Resource scheduling method, device and product

Abstract

The invention relates to the technical field of communication, and discloses a resource scheduling method, device and product. The method comprises the following steps of: screening all candidate nodes for a target combination by taking a combination of an analysis group and a region as granularity; determining a basic weight of each candidate node based on a preferred sequence corresponding to the region to which each candidate node belongs; calculating the geographic distance between the region to which each candidate node belongs and the target region; obtaining a current performance index and a historical performance expression score of each candidate node; and calculating a comprehensive score of the candidate nodes according to the geographic distance, the basic weight, the performance index and the historical performance expression score corresponding to each candidate node, so as to determine the candidate node with the highest priority as a target node corresponding to the target combination, and scheduling the CDN service request in the target area to the target node. By adopting the method, flexible scheduling of CDN node resources can be carried out based on accurate geographic positions, and the utilization rate of network resources and the accuracy of scheduling decisions are improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, specifically to a resource scheduling method, apparatus, and product. Background Technology

[0002] In large-scale CDN (Content Delivery Network) networks and edge computing environments, the efficiency and accuracy of node resource scheduling directly affect service quality and user experience. With the widespread application of 5G, IoT, and cloud computing technologies, network traffic is experiencing explosive growth, and users' demand for low-latency, high-reliability CDN services is becoming increasingly urgent.

[0003] Current CDN node resource scheduling strategies divide node resource pools based on whether they are within the same region or not. When selecting node resources, the scheduling system often uses a coarse approach based on network operators and resource pools within the same region, neglecting the influence of geographical factors. This leads to inaccurate resource scheduling decisions, low network resource utilization, and a negative impact on user experience. For example, when ranking available resource nodes, consideration is given only to network operators and regional divisions, ignoring the potential resource efficiency differences caused by geographical distances between different provinces within the same region. This results in selected nodes experiencing high network latency and low data transmission stability when providing CDN services, negatively affecting user experience. Summary of the Invention

[0004] The purpose of this application is to provide a resource scheduling method, apparatus, and product to solve the problem that in existing CDN scheduling systems, node resource scheduling strategies cannot accurately perceive geographical location, resulting in inaccurate resource scheduling decisions and affecting user experience.

[0005] To achieve the above objectives, the technical solution of this application is as follows: In a first aspect, embodiments of this application provide a resource scheduling method, which is applied to a CDN network scheduling system; the CDN network includes: multiple resolution groups and nodes deployed in different regions; the method includes: Using the combination of parsing groups and regions as the granularity, all candidate nodes in the CDN network are screened for the target combination; the target combination consists of a target region and its corresponding parsing group. The basic weight of each candidate node is determined based on the preferred order of the regions to which each candidate node belongs. Based on the location information of the target combination and the location information of each candidate node, calculate the geographical distance between the region to which each candidate node belongs and the target region; Obtain the current performance metrics and historical performance scores of each candidate node; the performance metrics represent the current service quality of the candidate node; the historical performance scores represent the stability of the service quality of the candidate node within a first time period. A comprehensive score is calculated for each candidate node based on its geographical distance, basic weight, performance indicators, and historical performance scores; the comprehensive score is used to characterize the priority of the candidate node. Based on the comprehensive scores of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, and CDN service requests within the target area are scheduled to the target node.

[0006] Optionally, all candidate nodes in the CDN network are selected for the target combination, including: Determine the region to which the target area belongs; the region must contain at least two sub-regions; Based on the parsing group, the corresponding business type is determined. According to the business type, a first type of node capable of providing the service type is obtained from the node resource pool of the region; and, according to the business type, a second type of node capable of providing the service type is obtained from the node resource pool of the neighboring province of the target region. The first type of nodes and the second type of nodes are added to the candidate set and deduplicated. All nodes in the deduplicated candidate set are then determined as candidate nodes.

[0007] Optionally, the method further includes: Obtain the sequence configuration information corresponding to the target combination; the sequence configuration information is used to specify at least one preferred region, and the preferred region has a corresponding preferred order; The process of selecting all candidate nodes in the CDN network for the target combination further includes: according to the service type, obtaining a third type of node that can provide the service type from the node resource pool of each preferred region and adding it to the candidate set; and deduplicating the candidate set.

[0008] Optionally, the sequence configuration information further includes an effective time period; determining the basic weight of each candidate node specifically includes: If the current time is outside the effective period, set the same first basic weight for each candidate node; Given that the current time is within the effective period, obtain the arrangement order of each preferred region in the sequence configuration information; According to the order of arrangement, from first to last, a second basic weight is set for the third type of node corresponding to each preferred region, from smallest to largest; the second basic weight is less than the first basic weight.

[0009] Optionally, based on the location information of the target combination and the location information of each candidate node, the geographical distance between the region to which each candidate node belongs and the target region is calculated, including: Obtain the latitude and longitude information of the target combination and the latitude and longitude information of each candidate node; Calculate the longitude difference and latitude difference between the latitude and longitude information of each candidate node and the latitude and longitude information of the target combination; Based on the Earth's radius and the longitude and latitude differences corresponding to each candidate node, the arc distance between the candidate node and the corresponding position of the target area is calculated.

[0010] Optionally, after calculating the arc distance between the candidate node and the corresponding position in the target region, the method further includes: The arc distance corresponding to each candidate node is compared with a first distance threshold and a second distance threshold to determine the value range of the arc distance; the second distance threshold is greater than the first distance threshold. Based on the value range of each arc distance, nonlinear normalization is performed on each arc distance to obtain the distance influence value corresponding to the candidate node; wherein, when the arc distance is within a first value range, the arc distance is expanded, and when the arc distance is within a second value range, the arc distance is reduced; the first value range is the value range less than the first distance threshold; the second value range is the value range greater than or equal to the second distance threshold.

[0011] Optionally, a comprehensive score is calculated for each candidate node based on its geographical distance, basic weight, performance metrics, and historical performance score, including: Calculate the order parameter based on the basic weight and segmentation cardinality corresponding to the candidate node; Based on the geographical factors and the distance influence values ​​corresponding to the candidate nodes, calculate the geographical parameter items; Based on the geographical factors and the current performance indicators of the candidate nodes, calculate the performance parameter items; Based on the historical performance scores and time factors of the candidate nodes, calculate the historical performance parameters. The comprehensive score of the candidate node is calculated based on the geographic parameter item, the performance parameter item, and the historical performance parameter item.

