Content distribution network scheduling model simulation verification method and device, and electronic equipment

Abstract

This application relates to the field of simulation verification technology, providing a method, apparatus, and electronic device for simulating and verifying a content delivery network (CDN) scheduling model. The method includes: constructing a digital twin CDN based on a real production environment of the CDN scheduling system; ensuring that the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment; performing a first simulation verification on a CDN scheduling model under a preset scenario based on the data from multiple twin nodes; if the result of the first simulation verification meets expectations, performing a second simulation verification on the CDN scheduling model under the preset scenario based on the data from multiple real nodes; and if the result of the second simulation verification meets expectations, applying the CDN scheduling model under the preset scenario to a real production environment. This application can fully verify the functionality of the CDN scheduling model, cover any required scenario, and will not cause any negative impact on the existing network.

Description

Technical Field

[0001] This application relates to the field of simulation verification technology, specifically to a simulation verification method, apparatus, and electronic device for a content delivery network scheduling model. Background Technology

[0002] A Content Delivery Network (CDN) is a distributed network architecture that can quickly and efficiently deliver requested content to users, while reducing the load on content provider servers and lowering network congestion.

[0003] The CDN scheduling system is the brain of the CDN. Based on factors such as the user's geographical location, network conditions, and server load, it uses intelligent decision-making and automated operations to route user requests to the most suitable edge nodes to reduce latency and improve response speed.

[0004] In recent years, many intelligent CDN scheduling models have been proposed for CDN scheduling systems. However, due to the complexity of actual networks, the diversity of services of massive service nodes and their real-time changes in status, it is impossible to fully construct a CDN that conforms to the actual situation in traditional test environments. As a result, the test environments on which each CDN scheduling model depends cannot match the actual situation of the live network. The model functions have problems such as insufficient verification and insufficient scenario coverage, and may even have a negative impact on the live network. Summary of the Invention

[0005] This application provides a content delivery network scheduling model simulation verification method, device, and electronic device to solve the technical problems that the test environment on which the CDN scheduling model depends cannot match the actual situation of the current network, the model function has insufficient verification, insufficient scenario coverage, etc., and may even have a negative impact on the current network.

[0006] In a first aspect, embodiments of this application provide a simulation verification method for a content delivery network scheduling model, including: A digital twin CDN is constructed based on the real production environment of the Content Delivery Network (CDN) scheduling system; the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on data from multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed; the CDN scheduling model under the preset scenario is obtained by performing parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in a test environment; If the results of the first simulation verification meet expectations, a second simulation verification is performed on the CDN scheduling model under the preset scenario based on data from multiple real nodes. If the results of the second simulation verification meet expectations, the CDN scheduling model under the preset scenario will be applied to the real production environment.

[0007] In one embodiment, after performing a first simulation verification of the CDN scheduling model under a preset scenario, the process includes: If the result of the first simulation verification does not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and the data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario until the result of the first simulation verification meets expectations.

[0008] In one embodiment, after performing a first simulation verification of the CDN scheduling model under a preset scenario, the process includes: If the result of the first simulation verification does not meet expectations, the parameters of the initial CDN scheduling model are adjusted, and the data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario, until the result of the first simulation verification meets expectations.

[0009] In one embodiment, after performing a second simulation verification on the CDN scheduling model under the preset scenario, the process includes: If the result of the second simulation verification does not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and the first simulation verification of the CDN scheduling model under the preset scenario is performed based on data from multiple twin nodes until the result of the second simulation verification meets expectations.

[0010] In one embodiment, constructing a digital twin CDN based on the real production environment of the Content Delivery Network (CDN) scheduling system includes: Collect data from real nodes in the actual production environment; The data from the real nodes are aggregated according to time and node to obtain aggregated data; The aggregated data is cleaned, corrected, and optimized to generate data for twin nodes; A digital twin CDN is constructed based on the data from the twin nodes.

[0011] In one embodiment, it also includes: If the difference between the simulation verification result and the expected result is less than the difference threshold, it is determined that the simulation verification result has met the expectation. The simulation verification results include the results of the first simulation verification and the results of the second simulation verification.

[0012] Secondly, embodiments of this application provide a content delivery network scheduling model simulation verification device, comprising: The digital twin CDN construction module is used to: construct a digital twin CDN based on the real production environment of the Content Delivery Network (CDN) scheduling system; the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment; The first simulation verification module is used to: perform a first simulation verification on the CDN scheduling model under a preset scenario based on data from multiple twin nodes; the CDN scheduling model under the preset scenario is obtained after parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in a test environment; The second simulation verification module is used to: perform a second simulation verification on the CDN scheduling model under the preset scenario based on data from multiple real nodes, provided that the result of the first simulation verification meets expectations. The CDN scheduling model application module is used to apply the CDN scheduling model under the preset scenario to the real production environment when the results of the second simulation verification meet expectations.

