally scattered across different components is organized into a unified data object, ensuring that each node corresponds to a complete resource description.

[0028] It should be noted that the final generated resource profile serves as the common input for subsequent resource demand analysis and resource pressure analysis. All subsequent steps are based on the resource profile for calculation, eliminating the need to repeatedly collect node resource information. This ensures that the data used in the resource analysis process remains consistent and provides a unified data foundation for the collaborative generation of resource configuration parameters.

[0029] Step S12: Based on the computing resource information and storage resource information in the resource profile, analyze the computing resource requirements of the node when running the storage service, and obtain the corresponding resource requirement results.

[0030] It should be noted that since different nodes in the target cloud platform undertake different storage tasks, their computing resource requirements also vary. If computing resources are configured solely based on the current CPU or memory utilization of a node, it cannot accurately reflect the actual computing resource needs of the storage service. Therefore, this embodiment analyzes the node resource requirements based on the constructed resource profile, providing a basis for generating subsequent computing resource configuration parameters.

[0031] In this embodiment, a resource profile (ResourceProfile) is read to obtain resource information such as the number of CPU cores, physical memory capacity, number of OSDs, storage capacity, disk type, and node role of the node. Using a resource demand model (ResourceDemandModel), the computational resource requirements of the node when running the storage service are analyzed based on the computational and storage resource information in the resource profile. Specifically, the process may include: analyzing the computational resource requirements of the node when running the storage service based on the number of OSDs, storage capacity, and disk type in the storage resource information of the resource profile, as well as the computational resource information and the node role, to obtain the resource demand result. The number of OSDs and the storage capacity are directly proportional to the computational resource requirements. The computational resource requirements corresponding to the node role being a preset control node are higher than those corresponding to the node role being a preset ordinary computing node. The CPU resource requirements corresponding to the disk type being a solid-state drive (SSD) are higher than those corresponding to the disk type being a hard disk drive (HDD).

[0032] Specifically, in the resource requirements analysis process, the computing resource requirements of the storage service are first analyzed based on the number of OSDs deployed on the nodes. Generally, the more OSDs there are, the more tasks the nodes undertake, such as data storage, data replication, data recovery, and data consistency maintenance, and the higher the demand for CPU and memory resources. Therefore, it is appropriate to increase the CPU and memory resource requirements of the nodes.

[0033] Secondly, analyze resource requirements based on the node's storage capacity. When a node manages a large amount of storage, it needs to handle more data read / write requests and metadata management tasks, thus increasing the demand for computing resources; when the storage capacity is small, the resource requirements should be appropriately reduced.

[0034] Furthermore, resource requirements are adjusted based on disk type. For example, nodes using SSDs (Solid State Drives) have higher data read and write performance, requiring relatively higher CPU resources to fully utilize their storage capabilities; nodes using HDDs (Hard Disk Drives) have relatively lower data processing capabilities, allowing for a more appropriate reduction in CPU resource requirements.

[0035] In addition, resource requirements are adjusted based on the node's role. For example, control nodes, in addition to providing storage services, also need to run platform management services, so their CPU and memory requirements are appropriately increased; while the resource requirements for ordinary computing nodes are mainly determined based on the scale of the storage service.

[0036] In one specific implementation, when a node has a large number of OSDs, a large storage capacity, and uses SSD disks, it is determined that the node has high resource requirements, and the CPU and memory resources that the host machine needs to reserve are increased accordingly.

[0037] In another specific implementation, when a node deploys only a small number of OSDs and has a small storage capacity, it is determined that the resource demand is low, and the requirement for reserved computing resources is reduced.

[0038] After completing the above analysis, a corresponding resource requirement result, StorageDemandSet, is generated. StorageDemandSet includes at least information such as node CPU resource requirements, memory resource requirements, and resource requirement levels, describing the computing resources required for a node to run storage services. Ultimately, StorageDemandSet serves as input data for subsequent resource stress assessment and computing resource configuration, providing a unified basis for resource requirements in subsequent processing.

[0039] Step S13: Based on the running status information and resource demand results in the resource profile, evaluate the current resource pressure status of the node to obtain the corresponding resource pressure results.

