Cloud resource scheduling method, device, equipment, storage medium and program product

Abstract

The application relates to the technical field of cloud computing, and provides a cloud resource scheduling method and device, equipment, a storage medium and a program product. The method comprises the following steps: performing network testing on each edge cloud node based on the IP address of a target user to obtain network quality parameters of each edge cloud node; calculating resource recommendation weights of each edge cloud node according to the network quality parameters, and selecting a target node with the best network quality based on the resource recommendation weights to schedule a resource pool of the target node to open resources for the user. The network quality parameters of the edge cloud node are obtained through network testing and are used for calculating the resource recommendation weights, so that the edge cloud node with the best network quality can be selected according to the resource recommendation weights of each edge node to open resources for the user, the service quality of the user is guaranteed, and the best scheduling of the cloud resources is realized.

Description

Technical Field

[0001] This application relates to the field of cloud computing technology, specifically to a cloud resource scheduling method, apparatus, device, storage medium, and program product. Background Technology

[0002] Currently, under the existing cloud computing service delivery model, different users subscribe to business resources and obtain cloud services through a unified portal across the entire network. The resource subscription process mainly involves the user selecting the required resource type, specifying a resource pool, or randomly assigning a resource pool to the user, and then activating and distributing resources based on the user-specified or randomly assigned pool. The user's specification of the resource pool is based on their subjective judgment; for the user, the activated resource pool may not necessarily meet their service quality requirements. For the central cloud, the resource pool activated for users is either user-specified or randomly assigned, making optimal scheduling of cloud resources impossible. Summary of the Invention

[0003] This application provides a cloud resource scheduling method, apparatus, device, storage medium, and program product to solve the technical problem that existing resource scheduling methods, which involve user-specified resource pools or random allocation of resource pools to users, cannot guarantee the quality of business services for users and cannot achieve optimal scheduling of cloud resources.

[0004] In a first aspect, embodiments of this application provide a cloud resource scheduling method, including:

[0005] Network tests are performed on each edge cloud node based on the IP address of the target user to obtain the network quality parameters of each edge cloud node.

[0006] Calculate the resource recommendation weight for each edge cloud node based on the network quality parameters;

[0007] Based on the resource recommendation weight, the target node with the best network quality is selected from each of the edge cloud nodes, and the resource pool of the target node is scheduled to activate resources for the target user.

[0008] In one embodiment, calculating the resource recommendation weight for each edge cloud node based on the network quality parameters includes:

[0009] Calculate the weighted average of the target parameters; the target parameters are any one of the network quality parameters.

[0010] Obtain the weight set corresponding to the network quality parameters; the weight set includes the weight value of each target parameter; based on the weight set, perform a weighted summation of the weighted average of each target parameter to obtain the resource recommendation weight of the edge cloud node.

[0011] In one embodiment, calculating the weighted average of the target parameters includes:

[0012] Obtain the parameter value range of the target parameter; the parameter value range is determined by the maximum and minimum parameter values ​​of the target parameter;

[0013] The parameter value range is divided into a preset number of sub-ranges, and the number of targets whose parameter values ​​fall within each of the sub-ranges is counted.

[0014] The target weight of the sub-interval is obtained based on the ratio of the target quantity to the total number of target parameter values; the total number of target parameter values ​​is the same as the number of network tests performed on each edge cloud node based on the IP address; the target parameter values ​​are normalized based on the median of the sub-interval and the target weight to obtain the weighted average of the target parameters.

[0015] In one embodiment, if the target nodes include multiple nodes, after selecting the target node with the best network quality from each of the edge cloud nodes based on the resource recommendation weight, the method further includes:

[0016] Calculate the standard deviation of the network quality parameters of any target node based on the weighted average of the network quality parameters of each target node;

[0017] The standard deviations of the network quality parameters of the target node are weighted and summed based on the weight set to obtain the verification value of the target node.

[0018] The target nodes are filtered based on the verification value.

[0019] In one embodiment, the step of performing network testing on each edge cloud node based on the target user's IP address to obtain the network quality parameters of each edge cloud node includes:

[0020] Based on a preset number of tests, test commands are generated based on the target user's IP address during each test.