[0013] 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 content delivery network scheduling model simulation verification method described in the first aspect.

[0014] Fourthly, embodiments of this application provide a computer program product, including a computer program, which, when executed by a processor, implements the steps of the content delivery network scheduling model simulation verification method described in the first aspect.

[0015] Fifthly, embodiments of this application provide a non-transitory computer-readable storage medium, including a computer program, which, when executed by a processor, implements the steps of the content delivery network scheduling model simulation verification method described in the first aspect.

[0016] The content delivery network (CDN) scheduling model simulation verification method, apparatus, and electronic equipment provided in this application construct a digital twin CDN based on the real production environment of the CDN scheduling system. The data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on the data of multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed. The CDN scheduling model under the preset scenario is obtained after parameter simulation and model simulation of the initial CDN scheduling model in the test environment. If the result of the first simulation verification meets expectations, a second simulation verification of the CDN scheduling model under the preset scenario is performed based on the data of multiple real nodes. If the result of the second simulation verification meets expectations, the CDN scheduling model under the preset scenario is applied to the real production environment. In this application, a digital twin CDN is added between the real production environment and the test environment. This digital twin CDN is a high-fidelity modeling and real-time mapping of the actual physical CDN in the digital space, which can reflect the status of the actual physical CDN in real time. The CDN scheduling model provided by the test environment is simulated and verified through the digital twin CDN. Then, the CDN scheduling model is simulated and verified again by combining the real-time data of the real production environment. Since the verification environment is separated from the live network and highly matched with the actual situation of the live network, the verification process needs to be verified by both twin data and real data and is strongly correlated with the preset scenario. This can fully verify the function of the CDN scheduling model, cover any required scenario, and will not have any negative impact on the live network. Attached Figure Description

[0017] 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.

[0018] Figure 1 This is one of the flowcharts illustrating the simulation verification method for the content delivery network scheduling model provided in this application embodiment; Figure 2 This is the second flowchart illustrating the simulation verification method for the content delivery network scheduling model provided in this application embodiment; Figure 3 This is a simulation verification system architecture diagram of the content delivery network scheduling model provided in the embodiments of this application; Figure 4 This is a schematic diagram of the verification and parameter tuning process in the digital twin platform provided in this application embodiment; Figure 5 This is a schematic diagram of the structure of the content delivery network scheduling model simulation verification device provided in the embodiments of this application; Figure 6 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] 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.

[0020] Figure 1 This is one of the flowcharts illustrating the simulation verification method for the content delivery network scheduling model provided in this application. (Refer to...) Figure 1 This application provides a simulation verification method for a content delivery network scheduling model, which may include: Step 101: Construct a digital twin CDN based on the real production environment of the Content Delivery Network (CDN) scheduling system; The data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment; Step 102: Based on data from multiple twin nodes, perform the first simulation verification of the CDN scheduling model under the preset scenario; The CDN scheduling model under the preset scenario was obtained by simulating the parameters and model of the initial CDN scheduling model under the preset scenario in the test environment. Step 103: If the results of the first simulation verification meet expectations, a second simulation verification is performed on the CDN scheduling model under the preset scenario based on data from multiple real nodes. Step 104: If the results of the second simulation verification meet expectations, apply the CDN scheduling model under the preset scenario to the real production environment.

[0021] In step 101, a digital twin is a virtual model, also known as a virtual representation or copy of a physical system or object in the physical world. The core of digital twin technology lies in achieving real-time bidirectional connection and synchronization between the physical and digital worlds, possessing characteristics such as real-time data synchronization, simulation and predictive analysis, and design testing. Initially applied to the health maintenance and assurance of aerospace vehicles, digital twin technology, with its continuous improvement and development, is now widely used in engineering construction and intelligent manufacturing. Its aim is to predict the response of equipment or network status in the actual environment and assess the next state through the operational analysis and full information interaction of a digitally mapped model.

[0022] In this step, a digital twin CDN is constructed based on the real production environment of the Content Delivery Network (CDN) scheduling system. That is, a digital twin CDN is constructed based on the actual physical CDN. The data of each twin node in the digital twin CDN is completely consistent with the data of each real node in the actual physical CDN, and can change in real time with the data of each real node.

[0023] Furthermore, the digital twin CDN can be built within a pre-defined digital twin platform, allowing the platform's built-in functions to construct the digital twin CDN and subsequently perform simulation verification of the model.