[0040] It should be noted that the current resource utilization rate of a node only reflects the current resource usage and cannot accurately reflect whether the node will have sufficient resources in the future. For example, a node may currently have low CPU utilization, but its resource demand may continue to increase due to the large amount of storage business it undertakes. In this case, judging the node's resource pressure solely based on CPU utilization may easily underestimate the actual resource pressure on the node. Therefore, in the resource pressure analysis process of this embodiment, not only is the current resource operating status of the node considered, but also the resource demand results are combined to conduct a comprehensive analysis of the node's resource pressure, providing a more accurate basis for the subsequent calculation of resource configuration parameters.

[0041] In this embodiment, the resource profile (ResourceProfile) and resource demand result (StorageDemandSet) are read to obtain resource operation status information such as node CPU utilization, memory utilization, and storage utilization, as well as resource demand information such as node CPU resource demand and memory resource demand. Using the resource pressure model (PressureModel), the current resource pressure status of the node is evaluated based on the operation status information and resource demand result in the resource profile, resulting in a corresponding resource pressure result. The specific process may include: first, determining the current resource usage of the node based on the CPU utilization, memory utilization, and storage utilization in the operation status information of the resource profile; then, determining the resource pressure level of the node based on the resource demand result and the resource usage, and generating the resource pressure result based on the resource pressure level.

[0042] Specifically, the node's resource usage is first analyzed based on its current CPU utilization, memory utilization, and storage utilization. When CPU utilization continues to rise, it indicates that the node's computing resources are gradually approaching saturation; when memory utilization continues to rise, it indicates that the memory resources available for business operations are decreasing; and when storage utilization is high, it indicates that the node's storage space is gradually becoming strained.

[0043] Subsequently, the impact of node resource demand on resource pressure was further analyzed based on the StorageDemandSet resource demand results. When node resource demand is high, even if the current resource utilization rate is still within the normal range, it is considered that the node's future resource consumption may continue to increase, so the resource pressure level should be appropriately increased; when node resource demand is low and the current resource utilization rate is low, the node resource pressure is considered low, and the resource pressure level can be appropriately reduced.

[0044] In one specific implementation, when a node's CPU utilization is 60%, although it has not yet reached the preset alarm threshold, but the node has deployed a large number of OSDs and the resource demand analysis results show that the CPU resource demand is continuously increasing, then it is generally judged that the node is under high resource pressure.

[0045] In another specific implementation, if a node's CPU utilization is 75%, but the node undertakes relatively few storage services and has low resource requirements, then the node's resource pressure is considered to be within the normal range, and there is no need to excessively adjust the resource configuration.

[0046] Furthermore, based on the comprehensive analysis results of resource operation status and resource demand, and according to preset rules, resource pressure levels are divided into three levels, such as low pressure, medium pressure and high pressure, to reflect the current resource tension of the nodes.

[0047] After completing the resource pressure analysis, a resource pressure result, PressureIndex, is generated. This result includes the node resource pressure value and its corresponding resource pressure level. The PressureIndex serves as input for generating subsequent resource configuration parameters. Together with the StorageDemandSet resource requirement result, it contributes to the generation of these parameters, ensuring that subsequent configuration parameters not only meet node resource requirements but also adapt to the node's current resource operating status, thus achieving collaborative analysis of resource requirements and resource pressure.

[0048] It is understood that in this embodiment, resource pressure is not determined solely based on the current resource utilization rate of a node, but rather by a combination of the node's resource utilization rate and resource demand analysis results. When both resource utilization and resource demand are high, the resource pressure level is increased; when resource utilization is low but resource demand is high, the resource pressure level is appropriately increased; when resource utilization is high but resource demand is low, the resource pressure level is determined based on resource operation trends; and when both resource utilization and resource demand are low, the resource pressure level is decreased.

[0049] Step S14: Based on the resource demand results and the resource pressure results, generate the target resource configuration parameters of the node through preset parameter mapping rules.

[0050] It should be noted that most existing technologies adjust computational resource configuration parameters based on fixed configuration templates or the current resource utilization rate of nodes, without comprehensively considering the correlation between node resource demand and resource pressure. Therefore, the generated configuration parameters are difficult to adapt to changes in node resource status. In contrast, this embodiment performs collaborative calculations on the results of resource demand analysis and resource pressure analysis, dynamically generating computational resource configuration parameters according to different resource states, making the computational resource configuration more consistent with the actual operating conditions of nodes.