[0021] The test command is sent to each edge cloud node to perform network testing, and the network quality parameters of each edge cloud node are obtained.

[0022] In one embodiment, before performing network testing on each edge cloud node based on the target user's IP address to obtain the network quality parameters of each edge cloud node, the method further includes:

[0023] Obtain information about the target user's global virtual private cloud and the specified IP address range, as well as the target user's configured business and resource requirements;

[0024] Based on the global virtual private cloud information, the IP address range, the resource requirements, and the business requirements, a resource ordering instruction for the target user is generated.

[0025] Secondly, embodiments of this application provide a cloud resource scheduling device, comprising:

[0026] The network testing module is used to perform network tests on each edge cloud node based on the IP address of the target user, and to obtain the network quality parameters of each edge cloud node.

[0027] The weight calculation module is used to calculate the resource recommendation weight of each edge cloud node based on the network quality parameters.

[0028] The resource scheduling module is used to select the target node with the best network quality from each of the edge cloud nodes based on the resource recommendation weight, and schedule the resource pool of the target node to enable resources for the target user.

[0029] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the steps of the cloud resource scheduling method described in the first aspect.

[0030] Fourthly, embodiments of this application provide a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the cloud resource scheduling method described in the first aspect.

[0031] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the cloud resource scheduling method described in the first aspect.

[0032] The cloud resource scheduling method, apparatus, device, storage medium, and program products provided in this application embodiment perform network tests on each edge cloud node using the user's IP address to obtain the network quality parameters of each edge cloud node. Based on the network quality parameters of each edge cloud node, a resource recommendation weight is calculated for each edge cloud node. Then, the edge cloud node with the best network quality is selected based on the resource recommendation weight, and the resources of the edge cloud node are scheduled to be activated. The activated resources are the edge cloud node that best meets the user's business needs, which can ensure the quality of user's business services and achieve optimal scheduling of cloud resources. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart illustrating the cloud resource scheduling method provided in an embodiment of this application;

[0035] Figure 2 This is a schematic diagram of the cloud resource scheduling process provided in the embodiments of this application;

[0036] Figure 3 This is a schematic diagram of the cloud resource scheduling device provided in the embodiments of this application;

[0037] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] It should be noted that in the description of this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a 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. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly, for example, as a fixed connection, a detachable connection, or an integral connection; a mechanical connection or an electrical connection; a direct connection or an indirect connection through an intermediate medium; or a connection within two elements. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0040] The terms "first," "second," etc., used in this invention are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0041] Understandably, cloud computing provides and manages scalable, elastic, and shared pools of physical and virtual resources in an on-demand, self-service manner, and offers network access. In the existing centralized cloud model, different users subscribe to services and obtain cloud services through a unified portal. With the emergence of new technologies such as 5G and the Internet of Things (IoT), new business applications with high requirements for latency, bandwidth, and data security have arisen, such as cloud VR, smart transportation, smart factories, smart healthcare, online education, and cloud gaming. However, centralized clouds, based on centralized resource management, are generally deployed in specific regions. Users accessing the centralized cloud via the internet need to traverse multiple regional layers to reach the cloud resource pool, involving multiple hops and transmission distances of up to hundreds of kilometers, making it difficult to guarantee the quality of service for latency-sensitive businesses.

[0042] For services like cloud gaming, autonomous driving, and smart healthcare, which require low latency, network stability, and zero or low packet loss in terms of network performance, and for services like smart factories, which require keeping customer data within the network, a distributed resource pool with unified resource management is needed to meet these business requirements. After constructing the distributed resource pool, ensuring the optimal service quality for users when resources are activated is crucial for guaranteeing service quality.

[0043] Based on this, this application provides a cloud resource scheduling method that can obtain the network quality parameters of each edge cloud node through network testing based on the user's IP address, determine the network quality of each edge cloud node, and then select the edge cloud node with the best network quality to activate resources for the user.

[0044] Specifically, Figure 1 This is a flowchart illustrating the cloud resource scheduling method provided in an embodiment of this application. (Refer to...) Figure 1 The cloud resource scheduling method provided in this application embodiment may include:

[0045] Step 100: Perform network tests on each edge cloud node based on the IP address of the target user to obtain the network quality parameters of each edge cloud node; the network quality parameters include path, latency, jitter and packet loss rate.