[0024] In step 102, since the CDN scheduling system needs to solve the problems of resource allocation and quality optimization in various business scenarios based on the real-time status of nodes and services in the actual physical CDN, such as traffic bursts, node failures, and service quality degradation, in the test environment, the preset business scenarios and node status and node failure scenarios under the preset business scenarios can be simulated with parameters, and the initial CDN scheduling model can be simulated to obtain a CDN scheduling model suitable for the preset business scenarios.

[0025] The initial CDN scheduling model can be any intelligent scheduling model, without limitation. In this embodiment, the initial CDN scheduling model can be a model that scores nodes based on their business data according to preset rules, and then schedules user requests based on the scores of each node. The preset rules can mainly include node indicator configuration items and their influence coefficients. The indicator configuration items can mainly include bandwidth ratio, first packet latency, download speed, abnormal code ratio, etc. The final score of a node can be determined based on the weighted sum of the scores of each indicator configuration item and the influence coefficients of each indicator configuration item, that is, it is determined according to the following formula: ; in, It is the node's final score. , , and These are the node's most recent scores in terms of bandwidth share, first packet latency, download speed, and error code share. , , and This represents the influence coefficient of the corresponding indicator configuration item.

[0026] In the test environment, parameter simulation and model simulation are performed on the initial CDN scheduling model for different preset business scenarios. The influence coefficients of the indicator configuration items and the indicator thresholds of the preset scores under each indicator configuration item can be obtained for different preset business scenarios, thus forming CDN scheduling models under different preset business scenarios.

[0027] Assuming the test environment targets scenarios 1, 2, and 3, parameter simulation and model simulation were performed on the initial CDN scheduling model. The resulting CDN scheduling models for the three scenarios are shown in the table below: Table 1 CDN scheduling models under different preset business scenarios

[0028] Among them, Scenario 1 is an overload scheduling scenario that mainly considers the high load of node bandwidth, and only includes the bandwidth ratio as the configuration item, with an impact coefficient of 1; Scenario 2 is a poor-quality scheduling scenario that comprehensively considers bandwidth ratio, first packet latency, download speed, and abnormal code ratio, and includes four configuration items: bandwidth ratio, first packet latency, download speed, and abnormal code ratio, with impact coefficients of 0.3, 0.2, 0.2, and 0.3, respectively; Scenario 3 is a poor-quality scheduling scenario that mainly considers abnormal status codes, and includes four configuration items: traffic ratio, first packet latency, download speed, and abnormal code ratio, with impact coefficients of 0.3, 0.1, 0.1, and 0.5, respectively.

[0029] In the above three scenarios: For the metric "Bandwidth Ratio," the score is as follows: 10 for 0% or higher and less than 10%; 9 for 10% or higher and less than 20%; 8 for 20% or higher and less than 30%; 7 for 30% or higher and less than 35%; 6 for 35% or higher and less than 40%; 5 for 40% or higher and less than 45%; 4 for 45% or higher and less than 50%; 3 for 50% or higher and less than 60%; 2 for 60% or higher and less than 65%; and 1 for 65% or higher. For the metric "First Packet Latency," the score is as follows: 10 for 30ms or less; 9 for 30ms or more but less than or equal to 50ms; 8 for 50ms or more but less than or equal to 100ms; 7 for 100ms or more but less than or equal to 150ms; 6 for 150ms or more but less than or equal to 180ms; 5 for 180ms or more but less than or equal to 200ms; 4 for 200ms or more but less than or equal to 220ms; 3 for 220ms or more but less than or equal to 250ms; 2 for 250ms or more but less than or equal to 300ms; 1 for 300ms or more but less than or equal to 330ms; and 0 for 330ms or more but less than or equal to 350ms. For the "Download Speed" metric, the score is as follows: 10 for 4000 KB / s or higher; 9 for 3500 KB / s or higher but less than 4000 KB / s; 8 for 3300 KB / s or higher but less than 3500 KB / s; 7 for 3000 KB / s or higher but less than 3300 KB / s; 6 for 2900 KB / s or higher but less than 3000 KB / s; 5 for 2700 KB / s or higher but less than 2900 KB / s; 4 for 2500 KB / s or higher but less than 2700 KB / s; 3 for 2400 KB / s or higher but less than 2500 KB / s; and 3 for 2200 KB / s or higher. A score of 2 is given when the speed is between 2400 KB / s and 2000 KB / s or higher and less than 2200 KB / s; a score of 1 is given when the speed is between 1800 KB / s or higher and less than 2000 KB / s. For the indicator configuration item "Percentage of Abnormal Status Codes", the score is 10 when it is less than or equal to 0.01%; 9 when it is greater than 0.01% and less than or equal to 0.05%; 8 when it is greater than 0.05% and less than or equal to 0.1%; 7 when it is greater than 0.1% and less than or equal to 0.3%; 6 when it is greater than 0.3% and less than or equal to 0.5%; 5 when it is greater than 0.5% and less than or equal to 1%; 4 when it is greater than 1% and less than or equal to 3%; 3 when it is greater than 3% and less than or equal to 5%; 2 when it is greater than 5% and less than or equal to 10%; 1 when it is greater than 10% and less than or equal to 30%; and 0 when it is greater than 30% and less than or equal to 50%.