[0051] In this embodiment, the resource demand result StorageDemandSet and the resource pressure result PressureIndex are read simultaneously. Based on pre-established parameter mapping rules, the number of reserved CPUs on the host machine, the reserved memory capacity on the host machine, the CPU over-allocation ratio, and the memory over-allocation ratio are calculated to generate the target resource configuration parameter TargetParameterSet. That is, the target resource configuration parameters for the node are generated based on the resource demand result and the resource pressure result, and through the preset parameter mapping rules. The specific process may include: first, determining the resource demand level of the node based on the resource demand result, and then determining the resource pressure level of the node based on the resource pressure result; then, adjusting the reserved resource amount and over-allocation ratio in the target resource configuration parameters based on the resource demand level and the resource pressure level, and through the preset parameter mapping rules.

[0052] Specifically, during the generation of the target resource configuration parameter TargetParameterSet, the current resource demand level of the node is first determined based on the resource demand result StorageDemandSet, and then the resource pressure level of the node is determined by combining the resource pressure result PressureIndex. Finally, the adjustment direction of each computing resource configuration parameter is determined based on the resource demand level and the resource pressure level.

[0053] In the first specific implementation, if the resource demand level and resource pressure level are both higher than the preset demand level and the preset pressure level, then the reserved resource amount is increased and the over-allocation ratio is reduced. Specifically, when the node's resource demand and resource pressure are both high, it indicates that the node not only requires more computing resources to support the operation of the storage service, but also that the current resources are already under high load. At this time, the number of reserved CPUs and the reserved memory capacity on the host machine are appropriately increased, and the CPU over-allocation ratio and memory over-allocation ratio are reduced, so that the host machine prioritizes ensuring the normal operation of the storage service and avoids node performance degradation due to excessive resource allocation.

[0054] In the second specific implementation, if the resource demand level is higher than the preset demand level but the resource pressure level is not higher than the preset pressure level, then the reserved resource amount is increased and the over-allocation ratio is increased. Specifically, when the node's resource demand is high but the resource pressure is low, it indicates that the node's current resources are relatively sufficient, but sufficient computing resources still need to be reserved for storage services in the future. At this time, a high host machine reserved CPU and host machine reserved memory configuration is maintained, while the CPU over-allocation ratio and memory over-allocation ratio are appropriately increased to improve the utilization rate of remaining computing resources while ensuring the resource demand of storage services.

[0055] In the third specific implementation, if the resource demand level is not higher than the preset demand level but the resource pressure level is higher than the preset pressure level, then the reserved resource amount is increased and the over-allocation ratio is maintained or reduced. Specifically, when the node resource demand is low but the resource pressure is high, it indicates that the current resource pressure mainly comes from business operations rather than storage service demands. In this case, the reserved resources on the host machine are appropriately increased, and the CPU over-allocation ratio and memory over-allocation ratio are controlled to avoid further increasing the node's computing load, allowing the node resources to gradually return to normal operating conditions.

[0056] In the fourth specific implementation, if the resource demand level is not higher than the preset demand level and the resource pressure level is not higher than the preset pressure level, then the amount of reserved resources is reduced and the over-allocation ratio is increased. Specifically, when the node's resource demand and resource pressure are low, it indicates that the node has a large resource margin. Accordingly, the number of reserved CPUs and the reserved memory capacity on the host machine are reduced, and the CPU over-allocation ratio and memory over-allocation ratio are appropriately increased, so that more computing resources can be allocated to virtual machines, improving the overall resource utilization of the node.

[0057] It should be noted that when resource demand or resource pressure changes, resource demand analysis and resource pressure analysis need to be re-executed, and the target resource configuration parameters need to be recalculated based on the latest analysis results. This allows the computing resource configuration to be dynamically adjusted according to changes in node resource status, rather than using fixed configuration parameters in the long term.

[0058] In one specific implementation, when a node adds multiple OSDs, the node's CPU and memory resource requirements increase significantly, and the node's resource pressure continues to rise. The system automatically increases the number of reserved CPUs and the reserved memory capacity on the host machine, and appropriately reduces the CPU over-allocation ratio and the memory over-allocation ratio to ensure that the new storage services have sufficient computing resources.