[0046] Step 200: Calculate the resource recommendation weight of each edge cloud node based on the network quality parameters;

[0047] Step 300: Select the target node with the best network quality from each of the edge cloud nodes based on the resource recommendation weight, and schedule the resource pool of the target node to enable resources for the target user.

[0048] Specifically, firstly, network tests are performed on each edge cloud node based on the target user's IP address to obtain the network quality parameters of each edge cloud node. The target user can be any cloud resource user, and their IP address can be included in their resource ordering instruction. Obtaining the target user's resource ordering instruction reveals the IP address of the target user included in the instruction. Optionally, the target user's resource ordering instruction is generated based on their resource ordering requirements, which must include the type and quantity of resources the target user wishes to order.

[0049] Network quality parameters for edge cloud nodes include path, latency, jitter, and packet loss rate. Latency, specifically, is the time required for a message or packet to travel from one end of the network to the other. Latency = Transmission Latency + Propagation Latency + Processing Latency + Queuing Latency. Latency is related to device performance and is positively correlated with the number of hops the data traverses, the data propagation distance, and the traffic volume. With the same traffic volume, selecting higher-performance devices; reducing the number of hops the data traverses and simplifying business processing; and reducing data transmission distance by deploying resources closer to users at edge nodes can effectively reduce latency.

[0050] Network jitter refers to the variation in latency that occurs during data packet transmission in network communication. It reflects network stability; the lower the jitter value, the more stable the network. Network jitter has a significant impact on the performance of network applications. For example, network jitter in online games can cause lag and movement errors, while network jitter in voice calls can cause audio noise and voice interruptions.

[0051] Packet loss rate refers to the probability of data packets being lost during network transmission, and is used to assess the stability and reliability of network transmission. Packet loss can occur due to network congestion, interference, signal attenuation, and other conditions. A higher packet loss rate indicates poorer network transmission stability and lower data transmission quality. Generally, the packet loss rate is 0% in a congestion-free environment, 1%–4% in mild congestion, and greater than 5% in severe congestion. In applications with high real-time requirements, such as voice calls and video conferencing, a low packet loss rate is required to ensure real-time and smooth transmission.

[0052] Furthermore, network tests are performed on each edge cloud node based on the target user's IP address to obtain network quality parameters for each edge cloud node. These parameters include path, latency, jitter, and packet loss rate. Optionally, for different business requirements, the network quality parameters obtained through network testing can be partially the same or different.

[0053] Furthermore, based on the obtained network quality parameters, the resource recommendation weight of each edge cloud node is calculated, and based on the resource recommendation weight, the target node with the best network quality is selected from each edge cloud node. In response to the resource ordering instruction of the target user, the resource pool of the target node is scheduled to open resources for the target user, and the opened resources are distributed to provide services to the target user.

[0054] Optionally, when calculating resource recommendation weights based on network quality parameters, edge cloud nodes with better network quality have higher resource recommendation weights, and edge cloud nodes with worse network quality have lower resource recommendation weights. When selecting the target node with the best network quality, this target node is the edge cloud node with the highest resource recommendation weight. Alternatively, when calculating resource recommendation weights based on network quality parameters, edge cloud nodes with better network quality have lower resource recommendation weights, and edge cloud nodes with worse network quality have higher resource recommendation weights. When selecting the target node with the best network quality, this target node is the edge cloud node with the lowest resource recommendation weight.

[0055] In this embodiment, network tests are performed on each edge cloud node using the user's IP address to obtain the network quality parameters of each edge cloud node. Based on the network quality parameters of each edge cloud node, a resource recommendation weight is calculated for each edge cloud node. Then, the edge cloud node with the best network quality is selected based on the resource recommendation weight, and the resources of the edge cloud node are scheduled to be activated. The activated resources are the edge cloud node that best meets the user's business needs, which can ensure the quality of user business services and achieve optimal scheduling of cloud resources.