[0012] Optionally, the method further includes: Determine the service type of the CDN service that the target combination needs to provide; the service type is any of the following: real-time interactive service, content distribution service, or service with high reliability requirements; When the business type is a real-time interactive business, set the corresponding first geographical factor; When the business type is a content distribution business, a corresponding second geographical factor is set; the second geographical factor is greater than the first geographical factor. When the business type is a high-reliability requirement business, a corresponding third geographical factor is set; the third geographical factor is greater than the second geographical factor.

[0013] Optionally, obtain the current performance metrics and historical performance scores of each candidate node, including: Obtain multiple performance test scores and corresponding test times for the candidate node within the first time period; the multiple performance test scores are obtained by conducting multiple tests on the service quality of the candidate node. Based on the multiple performance test scores and corresponding test times at the current moment, calculate the historical performance score of the candidate node within the first time period; and determine the performance test score obtained from the most recent test at the current moment as the performance index of the candidate node.

[0014] Optionally, based on the comprehensive scores of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, including: Based on the comprehensive scores of each candidate node, sort the candidate nodes from lowest to highest to generate a candidate list; Iterate through the candidate list and compare the current node load rate of each candidate node in the candidate list with the target load threshold. If the node load rate of the currently traversed candidate node is lower than the target load threshold, the candidate node is determined as the target node.

[0015] Optionally, after determining the candidate node as the target node, the method further includes: The real-time node load rate of the target node is continuously acquired at the first interval and compared with the target load threshold. If the real-time node load rate of the target node is greater than or equal to the target load threshold, the comprehensive scores of the other candidate nodes in the candidate list other than the target node are recalculated, and the candidate list is updated. Iterate through the updated candidate list and compare the current node load rate of each candidate node in the candidate list with the target load threshold; If the node load rate of the currently traversed candidate node is lower than the target load threshold, the candidate node is determined as the new target node.

[0016] Secondly, embodiments of this application provide a resource scheduling apparatus, deployed in a CDN network, for implementing the steps of the method provided in the first aspect of this application. The apparatus includes: The management module is configured to filter all candidate nodes in the CDN network at the granularity of the combination of parsing groups and regions; the target combination consists of a target region and its corresponding parsing group; and the basic weight of each candidate node is determined based on the preferred order of the regions to which each candidate node belongs. The distance calculation module is configured to calculate the geographical distance between the region to which each candidate node belongs and the target region based on the location information of the target combination and the location information of each candidate node; The monitoring module is configured to obtain the current performance indicators and historical performance scores of each candidate node; the performance indicators represent the current service quality of the candidate node; and the historical performance scores represent the stability of the service quality of the candidate node within a first time period. The comprehensive analysis module is configured to calculate the comprehensive score of each candidate node based on its geographical distance, basic weight, performance indicators, and historical performance scores; the comprehensive score is used to characterize the priority of the candidate nodes. The scheduling module is configured to determine the candidate node with the highest priority as the target node corresponding to the target combination based on the comprehensive score of each candidate node, and to schedule CDN service requests in the target area to the target node.

[0017] Thirdly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0018] Fourthly, embodiments of this application provide a server, which includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0019] Fifthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0020] The resource scheduling method provided in this application selects nodes to provide CDN services to users in the target region, using the combination of parsing groups and regions as the granularity. When selecting node resources, this application combines multiple dimensions of the candidate nodes, such as their geographical location, real-time performance indicators, and historical performance, with the user factor of a customizable node selection order for maintenance personnel. This comprehensive analysis of the resource advantages of each candidate node generates a comprehensive score for each candidate node. Then, based on the comprehensive scores of each candidate node, the optimal node resources are determined for the target combination, thereby generating a resource scheduling decision that accurately perceives geographical location and user flexibility needs, improving network resource utilization and enhancing the end-user experience in the target region. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a resource scheduling method proposed in an embodiment of this application; Figure 2 This is a flowchart of node resource scheduling in a scheduling system according to an embodiment of this application; Figure 3 This is a schematic diagram illustrating the calculation of the comprehensive score of candidate nodes in one embodiment of this application; Figure 4 This is a schematic diagram of a resource scheduling device proposed in an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. 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.

[0024] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0025] In the various embodiments of this application, it should be understood that the sequence number of each process described below does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0026] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects as detailed in this application.

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0028] In a CDN network, the scheduling system divides the network coverage area into several logical units called regions. Regions are typically based on geographical location (e.g., province, city) combined with network operators (e.g., China Telecom, China Mobile, China Unicom, etc.). For example, China Telecom Fujian, China Mobile Fujian, China Telecom Guangdong, China Mobile Guangdong, etc. A larger geographical unit division above the regional level is called a region. Each region has a node resource pool containing all nodes deployed within that region. The set of nodes in the node resource pools of all regions belonging to the same region is called region-wide resources. Generally, region-wide resources have lower network latency than non-region-wide resources. However, due to the fixed division of geographical regions, there are often situations where the network latency of nodes in neighboring provinces within a certain region is lower than that of nodes in more distant provinces or cities within the same region.

[0029] Traditional node resource scheduling schemes typically rely on a rough selection based on carrier and regional resource pools. This approach treats nodes within the same region as higher-priority primary resources and nodes outside the same region as lower-priority secondary resources. However, this scheduling method ignores geographical differences and fails to accurately distinguish the quality of resources within the same region, neighboring provinces, or even resources outside the same region. This results in insufficient precision in node resource scheduling decisions, leading to problems such as high network latency and low data transmission stability.

[0030] This application introduces the precise geographical location of nodes and user-defined selection order configuration, and combines multi-dimensional information to comprehensively sort node resources, thereby achieving efficient, flexible and accurate scheduling of node resources, improving transmission stability and latency, and enhancing network resource utilization.