[0059] In another specific implementation, when the storage business of a node decreases and the resource demand analysis results show a decrease in resource demand, while the node's CPU utilization and memory utilization continue to decrease, the host machine's reserved resources are reduced, and the CPU over-allocation ratio and memory over-allocation ratio are appropriately increased so that the released computing resources can be provided to virtual machines, thereby improving the overall resource utilization.

[0060] In this embodiment, a target resource configuration parameter TargetParameterSet is generated through the above analysis. TargetParameterSet includes at least the following computational resource configuration parameters: the number of CPUs reserved on the host machine, the capacity of memory reserved on the host machine, the CPU over-allocation ratio, and the memory over-allocation ratio. Subsequently, the target resource configuration parameters will serve as input data for subsequent collaborative constraint verification, further validating and correcting the generated results to ensure that the final output resource configuration parameters meet the node's resource capabilities and operational requirements.

[0061] Step S15: Perform preset constraint verification on the target resource configuration parameters, determine the final resource configuration parameters based on the verification results and the target resource configuration parameters, and configure the node's resources based on the final resource configuration parameters.

[0062] It should be noted that while the target resource configuration parameters can reflect the current resource requirements and pressure of a node, in practical applications, it is also necessary to ensure that the generated configuration parameters meet the actual resource capabilities of the node and maintain consistency among them. Therefore, this embodiment performs collaborative constraint verification after the target resource configuration parameters are generated to ensure that the final generated configuration parameters can be directly applied to the cloud platform's computing resource configuration.

[0063] In this embodiment, the target resource configuration parameter TargetParameterSet is read, and the resource profile ResourceProfile, resource demand result StorageDemandSet, and resource pressure result PressureIndex are read simultaneously. Pre-defined constraint verification is performed on the target resource configuration parameter. The specific process may include: first, verifying whether the reserved resource amount in the target resource configuration parameter exceeds the total resource capacity of the node to obtain a first verification result; then, verifying whether the reserved resource amount in the target resource configuration parameter meets the resource demand result to obtain a second verification result; next, verifying whether the over-allocation ratio matches the resource pressure result to obtain a third verification result; and finally, verifying whether the correlation between the reserved resource amount and the over-allocation ratio meets preset coordination conditions to obtain a fourth verification result.

[0064] Specifically, during the pre-defined constraint verification process, the resource capacity is first verified. Based on resource information such as the number of CPU cores and physical memory capacity of the node, it is checked whether the number of CPUs and the amount of memory reserved on the host machine exceed the actual resource range that the node can provide. If the reserved resources exceed the actual resource capacity of the node, the target resource configuration parameters are considered not to meet the resource capacity constraints, and the corresponding parameters need to be corrected.

[0065] Secondly, the resource requirements are verified. Based on the CPU and memory resource requirements calculated from the StorageDemandSet, it is determined whether the target resource configuration parameters can meet the resource requirements for the node to run the storage service. If the host machine's reserved resources are lower than the resource requirements, the corresponding reserved resources are appropriately increased to ensure the stable operation of the storage service.

[0066] Next, resource pressure is verified. The PressureIndex result is used to determine whether the CPU over-allocation ratio and memory over-allocation ratio match the current node's resource pressure. When node resource pressure is high, the CPU and memory over-allocation ratios are checked to ensure they are within preset ranges to prevent further strain on node resources due to excessive over-allocation. When node resource pressure is low, the over-allocation ratio is allowed to be appropriately increased to improve node computing resource utilization.

[0067] Furthermore, the relationships between the various configuration parameters are verified. Since there are interrelationships between the host machine's reserved CPU, reserved host machine memory, CPU over-allocation ratio, and memory over-allocation ratio, it is necessary to ensure that the adjustment direction of each parameter remains consistent. For example, when the host machine's reserved CPU increases, the CPU over-allocation ratio should not continue to increase; when the host machine's reserved memory decreases, it should still ensure that the resource demand analysis results are met, avoiding node resource configuration imbalances caused by inconsistencies in parameter configuration.

[0068] In one specific implementation, when it is detected that the number of reserved CPUs on the host machine has exceeded the number of CPU cores on the node, the number of reserved CPUs on the host machine is automatically adjusted to within the allowable range.

[0069] In another specific implementation, when the CPU over-allocation ratio exceeds the preset upper limit and the node resource pressure remains high, the CPU over-allocation ratio is appropriately reduced so that the configuration result matches the current resource status of the node.