[0056] In one embodiment, the target user's resource ordering instruction further includes the template user's business requirements and resource requirements, wherein the business requirements include the business type, and the resource requirements include the resource quantity and resource requirements. Based on this, before step 100, it is also necessary to obtain the target user's resource ordering instruction, including:

[0057] Step 101: Obtain the information of the full-domain virtual private cloud created by the target user and the specified IP address range, as well as the business requirements and resource requirements configured by the target user;

[0058] Step 102: Generate the resource ordering instruction for the target user based on the global virtual private cloud information, the IP address range, the resource requirements, and the business requirements.

[0059] When obtaining a target user's resource ordering instruction, the system first obtains the target user's global private virtual cloud information, the IP address range specified by the target user, and the target user's configured business and resource requirements. Based on the global private cloud information, the target user's created global private cloud information, and the target user's configured business and resource requirements, the system generates the target user's resource ordering instruction.

[0060] Optionally, in step 100, network tests are performed on each edge cloud node based on the target user's IP address to obtain the network quality parameters of each edge cloud node, specifically including:

[0061] Step 110: Based on a preset number of tests, generate a test command based on the target user's IP address during each test; Step 120: Send the test command to each edge cloud node to perform network testing and obtain the network quality parameters of each edge cloud node.

[0062] When conducting network testing on each edge cloud node based on the target user's IP address, each edge cloud node is tested multiple times. The number of tests for different edge cloud nodes can be the same or different, and there is no specific limitation here. Optionally, the number of tests can be preset, and all edge cloud nodes can be tested simultaneously in each test.

[0063] Specifically, based on a pre-defined test, during each test, a test command is generated based on the target user's IP address. This test command is then sent to each edge cloud node to perform network testing, thereby obtaining the network quality parameters of each edge cloud node. The test command is generated based on the user's IP address, combined with the target user's configured resource and service requirements. Different resource and / or different service requirements generate different test commands. Using this test command to perform network testing on each edge cloud node can simulate the user's service traffic under the required resource conditions, thereby determining the network quality of each edge cloud node for the target user's service.

[0064] In one embodiment, the network quality parameters of each edge cloud node are represented as an array of parameter values, where the parameter values ​​of different parameters obtained during the same network test constitute a parameter value array. Specifically, in step 200, the resource recommendation weight of each edge cloud node is calculated based on the obtained network quality parameters, which may include: step 210, calculating the weighted average of the target parameters; the target parameters are any of the network quality parameters; step 220, obtaining the weight set corresponding to the network quality parameters; the weight set includes the weight values ​​of each target parameter.

[0065] Step 230: Based on the weight set, perform a weighted summation of the weighted average values ​​of each target parameter to obtain the resource recommendation weight of the edge cloud node.

[0066] First, calculate the weighted average of the target parameters, where the target parameter for any edge cloud node is any one of the network quality parameters. Taking the network quality parameters of each edge cloud node obtained by performing K tests on each edge cloud node as an example, including path, latency, jitter, and packet loss rate, assuming there is a resource pool of m edge cloud nodes, perform K network tests on the resource pool of each edge cloud node, and save the test results of each edge cloud node for each test. The test results of the network quality parameters obtained by each edge cloud node in a single test form a parameter value array, represented as F: [N,D,J,L].

[0067] In array F, N represents the number of device hops (i.e., the path) from a resource server in the resource pool to the target user's client. A smaller N value indicates fewer devices traversed from the user's client to the resource pool at the edge cloud node, representing a lower potential network failure risk; conversely, a larger N value indicates a higher potential network failure risk. Assuming the probability of a single device operating without failure is P, then the probability of the entire link operating without failure is P_0.05. N The probability of a network failure is (1-P) N The value of N indicates that the risk of network failure increases exponentially. D represents delay, indicating the latency from a resource server in the resource pool to the target user's client. J represents jitter, indicating network jitter from a resource server in the resource pool to the target user's client. L represents packet loss, indicating the packet loss rate from a resource server in the resource pool to the target user's client. For applications such as web conferencing, the packet loss rate should be controlled below 1%, as packet loss has a significant impact on network communication quality.