[0031] The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0032] Figure 1 This is a flowchart of a resource scheduling method proposed in an embodiment of this application. The method is applied to a CDN network scheduling system; the CDN network includes multiple resolution groups and nodes deployed in different regions. Figure 1 As shown, the method includes: S1: Using the combination of parsing groups and regions as the granularity, filter all candidate nodes in the CDN network for the target combination; the target combination consists of a target region and its corresponding parsing group. S2: Determine the basic weight of each candidate node based on the preferred order corresponding to the region to which each candidate node belongs; S3: Based on the location information of the target combination and the location information of each candidate node, calculate the geographical distance between the region to which each candidate node belongs and the target region; S4: Obtain the current performance metrics and historical performance scores of each candidate node; the performance metrics represent the current service quality of the candidate node; the historical performance scores represent the stability of the service quality of the candidate node within a first time period; S5: Calculate the comprehensive score of each candidate node based on its geographical distance, basic weight, performance index, and historical performance score; the comprehensive score is used to characterize the priority of the candidate node. S6: Based on the comprehensive score of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, and CDN service requests in the target area are scheduled to the target node.

[0033] In this embodiment, node resource scheduling in the scheduling system is based on the combination of parsing groups and regions, and the system performs node resource scheduling for each combination. The system provides users with configuration management services, allowing users to customize the region preference order to achieve flexible node resource scheduling. When scheduling node resources for a target combination, firstly, based on the target region and corresponding parsing group in the target combination, all nodes meeting the requirements are selected from the CDN network as candidate nodes. Specifically, different parsing groups typically correspond to different service types within the CDN service. When selecting candidate nodes, it is necessary to consider the service type of the parsing group and the restrictions on node deployment locations imposed by the parsing group. For example, if the service type of the parsing group is real-time interactive service, and the restricted node deployment location is a neighboring province, nodes capable of supporting real-time interactive services are selected from all neighboring province node resource pools in the target region and determined as candidate nodes for the target combination.

[0034] Obtain multi-dimensional information about each candidate node, including the node's own information and the user-defined preferred order configuration. Specifically, this includes: the node's current performance metrics, historical performance scores, geographical distance between the node's region and the target region, and, if the node belongs to a user-specified preferred region, obtaining the preferred order of the preferred region to determine the basic weight of the candidate nodes within that preferred region.

[0035] Based on the multi-dimensional information obtained from each candidate node, a comprehensive score is calculated to determine the priority ranking of each candidate node when selecting the CDN service node (i.e., the target node) corresponding to the target combination. Finally, based on the comprehensive scores of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, providing the corresponding CDN service to users in that target area. Therefore, in subsequent CDN service operations, CDN service requests sent from user devices within the target area will be scheduled to this target node, which will then provide the corresponding service type CDN service to users in that area.

[0036] In this embodiment, the precise geographical location of candidate nodes and the user-defined regional preference order are incorporated into the resource scheduling decision. A comprehensive analysis is performed on multi-dimensional information, including node performance data, the geographical location of the node's region, and user-defined requirements, to determine the priority ranking of each candidate node. Compared to the coarse scheduling method in traditional solutions that only considers operator and regional divisions, this approach enables efficient, flexible, and precise scheduling of node resources, effectively improving transmission stability and latency, thereby increasing network resource utilization, enhancing user experience, reducing unnecessary resource waste, and lowering operating costs. Furthermore, because this solution incorporates user-defined preference order configurations into resource scheduling decisions, it can quickly respond to dynamic changes in business needs, improve resource scheduling efficiency, and has strong adaptability to scheduling requirements in different business scenarios.

[0037] This solution is applicable to a variety of network service scenarios, such as CDN scheduling system optimization, improving the online loading speed of video streaming, enhancing the smoothness of online games, and improving the access speed of large websites. It performs particularly well in multi-region and multi-service scenarios.

[0038] As one embodiment of this application, all candidate nodes in the CDN network are selected for the target combination, including: Determine the region to which the target area belongs; the region must contain at least two sub-regions; Based on the parsing group, the corresponding business type is determined. According to the business type, a first type of node capable of providing the service type is obtained from the node resource pool of the region; and, according to the business type, a second type of node capable of providing the service type is obtained from the node resource pool of the neighboring province of the target region. The first type of nodes and the second type of nodes are added to the candidate set and deduplicated. All nodes in the deduplicated candidate set are then determined as candidate nodes.

[0039] In one embodiment, when filtering candidate nodes, corresponding candidate nodes are selected from the node resource pools of neighboring provinces of the target region and the region to which the target region belongs. Specifically, the corresponding service type is determined according to the parsing group. Based on the service type, nodes capable of providing CDN services for that service type are obtained from the node resource pools of the region to which the target region belongs and the node resource pools of neighboring provinces, respectively, and are identified as candidate nodes and added to the candidate set. Since neighboring provinces may exist in the region to which the target region belongs, the candidate set needs to be deduplicated to avoid duplication. After deduplication, the remaining nodes in the candidate set are identified as candidate nodes.

[0040] As one embodiment of this application, the method further includes: Obtain the sequence configuration information corresponding to the target combination; the sequence configuration information is used to specify at least one preferred region, and the preferred region has a corresponding preferred order; The process of selecting all candidate nodes in the CDN network for the target combination further includes: according to the service type, obtaining a third type of node that can provide the service type from the node resource pool of each preferred region and adding it to the candidate set; and deduplicating the candidate set.

[0041] In one embodiment, the scheduling system supports differentiated settings for sequential configuration information based on a combination of parsing groups and regions. The parsing group specifies the parsing group for which a custom preferred order should be applied, and the region specifies the target region where the configuration is applied. In the sequential configuration information, the user specifies one or more preferred regions, causing the system to prioritize nodes in these preferred regions during resource scheduling. This alters the resource priority of specified regions in the original scheduling strategy (e.g., resources within the same region have higher priority than resources outside the same region), thereby enabling flexible adjustment of the resource scheduling strategy according to user needs. In this embodiment, the scheduling system supports dynamic updates to the sequential configuration information, which take effect immediately after the configuration information is updated. When the number of preferred regions is not less than two, the user, in addition to specifying the preferred regions, also specifies the preferred order for each preferred region, thereby determining the priority order of nodes selected in each preferred region. The earlier a preferred region appears in the preferred order, the higher the priority of its nodes.

[0042] For example, if the preferred region A is ranked first and the preferred region B is ranked second, then the priority of the candidate node belonging to preferred region A is higher than the priority of the candidate node belonging to preferred region B.

[0043] In this embodiment, based on user-defined sequence configuration information, when filtering candidate nodes corresponding to target combinations, the system also needs to consider the node resource pools of all preferred regions specified in the sequence configuration information. Third-type nodes that can provide services of that service type are selected from the node resource pools of each preferred region according to the service type and added to the candidate set, and the candidate set is deduplicated.