[0070] Accordingly, the final resource configuration parameters are determined based on the verification results and the target resource configuration parameters. The specific process may include: if all the verification results indicate that they pass, then the target resource configuration parameters are directly determined as the final resource configuration parameters; the verification results include a first verification result, a second verification result, a third verification result, and a fourth verification result; if any of the verification results indicate that they fail, then the target resource configuration parameters are corrected, and the final resource configuration parameters are determined based on the corrected resource configuration parameters.

[0071] Specifically, once all the above checks pass, the corresponding final resource configuration parameter ValidatedTargetParameterSet is generated and used as the final output result of the cloud platform computing resource configuration. If there are parameters that do not meet the constraints, the target configuration parameters are corrected according to the corresponding check results, and the check is re-executed until all configuration parameters meet the resource capacity, resource requirements, resource pressure, and parameter association constraints.

[0072] Understandably, the final resource configuration parameter ValidatedTargetParameterSet can simultaneously meet the node resource capabilities, storage service resource requirements, and the current resource operating status of the node, while maintaining the coordination and consistency between various computing resource configuration parameters. This allows the generated computing resource configuration parameters to be directly applied to the hyperconverged cloud platform, achieving collaborative optimization configuration between computing and storage resources.

[0073] As can be seen from the above, this embodiment first collects resource information of nodes in the target cloud platform, performs data preprocessing on the resource information, and constructs a resource profile corresponding to the node based on the processed resource information. The resource information includes computing resource information, storage resource information, and running status information. Then, based on the computing resource information and storage resource information in the resource profile, the computing resource requirements of the node when running storage services are analyzed to obtain the corresponding resource requirement results. Next, based on the running status information and resource requirement results in the resource profile, the current resource pressure status of the node is evaluated to obtain the corresponding resource pressure results. Subsequently, based on the resource requirement results and the resource pressure results, target resource configuration parameters of the node are generated through preset parameter mapping rules. Finally, preset constraint verification is performed on the target resource configuration parameters, and the final resource configuration parameters are determined based on the verification results and the target resource configuration parameters. The node is then configured with resources based on the final resource configuration parameters. As can be seen from the above, this embodiment collects resource information of nodes in the target cloud platform and performs data preprocessing to construct resource profiles corresponding to the nodes. The resource profiles integrate computing resource information, storage resource information, and operating status information, providing a unified data foundation for subsequent analysis. Based on this, the computing resource requirements of the nodes when running storage services are analyzed according to the computing resource information and storage resource information in the resource profiles, and the resource requirement results are obtained, realizing quantitative analysis of computing resource requirements based on the scale of storage services. Then, the current resource pressure status of the nodes is evaluated according to the operating status information and resource requirement results in the resource profiles, and the resource pressure results are obtained, realizing a comprehensive pressure assessment that combines resource demand trends with the current operating status. Subsequently, the target resource configuration parameters of the nodes are generated according to the resource requirement results and resource pressure results and through preset parameter mapping rules, realizing the collaborative mapping of resource requirements and resource pressure. Finally, the target resource configuration parameters are verified by preset constraints, and the final resource configuration parameters are determined based on the verification results before the nodes are configured. This ensures that the configuration parameters meet the node's resource capabilities and the coordination consistency between the parameters, avoiding the problem of configuring parameters only based on resource utilization or fixed configuration strategies, and making the generated configuration parameters more consistent with the actual operating conditions of the nodes. In this way, this embodiment can comprehensively consider resource demand and resource pressure to collaboratively generate resource configuration parameters for cloud platform nodes.

[0074] In this embodiment, the architecture diagram of the resource allocation system is shown below. Figure 2 As shown; among them, the resource profile serves as the foundation for shared data, while the resource demand model and resource pressure model jointly participate in the generation of resource configuration parameters. For the specific process of resource configuration on the cloud platform, please refer to... Figure 3 As shown.

[0075] It should be noted that the resource configuration system is applied to a hyperconverged cloud platform composed of OpenStack and Ceph; OpenStack is responsible for compute resource management, and Ceph is responsible for storage resource management. The system first collects resource information from each node of the cloud platform and constructs a unified resource profile. The resource profile serves as the data foundation for the entire collaborative optimization process, providing a unified data source for subsequent modules.