[0030] The scoring of the "Traffic Share" metric configuration item is similar to that of "Bandwidth Share" within each threshold range, so it will not be repeated here.

[0031] In the digital twin platform, based on data from multiple twin nodes in the digital twin CDN, the first simulation verification of the CDN scheduling model under the above three scenarios can be performed. The results of the first simulation verification are shown in the table below: Table 2. Simulation Verification Results for Scenario 1

[0032] The expected score and actual score in Table 2 are both bandwidth percentage scores. The actual score is calculated based on the CDN scheduling model in Scenario 1 in Table 1. The bandwidth percentage can be calculated based on the node capacity and the bandwidth used. Appropriate rounding has been performed in each calculation process.

[0033] Table 3. Simulation Verification Results for Scenario 2

[0034] The expected score and actual score in Table 3 are both comprehensive scores of each sub-item. The actual score is calculated based on the CDN scheduling model under scenario 2 in Table 1. The bandwidth ratio can be calculated based on the node capacity and the bandwidth used. All calculations have been rounded appropriately.

[0035] It should be noted that if at least one sub-item scores zero, the actual score is zero regardless of the scores of the other sub-items.

[0036] Table 4. Simulation Verification Results for Scenario 3

[0037] The expected score and actual score in Table 4 are both comprehensive scores of each sub-item. The actual score is calculated based on the CDN scheduling model under scenario 3 in Table 1. The traffic percentage can be calculated based on the node capacity and bandwidth used. Appropriate rounding has been performed in each calculation process.

[0038] It should be noted that if at least one sub-item scores zero, the actual score is zero regardless of the scores of the other sub-items.

[0039] In step 103, the first simulation verification result meets expectations. This can mean that the first simulation verification result of the CDN scheduling model under at least one preset scenario meets expectations. The standard for meeting expectations can be that the expected score and the actual score under the preset scenario are completely consistent, or that the difference between the expected score and the actual score is less than a difference threshold. This is not limited here. In this embodiment, the standard for meeting expectations can be set as the difference between the expected score and the actual score under the preset scenario being less than a difference threshold. For example, the absolute value of the difference between the expected score and the actual score under the preset scenario is less than an absolute value threshold.

[0040] As shown in Tables 2 to 4 in step 102, the expected scores and actual scores are completely consistent in the three preset scenarios. Therefore, the first simulation verification results of the CDN scheduling model in these three preset scenarios have all met expectations. Further, based on the data of multiple real nodes, the second simulation verification of the CDN scheduling model in these three preset scenarios can be carried out.

[0041] It should be noted that, since the data of nodes in the real production environment changes constantly, the data of real nodes may have changed since the first simulation verification of the CDN scheduling model under the preset scenario was achieved as expected. Therefore, it is necessary to collect the latest real node data for the second simulation verification.

[0042] In step 104, similar to the first simulation verification result, the second simulation verification result meets expectations. This can also mean that the second simulation verification result of at least one CDN scheduling model under a preset scenario meets expectations. The standard for meeting expectations can be that the expected score and the actual score under the preset scenario are completely consistent, or that the difference between the expected score and the actual score is less than a difference threshold; this is not limited here. In this embodiment, the standard for meeting expectations can be set as the difference between the expected score and the actual score under the preset scenario being less than a difference threshold. For example, the absolute value of the difference between the expected score and the actual score under the preset scenario is less than an absolute value threshold.

[0043] Furthermore, if the second simulation verification result of the CDN scheduling model under at least one preset scenario meets expectations, the CDN scheduling model under the preset scenario with the expected second simulation verification result is applied to the real production environment. The CDN scheduling model has been verified by both twin node data and real node data, and its scenario adaptability and algorithm correctness have been fully and comprehensively verified. It can perform optimal resource allocation and quality optimization based on the adapted business scenario in the real production environment.