[0044] As one embodiment of this application, the sequence configuration information further includes an effective time period; determining the basic weight of each candidate node specifically includes: If the current time is outside the effective period, set the same first basic weight for each candidate node; Given that the current time is within the effective period, obtain the arrangement order of each preferred region in the sequence configuration information; According to the order of arrangement, from first to last, a second basic weight is set for the third type of node corresponding to each preferred region, from smallest to largest; the second basic weight is less than the first basic weight.

[0045] In one embodiment, when setting the sequence configuration information, the user also includes an effective period, which indicates the effective and ineffective times of the preferred regions, supporting timed strategies and emergency scheduling needs. During the effective period, the base weights of candidate nodes in the preferred regions are adjusted from the default first base weight to a second base weight with higher priority. Outside the effective period, this sequence configuration information automatically expires, and the base weights of candidate nodes in each preferred region revert to the default base weights (first base weights). In this embodiment, a smaller value represents a higher priority weight, meaning the value of the second base weight is less than the value of the first base weight. The default base weight value can be customized as needed, for example, set to 99.

[0046] When multiple preferred regions are specified, the candidate nodes in each preferred region are assigned basic weights from smallest to largest according to their preferred order (i.e., from highest to lowest priority). For example, the candidate node in the first preferred region has a basic weight of 0, and the candidate node in the second preferred region has a basic weight of 1.

[0047] Optionally, the sequence configuration information also includes custom geographical and temporal factors. The geographical factor controls the weight of geographical distance on node resource ranking, while the temporal factor controls the weight of historical performance on node resource ranking. Users can adjust the values ​​of the geographical and temporal factors according to the needs of their applications to change the priority ranking of node resources.

[0048] As one embodiment of this application, the geographical distance between the region to which each candidate node belongs and the target region is calculated based on the location information of the target combination and the location information of each candidate node, including: Obtain the latitude and longitude information of the target combination and the latitude and longitude information of each candidate node; Calculate the longitude difference and latitude difference between the latitude and longitude information of each candidate node and the latitude and longitude information of the target combination; Based on the Earth's radius and the longitude and latitude differences corresponding to each candidate node, the arc distance between the candidate node and the corresponding position of the target area is calculated.

[0049] In one embodiment, the system obtains the location information of the regions where each candidate node is located through a management module. Optionally, when the scheduling system starts, the management module obtains the location information of the regions corresponding to each service node connected to the system. Based on the location information of each region, the distance calculation module pre-calculates the geographical distances between all regions and stores them in Excel or CSV format for easy retrieval later without recalculation, thereby accelerating the efficiency of node resource scheduling. The scheduling system supports adding, modifying, deleting, querying, batch importing, and exporting location information. If the location information of any region changes, the distance calculation module recalculates and updates all geographical distances related to that region.

[0050] In this embodiment, the location information of the region includes the following fields: Geographic Hierarchy ID: A unique identifier for a region; Geographic hierarchical name: A readable name of a region, such as a province name; Longitude: The longitude coordinates of this region, expressed in decimal system. Latitude: The latitude coordinates of this region, expressed in decimal system.

[0051] In this embodiment, the arc distance between the region to which each candidate node belongs and the target region is calculated, and this arc distance is taken as the geographical distance between the regions. Compared with the method of calculating the straight-line distance based on the latitude and longitude coordinates of the region, a more accurate true physical distance between regions can be obtained, making node resource decisions more accurate, thereby further reducing network latency and improving the stability and speed of data transmission.

[0052] Specifically, considering the curvature of the Earth, the Haversine algorithm is used to calculate the arc distance, and the specific expression is as follows: The intermediate variable a is calculated as follows: ; in, , These are the latitudes of the two locations; The difference in latitude between the two locations; This represents the difference in longitude between the two locations; The central angle c between the two positions is: ; The arc distance d between the two positions is: Where R is the Earth's radius, approximately 6371 kilometers.

[0053] Figure 2 This is a flowchart illustrating node resource scheduling in a scheduling system according to one embodiment of this application. For example... Figure 2As shown, the scheduling system obtains user-defined sequence configuration information through the configuration management service and determines the basic weights corresponding to candidate nodes in each preferred region based on this configuration information. It also obtains the location information of the region to which each candidate node belongs, as well as the location information of the target region, through the configuration management service, and calculates the geographical distance between each region and the target region. The system also monitors the node status of each candidate node through a monitoring module, obtaining real-time performance indicators and historical performance scores. Furthermore, based on the node's basic weight, geographical distance, performance indicators, and historical performance scores, a comprehensive score is calculated to determine the priority ranking of each candidate node. Since this embodiment uses smaller numerical values ​​to represent higher priorities, the candidate node with the lowest comprehensive score is the highest priority candidate node. The highest priority candidate node is determined as the target node corresponding to the target combination, and node resource scheduling is performed, scheduling CDN service requests of the corresponding business type from the target region to this target node.

[0054] As one embodiment of this application, after calculating the arc distance between the candidate node and the position corresponding to the target region, the method further includes: The arc distance corresponding to each candidate node is compared with a first distance threshold and a second distance threshold to determine the value range of the arc distance; the second distance threshold is greater than the first distance threshold. Based on the value range of each arc distance, nonlinear normalization is performed on each arc distance to obtain the distance influence value corresponding to the candidate node; wherein, when the arc distance is within a first value range, the arc distance is expanded, and when the arc distance is within a second value range, the arc distance is reduced; the first value range is the value range less than the first distance threshold; the second value range is the value range greater than or equal to the second distance threshold.

[0055] In one embodiment, to make the differences in geographical distance between candidate nodes more obvious, after calculating the arc distance (i.e., geographical distance) between each node and the location corresponding to the target area, the arc distance is non-linearly normalized to map the geographical distance to a larger range of values.

[0056] Specifically, a smaller first distance threshold (e.g., 50km) and a larger second distance threshold (e.g., 500km) are set. The arc distance of each candidate node is compared with the two distance thresholds, and the arc distance is normalized according to the value range of the arc distance.