[0076] Based on this, the system establishes a resource demand model (ResourceDemandModel) and a resource pressure model (PressureModel). The resource demand model is used to analyze the computing resources required for nodes to run storage services and generate a resource demand result (StorageDemandSet); the resource pressure model is used to assess resource pressure in conjunction with the node's running status and generate a resource pressure result (PressureIndex).

[0077] Subsequently, the TargetParameterSet Generation module integrates the resource demand results and resource pressure results, and generates the TargetParameterSet based on the preset parameter mapping rules.

[0078] Subsequently, the ConstraintCheckModel module further verifies the target resource configuration parameters by performing resource capacity constraints, resource demand constraints, resource pressure constraints, and parameter association constraints, generating the final configuration parameter set (ValidatedTargetParameterSet), which is then applied to the cloud platform computing resource configuration to achieve collaborative optimization between computing resources and storage resources.

[0079] As can be seen from the above, on the one hand, this embodiment constructs a resource profile and establishes a resource demand model and a resource pressure model to comprehensively analyze the node's resource demand and resource operation status. When generating computing resource configuration parameters, it considers both resource demand and resource pressure information simultaneously, avoiding the problem of existing technologies that only rely on resource utilization or fixed configuration strategies for parameter configuration. This makes the generated configuration parameters more consistent with the actual operation of the nodes. On the other hand, this embodiment dynamically generates computing resource configuration parameters such as host reserved CPU, host reserved memory, CPU over-allocation ratio, and memory over-allocation ratio based on changes in node resource demand and resource pressure. This allows the configuration parameters to automatically adjust with changes in node resource status, ensuring storage service resource requirements while improving computing resource utilization and reducing resource waste caused by excessive or insufficient resource configuration. Furthermore, after the target configuration parameters are generated, this embodiment performs unified verification of resource capacity, resource demand, resource pressure, and parameter relationships through collaborative constraint verification. This ensures that the final generated configuration parameters meet the node's resource capabilities and the collaborative relationships between parameters, improving the reliability of the configuration results and reducing resource conflicts and system operation risks caused by unreasonable configuration.

[0080] Accordingly, see Figure 4 As shown in the illustration, this application also provides a resource configuration device for a cloud platform, which may include: The resource information acquisition module 11 is used to acquire resource information of nodes in the target cloud platform, perform data preprocessing on the resource information, and construct a resource profile corresponding to the node based on the processed resource information; the resource information includes computing resource information, storage resource information, and running status information; The resource requirement result determination module 12 is used to analyze the amount of computing resources required by the node when running the storage service based on the computing resource information and the storage resource information in the resource profile, and to obtain the corresponding resource requirement result. The resource pressure result determination module 13 is used to evaluate the current resource pressure status of the node based on the running status information and the resource demand result in the resource profile, and obtain the corresponding resource pressure result; The resource configuration parameter generation module 14 is used to generate the target resource configuration parameters of the node based on the resource demand results and the resource pressure results, and through a preset parameter mapping rule. The constraint verification module 15 is used to perform preset constraint verification on the target resource configuration parameters, determine the final resource configuration parameters based on the verification results and the target resource configuration parameters, and configure the node based on the final resource configuration parameters.

[0081] In some specific embodiments, the resource information acquisition module 11 may include: The format standardization unit is used to standardize the format of the resource information and perform an integrity check on the standardized resource information to determine whether there is any missing information in the resource information. An information verification unit is used to complete the missing information if it exists, and to verify the validity of the completed resource information, and to determine the resource information that passes the validity verification as the processed resource information.

[0082] In some specific embodiments, the resource requirement result determination module 12 may include: The resource requirement analysis unit is used to analyze the computing resource requirements of the node when running the storage service based on the number of OSDs, storage capacity and disk type in the storage resource information in the resource profile, as well as the computing resource information and the node role of the node, and to obtain the resource requirement result. The number of OSDs and the storage capacity are both proportional to the computing resource requirements. The computing resource requirements are higher when the node role is a preset control node than when the node role is a preset ordinary computing node. The CPU resource requirements are higher when the disk type is a solid-state drive than when the disk type is a mechanical hard drive.

[0083] In some specific embodiments, the resource pressure result determination module 13 may include: The resource usage determination unit is used to determine the current resource usage of the node based on the CPU utilization, memory utilization, and storage utilization in the running status information of the resource profile. The resource pressure level determination unit is used to determine the resource pressure level of the node based on the resource demand result and the resource usage, and generate the resource pressure result based on the resource pressure level.