[0044] The content delivery network scheduling model simulation verification method provided in this embodiment constructs a digital twin CDN based on the real production environment of the content delivery network (CDN) scheduling system. The data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on the data of multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed. The CDN scheduling model under the preset scenario is obtained after parameter simulation and model simulation of the initial CDN scheduling model in the test environment. If the result of the first simulation verification meets expectations, a second simulation verification of the CDN scheduling model under the preset scenario is performed based on the data of multiple real nodes. If the result of the second simulation verification meets expectations, the CDN scheduling model under the preset scenario is applied to the real production environment. In this embodiment, a digital twin CDN is added between the real production environment and the test environment. This digital twin CDN is a high-fidelity modeling and real-time mapping of the actual physical CDN in the digital space, reflecting the real physical CDN's status in real time. The CDN scheduling model provided by the test environment is simulated and verified through the digital twin CDN, and then the CDN scheduling model is simulated and verified again by combining the real-time data of the real production environment. Since the verification environment is separated from the live network and highly matched with the actual situation of the live network, the verification process requires dual verification by twin data and real data and is strongly correlated with the preset scenario. This can fully verify the function of the CDN scheduling model, cover any required scenario, and will not have any negative impact on the live network.

[0045] In one embodiment, step 102 may be followed by: If the results of the first simulation verification do not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario until the results of the first simulation verification meet expectations.

[0046] If the result of the first simulation verification does not meet expectations, after adjusting the parameters of the CDN scheduling model under the preset scenario in the digital twin platform, return to step 102, and re-perform the first simulation verification of the CDN scheduling model under the preset scenario based on the data of multiple twin nodes, until the result of the first simulation verification meets expectations.

[0047] Taking the CDN scheduling model in Table 1 above as an example, after adjusting the influence coefficient of each indicator configuration item under each preset scenario and / or the indicator threshold of each preset score under each indicator configuration item, return to step 102, and re-perform the first simulation verification of the CDN scheduling model under the preset scenario based on the data of multiple twin nodes until the result of the first simulation verification reaches the expectation.

[0048] In this embodiment, if the results of the first simulation verification do not meet expectations, the parameters of the CDN scheduling model under the preset scenario are continuously adjusted on the digital twin platform to improve the scenario adaptability and algorithm correctness of the model until the results of the first simulation verification meet expectations, thereby verifying the scenario adaptability and algorithm correctness of the model on the twin node data.

[0049] In one embodiment, step 102 may be followed by: If the results of the first simulation verification do not meet expectations, the parameters of the initial CDN scheduling model are adjusted, and data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario, until the results of the first simulation verification meet expectations.

[0050] If the results of the first simulation verification do not meet expectations, after adjusting the parameters of the initial CDN scheduling model in the test environment, return to step 102 and re-perform the first simulation verification of the CDN scheduling model under the preset scenario based on the data of multiple twin nodes until the results of the first simulation verification meet expectations.

[0051] After adjusting the influence coefficients of each indicator configuration item in the initial CDN scheduling model and / or the indicator thresholds of each preset score under each indicator configuration item, return to step 102 and re-perform the first simulation verification of the CDN scheduling model under the preset scenario based on the data of multiple twin nodes until the result of the first simulation verification reaches the expectation; at this time, the CDN scheduling model under the preset scenario is obtained after performing parameter simulation and model simulation of the initial CDN scheduling model with adjusted parameters in the test environment.

[0052] In this embodiment, if the results of the first simulation verification do not meet expectations, a new CDN scheduling model under a preset scenario is generated by continuously adjusting the parameters of the initial CDN scheduling model in the test environment, and then the first simulation verification is performed on it until the results of the first simulation verification meet expectations, thereby verifying the scenario adaptability and algorithm correctness of the model on twin node data.

[0053] In one embodiment, step 103 may be followed by: If the results of the second simulation verification do not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario until the results of the second simulation verification meet expectations.

[0054] If the result of the second simulation verification does not meet expectations, after adjusting the parameters of the CDN scheduling model under the preset scenario in the digital twin platform, return to step 102, and re-perform the first simulation verification of the CDN scheduling model under the preset scenario based on the data of multiple twin nodes, until the result of the second simulation verification meets expectations.

[0055] Taking the CDN scheduling model in Table 1 above as an example, after adjusting the influence coefficient of each indicator configuration item under each preset scenario and / or the indicator threshold of each preset score under each indicator configuration item, return to step 102, and re-perform the first simulation verification of the CDN scheduling model under the preset scenario based on the data of multiple twin nodes, until the result of the second simulation verification reaches the expectation.

[0056] In this embodiment, when the results of the second simulation verification do not meet expectations, the parameters of the CDN scheduling model under the preset scenario are continuously adjusted on the digital twin platform to improve the model's scenario adaptability and algorithm correctness until the results of the second simulation verification meet expectations. This allows the model's scenario adaptability and algorithm correctness to be dually verified on both twin node data and real node data.