[0057] Specifically, the differences in arc distance d within the first value range (d < 50 km) are amplified to highlight the impact of geographical distance differences within this range; the differences in arc distance d within the second value range (d ≥ 500 km) are moderately compressed to reduce the impact of geographical distance differences within this range; and the differences in arc distance d within the third value range (50 km ≤ d < 500 km) are significantly compressed to minimize the impact of geographical distance factors when comparing different candidate nodes within this distance range. For example, the geographical differences between two candidate nodes whose geographical distances both exceed 500 km can be almost ignored.

[0058] In this embodiment, based on the Haversine algorithm for accurate distance calculation, the arc distance of each candidate node is nonlinearly normalized, making the scheduling system more sensitive to nearby node resources. When selecting target nodes in the future, the system will focus more on selecting nearby candidate nodes, while also reasonably handling distant resources, thereby improving the accuracy and efficiency of resource scheduling decisions.

[0059] In one embodiment, the first distance threshold is 50km, and the second distance threshold is 500km. The arc distance d (i.e., geographical distance) of the candidate nodes is mapped to achieve non-linear normalization, yielding the corresponding distance influence value normalizedDist. The specific method is as follows: Geographical distance within the first range (d < 50 km): normalizedDist = d÷50×200; R is the average radius of the Earth; Geographical distance within the second range (d ≥ 500km): normalizedDist = 600+(d-500)÷(R×π÷2-500)×399; Geographical distance within the third range (50km ≤ d < 500km): normalizedDist = 200+(d-50)÷450×400.

[0060] Furthermore, this embodiment also limits the upper limit of the distance influence value to 999, that is, limits the range of the distance influence value to [0, 999]. Therefore, when the value of the distance influence value is greater than this upper limit, the distance influence value is modified to 999. That is, if normalizedDist > 999, then normalizedDist = 999.

[0061] As one implementation of this application, a comprehensive score for each candidate node is calculated based on its geographical distance, basic weight, performance indicators, and historical performance scores, including: Calculate the order parameter based on the basic weight and segmentation cardinality corresponding to the candidate node; Based on the geographical factors and the distance influence values ​​corresponding to the candidate nodes, calculate the geographical parameter items; Based on the geographical factors and the current performance indicators of the candidate nodes, calculate the performance parameter items; Based on the historical performance scores and time factors of the candidate nodes, calculate the historical performance parameters. The comprehensive score of the candidate node is calculated based on the geographic parameter item, the performance parameter item, and the historical performance parameter item.

[0062] Figure 3 This is a schematic diagram illustrating the calculation of the comprehensive score of candidate nodes in one embodiment of this application. For example... Figure 3 As shown, in one embodiment, the comprehensive score of each candidate node is calculated based on its own multi-dimensional factors and user-defined order configuration information. The specific steps are as follows: (1) Determine the basic weight of the candidate nodes in each preferred region according to the preferred order in the sequential configuration information. For candidate nodes that do not belong to a preferred region, use the default first basic weight. Also, obtain the preset segmentation base. The segmentation base is used to amplify the numerical difference between different basic weights, thereby ensuring a clear separation between node resources of different priorities. Calculate the sequence parameter based on the basic weight and the segmentation base. This parameter indicates the importance of the preferred order configured by the user to the selection of the target node, where the smaller the value, the more important it is; (2) Obtain preset geographic factors based on the analytical group in the target combination. Perform nonlinear normalization based on the geographic distances corresponding to candidate nodes to obtain distance influence values. Calculate geographic parameter terms based on the distance influence values ​​and the geographic factors. These parameter terms indicate the importance of the precise geographic location of the node to the selection of target nodes, with smaller values ​​indicating greater importance; (3) Obtain the current performance indicators of the candidate nodes, and calculate the performance parameter based on the geographical factors and the performance indicators. This parameter indicates the importance of the current service quality of the node to the selection of the target node, where the smaller the value, the more important it is; (4) Obtain the historical performance score of the candidate node and the preset time factor. Based on the historical performance score and the time factor, calculate the historical performance parameter. This parameter indicates the importance of the historical service quality stability of the node to the selection of the target node, where the smaller the value, the more important it is; (5) Calculate the comprehensive score of the candidate node based on each parameter. Optionally, the expression for calculating the comprehensive score W of the candidate node is as follows: ; Where B is the basic weight; S is the segmentation base; and D is... Geographical factor; T is historical performance score; The time factor.

[0063] In this embodiment, by combining the multi-dimensional factors of the candidate nodes themselves and the factors of the user-defined preferred order, a comprehensive score is calculated to achieve a comprehensive analysis of the candidate nodes. This ensures that when selecting a target node, factors such as the node's precise geographical distance, the user's flexible customized preferred order, the node's own performance, and the stability of its historical service quality are considered simultaneously. This determines the optimal target node, enabling more flexible and accurate node resource scheduling decisions, improving the overall resource utilization of the network, and enhancing the user experience.

[0064] As one embodiment of this application, the method further includes: Determine the service type of the CDN service that the target combination needs to provide; the service type is any of the following: real-time interactive service, content distribution service, or service with high reliability requirements; When the business type is a real-time interactive business, set the corresponding first geographical factor; When the business type is a content distribution business, a corresponding second geographical factor is set; the second geographical factor is greater than the first geographical factor. When the business type is a high-reliability requirement business, a corresponding third geographical factor is set; the third geographical factor is greater than the second geographical factor.

[0065] In one embodiment, geographical factors are flexibly adjusted to adapt to business needs based on the business type corresponding to the target combination. Optionally, CDN business types include the following three categories: real-time interactive services, content distribution services, and services requiring high reliability. Real-time interactive services require low latency, two-way, and continuous communication, including video conferencing, live streaming, online gaming, and real-time audio and video calls; content distribution services include video-on-demand, software downloads, and access to news portals; and services requiring high reliability have extremely high requirements for service continuity, availability, and stability, including financial transaction services and emergency public services.

[0066] In this embodiment, corresponding geographical factors are set for the above three business types. The third geographical factor corresponding to high reliability requirements is greater than the second geographical factor corresponding to content distribution businesses. The second geographical factor corresponding to content distribution businesses is greater than the first geographical factor corresponding to real-time interactive businesses. Optionally, the value range of the geographical factors is set as follows, and users can set specific values ​​within the corresponding value range according to their actual needs: The range of the first geographic factor α corresponding to real-time interactive services is [0.2, 0.3]. The range of values ​​for the second geographic factor α corresponding to content distribution services is [0.4, 0.6]. The high reliability requirement requires the third geographic factor α to have a value range of [0.7, 0.8].