[0068] Taking latency D as an example, during K network tests, K parameter values ​​of latency D for a certain edge cloud node are obtained. The weighted average of latency D is calculated based on these K parameter values. Following this method, the weighted averages of other parameters in the network quality parameters of this edge cloud node are calculated, such as... and We obtain a weighted average array of network quality parameters for each edge cloud node.

[0069] Obtain the weight set corresponding to the network quality parameters. This weight set contains the weight values ​​of each target parameter in the network quality parameters. For any edge cloud node, based on the weight values ​​in this weight set, perform a weighted summation of the weighted average of each target parameter in its network quality parameters to obtain the resource recommendation weight of that edge cloud node.

[0070] For any edge cloud node, if the weighted average of each objective parameter in its network quality parameters is arranged into an array... The weight set corresponding to its network quality parameters is [λ1, λ2, λ3, λ4], where λ1, λ2, λ3, and λ4 are all configurable constants greater than 0, and λ1, λ2, λ3, and λ4 represent the weight values ​​of path N, latency D, jitter J, and packet loss rate L, respectively. Optionally, the weighted average of each target parameter in the network quality parameters of any edge cloud node is summed according to the following formula 1 to obtain the resource recommendation weight G of that edge cloud node:

[0071]

[0072] In Formula 1, the weighted average of path N is first logarithmized, and then a weighted sum is performed based on the logarithmic result. The resource recommendation weights calculated according to Formula 1 are used; a smaller G value indicates higher network quality. The recommended weight values ​​of the available resource pools for each edge cloud node are compared, and the resource pool with the smallest calculated G value is selected for resource activation for the target user.

[0073] Optionally, in step 210, calculating the weighted average of the target parameters may further include:

[0074] Step 211: Obtain the parameter value range of the target parameter; the parameter value range is determined by the maximum and minimum parameter values ​​of the target parameter;

[0075] Step 212: Divide the parameter value range into a preset number of sub-ranges, and count the number of targets whose parameter values ​​fall within each of the sub-ranges;

[0076] Step 213: Obtain the target weight of the sub-interval based on the ratio of the target quantity to the total number of parameter values ​​of the target parameters; the total number of parameter values ​​of the target parameters is the same as the number of network tests performed on each edge cloud node based on the IP address;

[0077] Step 214: Normalize the parameter values ​​of the target parameters based on the median of the sub-intervals and the target weights to obtain the weighted average value of the target parameters.

[0078] Based on multiple network tests on various edge cloud nodes, for the same target parameter in the network quality parameters, multiple parameter values ​​can be obtained. A parameter value range is formed based on its maximum and minimum values. The parameter value range of any target parameter is divided into a preset number of sub-ranges, and the number of targets whose parameter values ​​fall within each sub-range is counted. The target weight for each sub-range is obtained based on the ratio of the counted target values ​​to the total number of target parameter values. The total number of target parameter values ​​is the same as that used in the network tests on the edge cloud nodes. The target parameter values ​​are normalized based on the median of each sub-range and the calculated target weights to obtain a weighted average of the target parameters.

[0079] Specifically, taking latency D as the target parameter again, after performing K network tests on each edge cloud node, K parameter values ​​for latency D are obtained. Based on the maximum parameter value D... max and minimum parameter value D min The parameter value range that constitutes the time delay D is [D] max D min The parameter value range is divided into a preset number of sub-intervals, and the width of each sub-interval is: ΔD = (D max -D min ) / a, count the number of parameter values ​​in each sub-interval, that is, count D. min +(i-1)*ΔD≤D <D min The delay D of +i*ΔD has X parameters. i Then the parameter value of the time delay D is in the i-th subinterval [D min +(i-1)*ΔD,D min The probability within +i*ΔD) is: q i =X i / K, where the probability of the time delay D falling into each sub-interval is used as the weight of the sub-interval, and the median D of each sub-interval is taken. i =D min +(i-0.5)*ΔD, and perform data normalization according to the following formula 2 to obtain the weighted average value of the time delay D:

[0080]

[0081] Following the above method, the weighted average of each target parameter in the network quality parameters of any edge cloud node can be calculated, thereby obtaining the weighted average array corresponding to the network quality parameters of each edge cloud node.