[0057] Figure 2 This is the second flowchart illustrating the simulation verification method for the content delivery network scheduling model provided in this application. (Refer to...) Figure 2 In one embodiment, step 101 may include: Step 201: Collect data from real nodes in a real production environment; Step 202: Aggregate the data of the real nodes according to time and node to obtain aggregated data; Step 203: Clean, correct, and optimize the aggregated data to generate data for twin nodes; Step 204: Construct a digital twin CDN based on the data from the twin nodes.

[0058] Step 201 involves collecting data from real nodes in the actual physical CDN. This data can include node service metrics, node status metrics, alarm and fault metrics, etc. Node service metrics can include node service bandwidth, download speed, first packet latency, status codes, etc., while node status metrics can include node bandwidth capacity, node CPU status, node memory status, node disk status, etc. The data from these real nodes includes both real-time and historical data.

[0059] In step 202, data within a preset time period is selected from the data collected in step 201, and these data are aggregated and integrated according to nodes to obtain aggregated data.

[0060] In step 203, cleaning the aggregated data may include handling missing values, handling outliers, and format standardization. By identifying and processing "dirty data" in the data, the integrity and accuracy of the data are ensured. Correcting the cleaned data may include smoothing noise, logical correction, and data matching. By fixing errors or inconsistencies in the data, it is made to conform to the actual situation. Optimizing the corrected data may include feature engineering, data dimensionality reduction, and data sampling. By improving the data quality, it is made more suitable for subsequent modeling or analysis.

[0061] The data processed as described above can be stored as twin node data in the digital twin platform.

[0062] In step 204, a digital twin CDN is constructed based on high-quality twin node data. This digital twin CDN is a heterogeneous CDN, composed of nodes from different business scenarios, which can achieve accurate mapping to the actual physical CDN.

[0063] This embodiment collects data from real nodes in a real generation environment, aggregates and preprocesses it, and then constructs a digital twin CDN. It can build a digital twin CDN that accurately maps the actual physical CDN based on high-quality real node data.

[0064] Reference Figure 3 In one embodiment, based on a content delivery network scheduling model simulation verification system architecture consisting of a real production environment, a digital twin platform, and a testing environment, the content delivery network scheduling model simulation verification method of this application is briefly described as follows: In the test environment, parameters are simulated for preset business scenarios and node status and node failure scenarios under preset business scenarios, and the initial CDN scheduling model is simulated to obtain a CDN scheduling model suitable for preset business scenarios. The CDN scheduling model of the preset business scenarios is then fed back to the digital twin platform. Data from real nodes in a real production environment is collected, aggregated by time and node to obtain aggregated data, cleaned, corrected and optimized, and used to generate data for twin nodes, which is stored in a digital twin platform. Based on the data of the twin nodes, a digital twin CDN is built in the digital twin platform. The data of the real nodes may include node business indicators, node status indicators, alarm and fault indicators, etc. In a digital twin platform, scenario modeling and simulation, along with deep learning analysis, can be used to simulate and verify CDN scheduling models for predefined business scenarios. Specifically, for example... Figure 4 As shown: Based on data from multiple twin nodes in a digital twin CDN, a first simulation verification of the CDN scheduling model under a preset scenario is conducted. Determine whether the results of the first simulation verification meet expectations; If not, after adjusting the parameters of the CDN scheduling model under the preset scenario on the digital twin platform, the first simulation verification step of the CDN scheduling model under the preset scenario is returned based on the data of multiple twin nodes in the digital twin CDN, until the result of the first simulation verification reaches the expected result. If so, save the parameters of the CDN scheduling model under the preset scenario, and perform a second simulation verification of the CDN scheduling model under the preset scenario based on data from multiple real nodes in the real production environment. Determine whether the results of the second simulation verification meet expectations; If not, after adjusting the parameters of the CDN scheduling model under the preset scenario on the digital twin platform, the first simulation verification step of the CDN scheduling model under the preset scenario is returned based on the data of multiple twin nodes in the digital twin CDN, until the result of the second simulation verification reaches the expectation. If so, the parameters of the CDN scheduling model under the preset scenario will be fed back to the real production environment, so as to realize the application of the CDN scheduling model under the preset scenario in the real production environment.

[0065] This embodiment is based on a simulation verification system architecture for a content delivery network scheduling model. It can not only assist in the simulation verification of the CDN scheduling model in various preset scenarios, but also continuously adjust the parameters of the CDN scheduling model in the dual verification process to obtain the CDN scheduling model with optimal function. This solves the problem of not being able to accurately configure the threshold values ​​of various model indicators in complex production environments.