[0067] As one implementation of this application, the current performance metrics and historical performance scores of each candidate node are obtained, including: Obtain multiple performance test scores and corresponding test times for the candidate node within the first time period; the multiple performance test scores are obtained by conducting multiple tests on the service quality of the candidate node. Based on the multiple performance test scores and corresponding test times at the current moment, calculate the historical performance score of the candidate node within the first time period; and determine the performance test score obtained from the most recent test at the current moment as the performance index of the candidate node.

[0068] In one embodiment, the monitoring module of the scheduling system acquires performance data for each candidate node at first time intervals, including: the node's current performance metrics, historical performance test scores, and corresponding test times. The historical performance test scores are obtained by conducting multiple performance tests on the service node within the past first time interval, covering dimensions such as response time, packet loss rate, throughput, and availability. Each test score has a corresponding timestamp (i.e., test time). Based on these historical performance test scores, a historical performance score for the candidate node is calculated to reflect the stability of the node's service quality within the past first time interval. Furthermore, the performance test score obtained from the most recent performance test is used as the node's current performance metric.

[0069] In this embodiment, the historical performance score T is calculated using an exponential decay model, as detailed below: ; in, The score for the i-th performance test; For testing time; The current time; This is the attenuation coefficient.

[0070] As one implementation of this application, based on the comprehensive score of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, including: Based on the comprehensive scores of each candidate node, sort the candidate nodes from lowest to highest to generate a candidate list; Iterate through the candidate list and compare the current node load rate of each candidate node in the candidate list with the target load threshold. If the node load rate of the currently traversed candidate node is lower than the target load threshold, the candidate node is determined as the target node.

[0071] In one embodiment, based on the comprehensive scores of each candidate node, the candidate node with the best comprehensive score is selected as the target node corresponding to the target combination. Specifically, the comprehensive scores of each candidate node are sorted from low to high, and a candidate list is generated. The candidate node ranked first in the candidate list (i.e., with the lowest comprehensive score) is determined as the target node corresponding to the target combination.

[0072] In one embodiment, the current load rate of candidate nodes is also considered when selecting a target node. The candidate list is traversed, and the candidate node with the lowest overall score is selected first. If the load rate of this node is within an acceptable range, then this node is selected as the target node to provide CDN services. If the load rate of this node is too high or it fails, the remaining nodes are considered in order of their list ranking.

[0073] Specifically, the load rate of the currently traversed candidate nodes is compared with the target load threshold. If the load rate of a candidate node is less than the target load threshold, the candidate node is selected as the target node; if the load rate of a candidate node is greater than or equal to the target load threshold, the node is avoided as the target node. This is because when a node's load rate is close to the target load threshold, it indicates that the node is under high load. If this node is selected as the target node for the target combination to provide CDN services, it will further increase the load on the node, increasing the risk of node downtime or failure.

[0074] By comprehensively analyzing node load rate and overall score, we can ensure that the selected target nodes provide CDN services to the target area for as long as possible, thereby further enhancing the stability of CDN services in the target area.

[0075] As one embodiment of this application, after determining the candidate node as the target node, the method further includes: The real-time node load rate of the target node is continuously acquired at the first interval and compared with the target load threshold. If the real-time node load rate of the target node is greater than or equal to the target load threshold, the comprehensive scores of the other candidate nodes in the candidate list other than the target node are recalculated, and the candidate list is updated. Iterate through the updated candidate list and compare the current node load rate of each candidate node in the candidate list with the target load threshold; If the node load rate of the currently traversed candidate node is lower than the target load threshold, the candidate node is determined as the new target node.

[0076] Optionally, when the target node is determined to be the CDN service node of the target combination, if the target node experiences node failure, downtime, or excessive load during subsequent business operations, the method provided in the above embodiments can be used to select the current optimal candidate node from the candidate list corresponding to the target combination and replace the target node.

[0077] In one embodiment, during CDN service operation, the real-time node load rate of the target node is monitored at a first interval. When the load of the target node reaches the target load threshold (e.g., 90%), the current comprehensive score of each candidate node in the candidate list of the target combination is recalculated, and the node with the lowest comprehensive score is selected as the new target node to continue providing CDN services to users in the target area. In practical applications, the target load threshold can be set according to actual business needs, and this application does not impose any restrictions on it.

[0078] The following is a practical example to illustrate the above solution. The target area is "China Telecom Sichuan", and the corresponding resolution group is "live_stream" used to provide live streaming services. If node resources within the target area's coverage area experience anomalies, the scheduling system needs to select a suitable node from the node resource pool corresponding to the target combination "live_stream-China Telecom Sichuan" to replace it, ensuring the service quality of the target area. The maintenance personnel's customized sequence configuration information is shown in Table 1 below, where the priority of each preferred area is arranged from high to low and separated by the symbol "|".

[0079] Table 1

[0080] As shown in Table 1, when selecting nodes for the target combination within the validity period (effective-expired) of the sequential configuration information, candidate nodes from each preferred region will be considered in the following order: China Telecom Sichuan, China Telecom Chongqing, China Telecom Shaanxi, China Telecom Yunnan, China Telecom Hubei, and China Telecom Shanghai. The geographical factor is set to 0.7, indicating that geographical distance plays a significant role in calculating the comprehensive score of each candidate node. The specific process for selecting the optimal target node for the target combination is as follows: (1) Based on the above-mentioned sequential configuration information, select all candidate nodes corresponding to the target combination from the node resource pools of the same region, neighboring provinces, and the node resource pools of the specified preferred regions. Obtain the current performance indicators and load rates of each candidate node. Also, obtain the historical performance test scores of the candidate nodes and calculate the historical performance scores, as shown in Table 2 below.