[0082] In one embodiment, since the network quality parameters obtained from multiple network tests are simplified through weighted averaging and normalization, multiple edge cloud nodes may have the same resource recommendation weight, i.e., the same G value in Formula 1. When multiple target nodes with the best network quality (i.e., the target nodes with the smallest G value) are selected from the edge cloud nodes based on the resource recommendation weight, step 300, after selecting the target node with the best network quality from the edge cloud nodes based on the resource recommendation weight, may further include:

[0083] Step 301: Calculate the standard deviation of the network quality parameters of the target node based on the weighted average of each target parameter in the network quality parameters of any target node;

[0084] Step 302: Based on the weight set, perform a weighted summation of the standard deviations of the network quality parameters of the target node to obtain the verification value of the target node;

[0085] Step 303: Filter the target nodes according to the verification value.

[0086] Based on the weighted average of the network quality parameters of any target node, the standard deviation of the network quality parameters of that target node is calculated. Based on the calculated standard deviation of the network quality parameters of each target node, the target nodes are screened, and the best edge cloud node is selected from multiple target nodes for resource activation.

[0087] Optionally, based on K network tests of each edge cloud node, the standard deviation of each target parameter is calculated according to the weighted average of the network quality parameters of each edge cloud node, thus obtaining the standard deviation array corresponding to the edge cloud node. Taking the weighted average array of the edge cloud nodes as an example, the corresponding standard deviation array is H: [σ N ,σ D ,σ J ,σ L ].

[0088] Optionally, when filtering multiple target nodes with the same G value, the standard deviation array H of each target node is weighted and summed based on the weight set [λ1,λ2,λ3,λ4], and the resulting value is used as the verification value of the target node. Then, the target node with the smallest verification value is selected from the multiple target nodes for resource activation. Optionally, another weight set different from the weight set [λ1,λ2,λ3,λ4] can be used to perform a weighted summation of the standard deviation array to obtain the verification value, and then the target node with the smallest verification value is selected for resource activation. There are no specific limitations on this method.

[0089] In one embodiment, refer to Figure 2The cloud resource scheduling process shown first involves obtaining the resource ordering instruction from the target user. Based on the business requirements, resource requirements, and IP address in the ordering instruction, a network test instruction is generated. This instruction then performs K network tests on m edge cloud nodes (from edge cloud 1 to edge cloud m), obtaining K sets of network quality parameter values ​​for each edge cloud node. The parameter values ​​of different network quality parameters obtained during the same test are grouped into an array, resulting in K arrays for each edge cloud node, i.e., K arrays F1 to K arrays F... m Based on the K groups of F1 calculations for edge cloud node 1, a weighted average array is calculated. Following this method, a weighted average array for each edge cloud node is calculated, resulting in... Then, based on the preset weight set, the weighted average of the parameters in each weighted average array is summed to obtain the resource recommendation weight corresponding to each edge cloud node, i.e., G1-G. m Simultaneously, based on the calculated weighted average array, the standard deviation array of edge cloud 1 - edge cloud m is calculated, i.e., H1-H m Based on a preset set of weights, the standard deviation array is weighted and summed to calculate the check value of edge cloud 1 - edge cloud m, thus obtaining CHECK1 - CHECK. m .

[0090] Then, the target node with the smallest resource recommendation weight, i.e., G, is selected from edge clouds 1 to edge clouds m. min The corresponding target node. Determine if the target node is unique. If unique, the resource pool of that unique target node is selected as the optimal resource pool, and the optimal resource pool is used to activate resources for the target user. If the target node selected based on resource recommendation weights is not unique, the checksums of multiple target nodes are compared, and the target node with the smallest checksum is selected, i.e., CHECK. min The corresponding target node is designated as the optimal edge cloud node, and the resource pool of this optimal edge cloud node is designated as the optimal resource pool. This optimal resource pool is then used to activate resources for the target user.

[0091] In this embodiment, by conducting network tests on user IP addresses and calculating weights using network quality parameters, the system comprehensively considers the number of devices the user passes through to the resource pool and the network quality to arrive at a recommended weight value for opening the resource pool. This approach facilitates a comprehensive evaluation of each edge cloud node based on factors such as user distance, remaining resources, latency, jitter, and packet loss rate. It intelligently selects the edge nodes to provide business services to the user, completes the opening of business resources, effectively guarantees the quality of user business services, and achieves optimal scheduling of cloud resources from multiple aspects, including business service quality and resource requirements.