[0066] The following describes the content delivery network scheduling model simulation verification device provided in the embodiments of this application. The content delivery network scheduling model simulation verification device described below can be referred to in correspondence with the content delivery network scheduling model simulation verification method described above.

[0067] Figure 5 This is a schematic diagram of the structure of the content delivery network scheduling model simulation verification device provided in this application embodiment. (Refer to...) Figure 5 This application provides a simulation and verification device for a content delivery network scheduling model, which may include: The digital twin CDN construction module 501 is used to: construct a digital twin CDN based on the real production environment of the content delivery network (CDN) scheduling system; the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment; The first simulation verification module 502 is used to: perform a first simulation verification on the CDN scheduling model under a preset scenario based on the data of multiple twin nodes; the CDN scheduling model under the preset scenario is obtained after parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in the test environment; The second simulation verification module 503 is used to: perform a second simulation verification on the CDN scheduling model under the preset scenario based on the data of multiple real nodes, provided that the result of the first simulation verification meets expectations. CDN scheduling model application module 504 is used to: apply the CDN scheduling model under the preset scenario to the real production environment when the result of the second simulation verification meets expectations.

[0068] The content delivery network scheduling model simulation verification device provided in this embodiment constructs a digital twin CDN based on the real production environment of the content delivery network (CDN) scheduling system. The data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on the data of multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed. The CDN scheduling model under the preset scenario is obtained after parameter simulation and model simulation of the initial CDN scheduling model in the test environment. If the result of the first simulation verification meets expectations, a second simulation verification of the CDN scheduling model under the preset scenario is performed based on the data of multiple real nodes. If the result of the second simulation verification meets expectations, the CDN scheduling model under the preset scenario is applied to the real production environment. In this embodiment, a digital twin CDN is added between the real production environment and the test environment. This digital twin CDN is a high-fidelity modeling and real-time mapping of the actual physical CDN in the digital space, reflecting the real physical CDN's status in real time. The CDN scheduling model provided by the test environment is simulated and verified through the digital twin CDN, and then the CDN scheduling model is simulated and verified again by combining the real-time data of the real production environment. Since the verification environment is separated from the live network and highly matched with the actual situation of the live network, the verification process requires dual verification by twin data and real data and is strongly correlated with the preset scenario. This can fully verify the function of the CDN scheduling model, cover any required scenario, and will not have any negative impact on the live network.

[0069] In one embodiment, the first simulation verification module 502 is specifically used for: If the result of the first simulation verification does not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and the first simulation verification of the CDN scheduling model under the preset scenario is performed based on data from multiple twin nodes until the result of the first simulation verification meets expectations. In one embodiment, the first simulation verification module 502 is specifically used for: If the result of the first simulation verification does not meet expectations, the parameters of the initial CDN scheduling model are adjusted, and the data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario, until the result of the first simulation verification meets expectations.

[0070] In one embodiment, the second simulation verification module 503 is specifically used for: If the result of the second simulation verification does not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and the first simulation verification of the CDN scheduling model under the preset scenario is performed based on data from multiple twin nodes until the result of the second simulation verification meets expectations.

[0071] In one embodiment, the digital twin CDN construction module 501 is specifically used for: Collect data from real nodes in the actual production environment; The data from the real nodes are aggregated according to time and node to obtain aggregated data; The aggregated data is cleaned, corrected, and optimized to generate data for twin nodes; A digital twin CDN is constructed based on the data from the twin nodes.

[0072] In one embodiment, it also includes: If the difference between the simulation verification result and the expected result is less than the difference threshold, it is determined that the simulation verification result has met the expectation. The simulation verification results include the results of the first simulation verification and the results of the second simulation verification.

[0073] Figure 6 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 can call a computer program in the memory 630 to execute the steps of the content delivery network scheduling model simulation verification method, such as including: A digital twin CDN is constructed based on the real production environment of the Content Delivery Network (CDN) scheduling system; the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on data from multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed; the CDN scheduling model under the preset scenario is obtained by performing parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in a test environment; If the results of the first simulation verification meet expectations, a second simulation verification is performed on the CDN scheduling model under the preset scenario based on data from multiple real nodes. If the results of the second simulation verification meet expectations, the CDN scheduling model under the preset scenario will be applied to the real production environment.

[0074] Furthermore, the logical instructions in the aforementioned memory 630 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.

[0075] 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 content delivery network scheduling model simulation verification method provided in the above embodiments, such as including: A digital twin CDN is constructed based on the real production environment of the Content Delivery Network (CDN) scheduling system; the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on data from multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed; the CDN scheduling model under the preset scenario is obtained by performing parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in a test environment; If the results of the first simulation verification meet expectations, a second simulation verification is performed on the CDN scheduling model under the preset scenario based on data from multiple real nodes. If the results of the second simulation verification meet expectations, the CDN scheduling model under the preset scenario will be applied to the real production environment.