[0084] In some specific embodiments, the resource configuration parameter generation module 14 may include: The demand pressure level determination submodule is used to determine the resource demand level of the node based on the resource demand results, and to determine the resource pressure level of the node based on the resource pressure results. The resource configuration parameter adjustment submodule is used to adjust the reserved resource amount and over-allocation ratio in the target resource configuration parameters according to the resource demand level and the resource pressure level, and through the preset parameter mapping rules.

[0085] In some specific implementations, the resource configuration parameter adjustment submodule may include: The first adjustment unit is used to increase the amount of reserved resources and reduce the over-allocation ratio if the resource demand level is higher than the preset demand level and the resource pressure level is higher than the preset pressure level. The second adjustment unit is used to increase the amount of reserved resources and increase the over-allocation ratio if the resource demand level is higher than the preset demand level and the resource pressure level is not higher than the preset pressure level. The third adjustment unit is used to increase the amount of reserved resources and maintain or reduce the over-allocation ratio if the resource demand level is not higher than the preset demand level and the resource pressure level is higher than the preset pressure level. The fourth adjustment unit is used to reduce the amount of reserved resources and increase the over-allocation ratio if the resource demand level is not higher than the preset demand level and the resource pressure level is not higher than the preset pressure level.

[0086] In some specific embodiments, the constraint verification module 15 may include: The first verification result determination unit is used to verify whether the amount of reserved resources in the target resource configuration parameters exceeds the total resource capacity of the node to obtain a first verification result, and to verify whether the amount of reserved resources in the target resource configuration parameters meets the resource requirements to obtain a second verification result. The third verification result determination unit is used to verify whether the over-allocation ratio matches the resource pressure result to obtain the third verification result, and to verify whether the correlation between the reserved resource amount and the over-allocation ratio meets the preset coordination conditions to obtain the fourth verification result. The final resource configuration parameter determination unit is used to directly determine the target resource configuration parameter as the final resource configuration parameter if all the verification results indicate that they pass; the verification results include a first verification result, a second verification result, a third verification result, and a fourth verification result; The target resource configuration parameter correction unit is used to correct the target resource configuration parameters if any of the verification results fail, and to determine the final resource configuration parameters based on the corrected resource configuration parameters.

[0087] Furthermore, embodiments of this application also disclose an electronic device, Figure 5This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the cloud platform resource allocation method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0088] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0089] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0090] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the resource configuration method of the cloud platform executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program capable of performing other specific tasks.

[0091] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned resource allocation method for a cloud platform. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0092] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0093] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0095] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0096] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A resource allocation method for a cloud platform, characterized in that, include: Collect resource information of nodes in the target cloud platform, perform data preprocessing on the resource information, and construct a resource profile corresponding to the node based on the processed resource information; Resource information includes computing resource information, storage resource information, and operational status information; Based on the computing resource information and storage resource information in the resource profile, analyze the computing resource requirements of the node when running storage services, and obtain the corresponding resource requirement results. Based on the running status information and resource demand results in the resource profile, the current resource pressure status of the node is evaluated to obtain the corresponding resource pressure results; Based on the resource demand results and the resource pressure results, the target resource configuration parameters of the node are generated through preset parameter mapping rules; The target resource configuration parameters are subjected to preset constraint verification, and the final resource configuration parameters are determined based on the verification results and the target resource configuration parameters. The node is then configured with resources based on the final resource configuration parameters.

2. The resource allocation method for a cloud platform according to claim 1, characterized in that, The data preprocessing of the resource information includes: The format of the resource information is standardized, and the standardized resource information is then subjected to an integrity check to determine whether there is any missing information in the resource information. If the missing information exists, it is filled in, and the completed resource information is validated. The resource information that passes the validation is determined as the processed resource information.

3. The resource allocation method for a cloud platform according to claim 1, characterized in that, The step of analyzing the computing resource requirements of the node when running storage services based on the computing resource information and storage resource information in the resource profile includes: Based on the number of OSDs, storage capacity, and disk type in the storage resource information of the resource profile, as well as the computing resource information and the node role of the node, the computing resource requirements of the node when running the storage service are analyzed to obtain the resource requirement result. The number of OSDs and the storage capacity are both proportional to the computing resource requirements. The computing resource requirements are higher when the node role is a preset control node than when the node role is a preset ordinary computing node. The CPU resource requirements are higher when the disk type is a solid-state drive than when the disk type is a mechanical hard drive.