[0092] The cloud resource scheduling device provided in the embodiments of this application is described below. The cloud resource scheduling device described below can be referred to in correspondence with the cloud resource scheduling method described above.

[0093] Reference Figure 3 The cloud resource scheduling device provided in this application embodiment includes:

[0094] Network testing module 10 is used to perform network tests on each edge cloud node based on the IP address of the target user, and obtain the network quality parameters of each edge cloud node; the network quality parameters include path, latency, jitter and packet loss rate;

[0095] The weight calculation module 20 is used to calculate the resource recommendation weight of each edge cloud node based on the network quality parameters.

[0096] The resource scheduling module 30 is used to select the target node with the best network quality from each of the edge cloud nodes based on the resource recommendation weight, and schedule the resource pool of the target node to enable resources for the target user.

[0097] In one embodiment, the cloud resource scheduling device further includes an instruction acquisition module, used for:

[0098] Obtain information about the target user's global virtual private cloud and the specified IP address range, as well as the target user's configured business and resource requirements;

[0099] Based on the global virtual private cloud information, the IP address range, the resource requirements, and the business requirements, a resource ordering instruction for the target user is generated.

[0100] In one embodiment, the network testing module 10 is further configured to:

[0101] Based on a preset number of tests, test commands are generated based on the target user's IP address during each test.

[0102] The test command is sent to each edge cloud node to perform network testing, and the network quality parameters of each edge cloud node are obtained.

[0103] In one embodiment, the weight calculation module 20 is further configured to:

[0104] Calculate the weighted average of the target parameters; the target parameters are any one of the network quality parameters.

[0105] Obtain the weight set corresponding to the network quality parameters; the weight set includes the weight value of each target parameter; based on the weight set, perform a weighted summation of the weighted average of each target parameter to obtain the resource recommendation weight of the edge cloud node.

[0106] In one embodiment, the weight calculation module 20 is further configured to:

[0107] Obtain the parameter value range of the target parameter; the parameter value range is determined by the maximum and minimum parameter values ​​of the target parameter;

[0108] The parameter value range is divided into a preset number of sub-ranges, and the number of targets whose parameter values ​​fall within each of the sub-ranges is counted.

[0109] The target weight of the sub-interval is obtained based on the ratio of the target quantity to the total number of target parameter values; the total number of target parameter values ​​is the same as the number of network tests performed on each edge cloud node based on the IP address; the target parameter values ​​are normalized based on the median of the sub-interval and the target weight to obtain the weighted average of the target parameters.

[0110] In one embodiment, if the target nodes include multiple nodes, the resource scheduling module 30 is further configured to:

[0111] Calculate the standard deviation of the network quality parameters of any target node based on the weighted average of the network quality parameters of the target node.

[0112] The standard deviations of the network quality parameters of the target node are weighted and summed based on the weight set to obtain the verification value of the target node.

[0113] The target nodes are filtered based on the verification value.

[0114] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call a computer program in the memory 430 to execute steps of a cloud resource scheduling method, such as:

[0115] Network tests are performed on each edge cloud node based on the IP address of the target user to obtain the network quality parameters of each edge cloud node.

[0116] Calculate the resource recommendation weight for each edge cloud node based on the network quality parameters;

[0117] Based on the resource recommendation weight, the target node with the best network quality is selected from each of the edge cloud nodes, and the resource pool of the target node is scheduled to activate resources for the target user.

[0118] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the cloud resource scheduling method provided in the above embodiments, such as including:

[0120] Network tests are performed on each edge cloud node based on the IP address of the target user to obtain the network quality parameters of each edge cloud node.

[0121] Calculate the resource recommendation weight for each edge cloud node based on the network quality parameters;

[0122] Based on the resource recommendation weight, the target node with the best network quality is selected from each of the edge cloud nodes, and the resource pool of the target node is scheduled to activate resources for the target user.