[0076] On the other hand, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program thereon. The computer program is used to cause a processor to execute the steps of the content delivery network scheduling model simulation verification method provided in the above embodiments, including, for example: A digital twin CDN is constructed based on the real production environment of the Content Delivery Network (CDN) scheduling system; the data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment. Based on data from multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed; the CDN scheduling model under the preset scenario is obtained by performing parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in a test environment; If the results of the first simulation verification meet expectations, a second simulation verification is performed on the CDN scheduling model under the preset scenario based on data from multiple real nodes. If the results of the second simulation verification meet expectations, the CDN scheduling model under the preset scenario will be applied to the real production environment.

[0077] The non-transitory computer-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)).

[0078] 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.

[0079] 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.

[0080] 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 simulation verification method for a content delivery network scheduling model, characterized in that, include: Based on the real production environment of the Content Delivery Network (CDN) scheduling system, construct a digital twin CDN; The data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment; Based on data from multiple twin nodes, a first simulation verification of the CDN scheduling model under a preset scenario is performed; The CDN scheduling model under the preset scenario is obtained by performing parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in the test environment; If the results of the first simulation verification meet expectations, a second simulation verification is performed on the CDN scheduling model under the preset scenario based on data from multiple real nodes. If the results of the second simulation verification meet expectations, the CDN scheduling model under the preset scenario will be applied to the real production environment.

2. The simulation verification method for the content delivery network scheduling model according to claim 1, characterized in that, After performing the first simulation verification of the CDN scheduling model under the preset scenario, the process includes: If the result of the first simulation verification does not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and the data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario until the result of the first simulation verification meets expectations.

3. The simulation verification method for the content delivery network scheduling model according to claim 1, characterized in that, After performing the first simulation verification of the CDN scheduling model under the preset scenario, the process includes: If the result of the first simulation verification does not meet expectations, the parameters of the initial CDN scheduling model are adjusted, and the data based on multiple twin nodes is returned to perform the first simulation verification of the CDN scheduling model under the preset scenario, until the result of the first simulation verification meets expectations.

4. The simulation verification method for the content delivery network scheduling model according to claim 1, characterized in that, After performing a second simulation verification on the CDN scheduling model under the preset scenario, the process includes: If the result of the second simulation verification does not meet expectations, the parameters of the CDN scheduling model under the preset scenario are adjusted, and the first simulation verification of the CDN scheduling model under the preset scenario is performed based on data from multiple twin nodes until the result of the second simulation verification meets expectations.

5. The simulation verification method for the content delivery network scheduling model according to claim 1, characterized in that, The construction of a digital twin CDN based on the real production environment of the Content Delivery Network (CDN) scheduling system includes: Collect data from real nodes in the actual production environment; The data from the real nodes are aggregated according to time and node to obtain aggregated data; The aggregated data is cleaned, corrected, and optimized to generate data for twin nodes; A digital twin CDN is constructed based on the data from the twin nodes.

6. The simulation verification method for the content delivery network scheduling model according to claim 1, characterized in that, Also includes: If the difference between the simulation verification result and the expected result is less than the difference threshold, it is determined that the simulation verification result has met the expectation. The simulation verification results include the results of the first simulation verification and the results of the second simulation verification.

7. A simulation and verification device for a content delivery network scheduling model, characterized in that, include: The digital twin CDN building module is used to: build a digital twin CDN based on the real production environment of the Content Delivery Network (CDN) scheduling system; The data of the twin nodes in the digital twin CDN is consistent with the data of the real nodes in the real production environment; The first simulation verification module is used to: perform a first simulation verification of the CDN scheduling model under a preset scenario based on data from multiple twin nodes; The CDN scheduling model under the preset scenario is obtained by performing parameter simulation and model simulation of the initial CDN scheduling model under the preset scenario in the test environment; The second simulation verification module is used to: perform a second simulation verification on the CDN scheduling model under the preset scenario based on data from multiple real nodes, provided that the result of the first simulation verification meets expectations. The CDN scheduling model application module is used to apply the CDN scheduling model under the preset scenario to the real production environment when the results of the second simulation verification meet expectations.

8. 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 content delivery network scheduling model simulation verification method according to any one of claims 1 to 6.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the content delivery network scheduling model simulation verification method according to any one of claims 1 to 6.

10. 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 content delivery network scheduling model simulation verification method according to any one of claims 1 to 6.

Images