4. The resource allocation method for a cloud platform according to claim 1, characterized in that, The step of assessing the current resource pressure status of the node based on the operational status information and resource demand results in the resource profile, and obtaining the corresponding resource pressure results, includes: Based on the CPU utilization, memory utilization, and storage utilization in the running status information of the resource profile, the current resource usage of the node is determined; The resource pressure level of the node is determined based on the resource demand results and the resource usage, and the resource pressure result is generated based on the resource pressure level.

5. The resource allocation method for a cloud platform according to any one of claims 1 to 4, characterized in that, The step of generating the target resource configuration parameters for the node based on the resource demand results and the resource pressure results, and through a preset parameter mapping rule, includes: The resource demand level of the node is determined based on the resource demand results, and the resource pressure level of the node is determined based on the resource pressure results. Based on the resource demand level and the resource pressure level, the reserved resource amount and over-allocation ratio in the target resource configuration parameters are adjusted according to the preset parameter mapping rules.

6. The resource allocation method for a cloud platform according to claim 5, characterized in that, The step of adjusting the reserved resource amount and over-allocation ratio in the target resource configuration parameters according to the resource demand level and the resource pressure level, and through the preset parameter mapping rules, includes: If the resource demand level is higher than the preset demand level and the resource pressure level is higher than the preset pressure level, then increase the amount of reserved resources and decrease the over-allocation ratio; If the resource demand level is higher than the preset demand level and the resource pressure level is not higher than the preset pressure level, then increase the amount of reserved resources and increase the over-allocation ratio; If the resource demand level is not higher than the preset demand level and the resource pressure level is higher than the preset pressure level, then increase the amount of reserved resources and maintain or reduce the over-allocation ratio; If the resource demand level is not higher than the preset demand level and the resource pressure level is not higher than the preset pressure level, then the amount of reserved resources is reduced and the over-allocation ratio is increased.

7. The resource allocation method for a cloud platform according to claim 6, characterized in that, The step of performing preset constraint verification on the target resource configuration parameters includes: A first verification result is obtained by verifying whether the reserved resource amount in the target resource configuration parameters exceeds the total resource capacity of the node, and a second verification result is obtained by verifying whether the reserved resource amount in the target resource configuration parameters meets the resource requirements. A third verification result is obtained by verifying whether the over-allocation ratio matches the resource pressure result, and a fourth verification result is obtained by verifying whether the correlation between the reserved resource amount and the over-allocation ratio meets the preset coordination conditions. Accordingly, determining the final resource configuration parameters based on the verification result and the target resource configuration parameters includes: If all the verification results indicate that they pass, then the target resource configuration parameters are directly determined as the final resource configuration parameters; the verification results include a first verification result, a second verification result, a third verification result, and a fourth verification result; If any of the verification results fails, the target resource configuration parameters are corrected, and the final resource configuration parameters are determined based on the corrected resource configuration parameters.

8. A resource allocation device for a cloud platform, characterized in that, include: The resource information acquisition module is used to collect resource information of nodes in the target cloud platform, perform data preprocessing on the resource information, and construct a resource profile corresponding to the node based on the processed resource information; the resource information includes computing resource information, storage resource information, and operating status information; The resource requirement result determination module is used to analyze the amount of computing resources required by the node when running the storage service based on the computing resource information and the storage resource information in the resource profile, and to obtain the corresponding resource requirement result. The resource pressure result determination module is used to evaluate the current resource pressure status of the node based on the running status information and resource demand results in the resource profile, and obtain the corresponding resource pressure result. The resource configuration parameter generation module is used to generate the target resource configuration parameters of the node based on the resource demand results and the resource pressure results, and through a preset parameter mapping rule. The constraint verification module is used to perform preset constraint verification on the target resource configuration parameters, determine the final resource configuration parameters based on the verification results and the target resource configuration parameters, and configure the node based on the final resource configuration parameters.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory; wherein the memory is used to store a computer program, which is loaded and executed by the processor to implement the resource configuration method of the cloud platform as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer programs, which, when executed by a processor, implement the resource configuration method of the cloud platform as described in any one of claims 1 to 7.

Images