[0123] On the other hand, embodiments of this application also provide a processor-readable storage medium storing a computer program, the computer program being used to cause a processor to execute the steps of the cloud resource scheduling method provided in the above embodiments, for example including:

[0124] Network tests are performed on each edge cloud node based on the IP address of the target user to obtain the network quality parameters of each edge cloud node.

[0125] Calculate the resource recommendation weight for each edge cloud node based on the network quality parameters;

[0126] Based on the resource recommendation weight, the target node with the best network quality is selected from each of the edge cloud nodes, and the resource pool of the target node is scheduled to activate resources for the target user.

[0127] The processor-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0128] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0129] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A cloud resource scheduling method, characterized in that, include: Network tests are performed on each edge cloud node based on the IP address of the target user to obtain the network quality parameters of each edge cloud node. Calculate the resource recommendation weight for each edge cloud node based on the network quality parameters; Based on the resource recommendation weight, the target node with the best network quality is selected from each of the edge cloud nodes, and the resource pool of the target node is scheduled to open resources for the target user. Before performing network tests on each edge cloud node based on the target user's IP address to obtain the network quality parameters of each edge cloud node, the method further includes: Obtain information about the target user's global virtual private cloud and the specified IP address range, as well as the target user's configured business and resource requirements; Based on the global virtual private cloud information, the IP address range, the resource requirements, and the business requirements, a resource ordering instruction for the target user is generated.

2. The cloud resource scheduling method according to claim 1, characterized in that, The calculation of resource recommendation weights for each edge cloud node based on the network quality parameters includes: Calculate the weighted average of the target parameters; the target parameters are any one of the network quality parameters. Obtain the weight set corresponding to the network quality parameters; the weight set includes the weight value of each target parameter; The resource recommendation weight of the edge cloud node is obtained by weighting and summing the weighted average values ​​of each target parameter based on the weight set.

3. The cloud resource scheduling method according to claim 2, characterized in that, The calculation of the weighted average of the target parameters includes: Obtain the parameter value range of the target parameter; the parameter value range is determined by the maximum and minimum parameter values ​​of the target parameter; The parameter value range is divided into a preset number of sub-ranges, and the number of targets whose parameter values ​​fall within each of the sub-ranges is counted. The target weight of the sub-interval is obtained based on the ratio of the target quantity to the total number of parameter values ​​of the target parameters; the total number of parameter values ​​of the target parameters is the same as the number of network tests performed on each edge cloud node based on the IP address. The target parameter values ​​are normalized based on the median of the sub-intervals and the target weights to obtain the weighted average of the target parameters.

4. The cloud resource scheduling method according to claim 2, characterized in that, If the target nodes include multiple nodes, after selecting the target node with the best network quality from each edge cloud node based on the resource recommendation weight, the method further includes: Calculate the standard deviation of the network quality parameters of any target node based on the weighted average of the network quality parameters of each target node; The standard deviations of the network quality parameters of the target node are weighted and summed based on the weight set to obtain the verification value of the target node. The target nodes are filtered based on the verification value.

5. The cloud resource scheduling method according to claim 1, characterized in that, The network testing of each edge cloud node based on the target user's IP address yields network quality parameters for each edge cloud node, including: Based on a preset number of tests, test commands are generated based on the target user's IP address during each test. The test command is sent to each edge cloud node to perform network testing, and the network quality parameters of each edge cloud node are obtained.

6. A cloud resource scheduling device, characterized in that, include: The network testing module is used to perform network tests on each edge cloud node based on the IP address of the target user, and to obtain the network quality parameters of each edge cloud node. The weight calculation module is used to calculate the resource recommendation weight of each edge cloud node based on the network quality parameters. The resource scheduling module is used to select the target node with the best network quality from each of the edge cloud nodes based on the resource recommendation weight, and schedule the resource pool of the target node to enable resources for the target user. It also includes an instruction retrieval module, used for: Obtain information about the target user's global virtual private cloud and the specified IP address range, as well as the target user's configured business and resource requirements; Based on the global virtual private cloud information, the IP address range, the resource requirements, and the business requirements, a resource ordering instruction for the target user is generated.

7. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the cloud resource scheduling method according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the cloud resource scheduling method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the cloud resource scheduling method as described in any one of claims 1 to 5.

Images