[0081] Table 2

[0082] (2) Calculate the geographical distance (arc distance) between each candidate node (host group) and the target area (Telecom Sichuan), and perform nonlinear normalization to obtain the corresponding distance influence value. The results are as follows: Chengdu-Chengdu (Node-SC-01): 0 km; Distance impact value: 0; Chengdu to Mianyang (Node-SC-02): 122.7 km; Distance influence value 491 (close-up magnification); Chengdu to Nanchong (Node-SC-03): 168.3 km; Distance influence value 546 (close-range magnification); Chengdu to Chongqing (Node-CQ-01): 270.9 km; Distance impact value 598 (medium distance); Chengdu-Xi'an (Node-SX-01): 658.7 km; Distance impact value 694 (long-distance compression); Chengdu to Kunming (Node-YN-01): 634.2 km; Distance impact value 687 (long-distance compression); Chengdu-Wuhan (Node-HB-01): 973.5 km; Distance impact value 752 (long-distance compression); Chengdu-Beijing (Node-BJ-01): 1513.7 km; Distance impact value 836 (long-distance compression); Chengdu-Shanghai (Node-SH-01): 1661.5 km; Distance impact value 862 (long-distance compression).

[0083] (3) Set the corresponding basic weights for each candidate node according to the user-defined preference order. Among them, nodes HG-SC-01, HG-SC-02, and HG-SC-03 are nodes in Sichuan Province that are ranked first in the preference order, so they are all assigned a basic weight of 0. The basic weights are assigned to each preference area in the following order: the basic weight of node HG-CQ-01 is 1; the basic weight of node HG-SX-01 is 2; the basic weight of node HG-YN-01 is 3; the basic weight of node HG-HB-01 is 4; and the basic weight of node HG-SH-01 is 5. Node HG-BJ-01 does not belong to the preference area, so the default basic weight is set to 99.

[0084] (4) The business type corresponding to the parsing group is live video streaming, which belongs to the real-time interactive business. The corresponding first geographical factor α is 0.3, the time factor β is 0.2, and the segmentation base S is 1000. Based on this, according to the performance indicators, historical performance scores, distance influence values, and basic weights obtained in the above steps, the comprehensive scores of each candidate node are calculated as follows: The overall score for Hg-SC-01 is 433; for Hg-SC-02 it is 1344; for Hg-SC-03 it is 1382; for Hg-CQ-01 it is 2419; for Hg-SX-01 it is 3486; for Hg-YN-01 it is 4481; for Hg-HB-01 it is 5526; for Hg-BJ-01 it is 99585; and for Hg-SH-01 it is 7604.

[0085] (5) Based on the comprehensive score of each candidate node, sort them in ascending order of score to generate a candidate list. The lower the comprehensive score, the higher the priority of the candidate node. Iterate through the candidate list and compare the current load rate of each candidate node with the target load threshold (90%). If the load rate of a node is lower than the target load threshold, it can be selected as the CDN service node for the target combination (i.e., the target node). Finally, the node Hg-SC-01 with a load rate of 78% is determined as the CDN service node for the "live_stream" parsing group in the Sichuan region of China Telecom.

[0086] If the Hg-SC-01 node experiences excessive load (reaching the target load threshold) or fails, the overall score of each candidate node in the candidate list will be recalculated, and a replacement node will be selected based on the score ranking.

[0087] Based on the same inventive concept, one embodiment of this application provides a resource scheduling device for a scheduling system deployed in a CDN network. Figure 4This is a schematic diagram of a resource scheduling device 100 according to an embodiment of this application. Figure 4 As shown, the device includes: Management module 101 is configured to filter all candidate nodes in the CDN network at the granularity of the combination of parsing groups and regions; the target combination consists of a target region and its corresponding parsing group; and the basic weight of each candidate node is determined based on the preferred order of the regions to which each candidate node belongs. The distance calculation module 102 is configured to calculate the geographical distance between the region to which each candidate node belongs and the target region based on the location information of the target combination and the location information of each candidate node. The monitoring module 103 is configured to obtain the current performance indicators and historical performance scores of each candidate node; the performance indicators represent the current service quality of the candidate node; and the historical performance scores represent the stability of the service quality of the candidate node within a first time period. The comprehensive analysis module 104 is configured to calculate the comprehensive score of each candidate node based on its geographical distance, basic weight, performance index, and historical performance score; the comprehensive score is used to characterize the priority of the candidate nodes. The scheduling module 105 is configured to determine the candidate node with the highest priority as the target node corresponding to the target combination based on the comprehensive score of each candidate node, and to schedule CDN service requests in the target area to the target node.

[0088] Based on the same inventive concept, one embodiment of this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the resource scheduling method described in any of the above embodiments of this application.

[0089] Based on the same inventive concept, one embodiment of this application provides a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the resource scheduling method as described in any of the above embodiments of this application.

[0090] Based on the same inventive concept, one embodiment of this application provides a server. The server includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When executed by the processor, the computer program implements the steps of the resource scheduling method described in any of the above embodiments of this application.

[0091] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here. The above descriptions are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0092] For the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and components involved are not necessarily essential to this application.

[0093] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0095] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0096] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0097] Although preferred embodiments of the embodiments of this application have been described, those skilled in the art, once they understand the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, this application is to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of this application.

[0098] 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 terminal device 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 terminal device. 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 terminal device that includes said element.

[0099] The resource scheduling method, apparatus, and product 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 method 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 scheduling method, characterized in that, A scheduling system applied to CDN networks; The CDN network includes: multiple resolution groups and nodes deployed in different regions; the method includes: Using the combination of parsing groups and regions as the granularity, all candidate nodes in the CDN network are screened for the target combination; the target combination consists of a target region and its corresponding parsing group. The basic weight of each candidate node is determined based on the preferred order of the regions to which each candidate node belongs. Based on the location information of the target combination and the location information of each candidate node, calculate the geographical distance between the region to which each candidate node belongs and the target region; Obtain the current performance metrics and historical performance scores of each candidate node; the performance metrics represent the current service quality of the candidate node; the historical performance scores represent the stability of the service quality of the candidate node within a first time period. A comprehensive score is calculated for each candidate node based on its geographical distance, basic weight, performance indicators, and historical performance scores; the comprehensive score is used to characterize the priority of the candidate node. Based on the comprehensive scores of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, and CDN service requests within the target area are scheduled to the target node.

2. The resource scheduling method according to claim 1, characterized in that, All candidate nodes in the CDN network are selected based on the target combination, including: Determine the region to which the target area belongs; the region must contain at least two sub-regions; Based on the parsing group, the corresponding business type is determined. According to the business type, a first type of node capable of providing the service type is obtained from the node resource pool of the region; and, according to the business type, a second type of node capable of providing the service type is obtained from the node resource pool of the neighboring province of the target region. The first type of nodes and the second type of nodes are added to the candidate set and deduplicated. All nodes in the deduplicated candidate set are then determined as candidate nodes.

3. The resource scheduling method according to claim 2, characterized in that, The method further includes: Obtain the sequence configuration information corresponding to the target combination; the sequence configuration information is used to specify at least one preferred region, and the preferred region has a corresponding preferred order; The process of selecting all candidate nodes in the CDN network for the target combination further includes: according to the service type, obtaining a third type of node that can provide the service type from the node resource pool of each preferred region and adding it to the candidate set; and deduplicating the candidate set.

4. The resource scheduling method according to claim 3, characterized in that, The sequence configuration information also includes the effective time period; determining the basic weight of each candidate node specifically includes: If the current time is outside the effective period, set the same first basic weight for each candidate node; Given that the current time is within the effective period, obtain the arrangement order of each preferred region in the sequence configuration information; According to the order of arrangement, from first to last, a second basic weight is set for the third type of node corresponding to each preferred region, from smallest to largest; the second basic weight is less than the first basic weight.

5. The resource scheduling method according to claim 1, characterized in that, Based on the location information of the target combination and the location information of each candidate node, the geographical distance between the region to which each candidate node belongs and the target region is calculated, including: Obtain the latitude and longitude information of the target combination and the latitude and longitude information of each candidate node; Calculate the longitude difference and latitude difference between the latitude and longitude information of each candidate node and the latitude and longitude information of the target combination; Based on the Earth's radius and the longitude and latitude differences corresponding to each candidate node, the arc distance between the candidate node and the corresponding position of the target area is calculated.

6. The resource scheduling method according to claim 5, characterized in that, After calculating the arc distance between the candidate node and the corresponding position in the target region, the method further includes: The arc distance corresponding to each candidate node is compared with a first distance threshold and a second distance threshold to determine the value range of the arc distance; the second distance threshold is greater than the first distance threshold. Based on the value range of each arc distance, nonlinear normalization is performed on each arc distance to obtain the distance influence value corresponding to the candidate node; wherein, when the arc distance is within a first value range, the arc distance is expanded, and when the arc distance is within a second value range, the arc distance is reduced; the first value range is the value range less than the first distance threshold; the second value range is the value range greater than or equal to the second distance threshold.

7. The resource scheduling method according to claim 6, characterized in that, Based on the geographical distance, basic weight, performance indicators, and historical performance scores of each candidate node, a comprehensive score is calculated for each candidate node, including: Calculate the order parameter based on the basic weight and segmentation cardinality corresponding to the candidate node; Based on the geographical factors and the distance influence values ​​corresponding to the candidate nodes, calculate the geographical parameter items; Based on the geographical factors and the current performance indicators of the candidate nodes, calculate the performance parameter items; Based on the historical performance scores and time factors of the candidate nodes, calculate the historical performance parameters. The comprehensive score of the candidate node is calculated based on the geographic parameter item, the performance parameter item, and the historical performance parameter item.

8. The resource scheduling method according to claim 7, characterized in that, Also includes: Determine the service type of the CDN service that the target combination needs to provide; the service type is any of the following: real-time interactive service, content distribution service, or service with high reliability requirements; When the business type is a real-time interactive business, set the corresponding first geographical factor; When the business type is a content distribution business, a corresponding second geographical factor is set; The second geographical factor is greater than the first geographical factor; When the business type is a high-reliability requirement business, a corresponding third geographical factor is set; The third geographical factor is greater than the second geographical factor.

9. The resource scheduling method according to claim 1, characterized in that, Obtain the current performance metrics and historical performance scores of each candidate node, including: Obtain multiple performance test scores and corresponding test times for the candidate node within the first time period; the multiple performance test scores are obtained by conducting multiple tests on the service quality of the candidate node. Based on the multiple performance test scores and corresponding test times at the current moment, calculate the historical performance score of the candidate node within the first time period; and determine the performance test score obtained from the most recent test at the current moment as the performance index of the candidate node.

10. The resource scheduling method according to claim 1, characterized in that, Based on the comprehensive scores of each candidate node, the candidate node with the highest priority is determined as the target node corresponding to the target combination, including: Based on the comprehensive scores of each candidate node, sort the candidate nodes from lowest to highest to generate a candidate list; Iterate through the candidate list and compare the current node load rate of each candidate node in the candidate list with the target load threshold. If the node load rate of the currently traversed candidate node is lower than the target load threshold, the candidate node is determined as the target node.

11. The resource scheduling method according to claim 10, characterized in that, After determining the candidate node as the target node, the method further includes: The real-time node load rate of the target node is continuously acquired at the first interval and compared with the target load threshold. If the real-time node load rate of the target node is greater than or equal to the target load threshold, the comprehensive scores of the other candidate nodes in the candidate list other than the target node are recalculated, and the candidate list is updated. Iterate through the updated candidate list and compare the current node load rate of each candidate node in the candidate list with the target load threshold; If the node load rate of the currently traversed candidate node is lower than the target load threshold, the candidate node is determined as the new target node.

12. A resource scheduling device, characterized in that, A scheduling system deployed in a CDN network, used to perform the method as described in any one of claims 1-11, comprising: The management module is configured to filter all candidate nodes in the CDN network at the granularity of the combination of parsing groups and regions; the target combination consists of a target region and its corresponding parsing group; and the basic weight of each candidate node is determined based on the preferred order of the regions to which each candidate node belongs. The distance calculation module is configured to calculate the geographical distance between the region to which each candidate node belongs and the target region based on the location information of the target combination and the location information of each candidate node; The monitoring module is configured to obtain the current performance indicators and historical performance scores of each candidate node; the performance indicators represent the current service quality of the candidate node; and the historical performance scores represent the stability of the service quality of the candidate node within a first time period. The comprehensive analysis module is configured to calculate the comprehensive score of each candidate node based on its geographical distance, basic weight, performance indicators, and historical performance scores; the comprehensive score is used to characterize the priority of the candidate nodes. The scheduling module is configured to determine the candidate node with the highest priority as the target node corresponding to the target combination based on the comprehensive score of each candidate node, and to schedule CDN service requests in the target area to the target node.

13. 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 method according to any one of claims 1-11.

14. A server, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method as described in any one of claims 1-11.

15. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1-11.

Images