CDN edge IP network quality scheduling method and device, equipment and medium

Abstract

The application provides a CDN edge IP network quality scheduling method, device and equipment and medium. The method collects the network quality score of a target edge IP, judges whether to perform a bandwidth traction operation on the target edge IP, avoids directly replacing the edge IP as a whole, and improves the availability of the edge IP. When it is necessary to perform a bandwidth traction operation on the target edge IP, corresponding bandwidth response proportions are allocated to the target edge IP and a replacement edge IP, and a random number generated by a random number generation method points to the first bandwidth response proportion or the second bandwidth response proportion, and then the corresponding edge IP response content acceleration request is called. Thus, the overloaded traffic is led to the idle edge IP in the form of traction bandwidth, the network service quality of the edge IP is improved, direct replacement of the edge IP is avoided, the network request response efficiency is improved, the bandwidth cost is saved, and the availability of the edge IP is improved.

Description

Technical Field

[0001] This application relates to the field of network scheduling technology, and in particular to a CDN edge IP network quality scheduling method, apparatus, device and medium. Background Technology

[0002] A CDN (Content Delivery Network) is a network architecture that provides users with fast, efficient, and secure internet content delivery services through technologies such as distributed deployment, content caching, and intelligent scheduling. The main functions of a CDN are to alleviate internet network congestion, reduce website response time, improve user experience, and ensure the security and reliability of internet content.

[0003] The CDN edge IP network quality scheduling method is a technique based on CDN technology that performs node load balancing scheduling for specific application scenarios to optimize network resource utilization and user experience. This technology has the following characteristics:

[0004] 1) Based on the geographical location information of IP addresses, the optimal CDN node selection and dynamic load balancing scheduling are achieved;

[0005] 2) Based on the load status of CDN nodes, timely load balancing scheduling is performed to improve the utilization efficiency of node resources and network stability;

[0006] 3) The refined load balancing scheduling strategy can meet the needs of different application scenarios and users, and improve the adaptability and scalability of CDN.

[0007] If a network quality issue arises with an edge IP, the current scheduling method will be to directly replace the edge IP. This scheduling method will bring several problems:

[0008] 1) Degraded availability: Sometimes the network quality of edge IPs is abnormal only due to sudden LVS scheduling congestion. During this period, diverting some bandwidth to healthy edge IPs can solve the problem. Replacing the entire IP will lead to a waste of resources.

[0009] 2) Increased costs: Some edge IPs carry a large amount of bandwidth. If the entire IP is replaced, the bandwidth cost of the edge node to which the IP belongs will increase.

[0010] Therefore, improving the availability of edge IP has become an urgent technical problem to be solved. Summary of the Invention

[0011] This application provides a CDN edge IP network quality scheduling method, apparatus, device, and storage medium, aiming to improve the availability of edge IP.

[0012] In a first aspect, this application provides a CDN edge IP network quality scheduling method, the method comprising:

[0013] When a client sends a request to accelerate content for a target edge IP, the system determines whether bandwidth pulling operation needs to be performed on the target edge IP based on the network quality score of the target edge IP.

[0014] When it is necessary to perform the bandwidth pulling operation on the target edge IP, obtain the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP;

[0015] Generate random numbers based on random number generation methods;

[0016] When the random number corresponds to the first bandwidth response ratio, the target edge IP is invoked to respond to the content acceleration request;

[0017] When the random number corresponds to the second bandwidth response ratio, the replacement edge IP is invoked to respond to the content acceleration request.

[0018] Secondly, this application also provides a CDN edge IP network quality scheduling device, the CDN edge IP network quality scheduling device comprising:

[0019] The bandwidth traction execution judgment module is used to determine whether a bandwidth traction operation needs to be performed on the target edge IP based on the network quality score of the target edge IP when a content acceleration request for the target edge IP is received from the client.

[0020] The data acquisition module is used to acquire, when it is necessary to perform the bandwidth pulling operation on the target edge IP, the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP;

[0021] The random number generation module is used to generate random numbers based on random number generation methods.

[0022] The target edge IP response module is used to invoke the target edge IP to respond to the content acceleration request when the random number corresponds to the first bandwidth response ratio;

[0023] The Replace Edge IP Response Module is used to invoke the Replace Edge IP Response to the content acceleration request when the random number corresponds to the second bandwidth response ratio.

[0024] Thirdly, this application also provides a computer device, the computer device including a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the CDN edge IP network quality scheduling method as described above.

[0025] Fourthly, this application also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of the CDN edge IP network quality scheduling method described above.

[0026] This application provides a CDN edge IP network quality scheduling method, apparatus, device, and storage medium. The method includes, upon receiving a content acceleration request for a target edge IP initiated by a client, determining whether a bandwidth redirection operation needs to be performed on the target edge IP based on the network quality score of the target edge IP; if the bandwidth redirection operation needs to be performed on the target edge IP, obtaining a replacement edge IP corresponding to the target edge IP, a first bandwidth response ratio corresponding to the target edge IP, and a second bandwidth response ratio corresponding to the replacement edge IP; generating a random number based on a random number generation method; if the random number corresponds to the first bandwidth response ratio, then calling the target edge IP to respond to the content acceleration request; if the random number corresponds to the second bandwidth response ratio, then calling the replacement edge IP to respond to the content acceleration request. Through this method, this application, by collecting the network quality score of the target edge IP and determining whether to perform a bandwidth redirection operation on the target edge IP, avoids directly replacing the entire edge IP and improves the availability of edge IPs. When bandwidth diversion is required for a target edge IP, corresponding bandwidth response ratios are allocated to the target edge IP and the replacement edge IP. Based on the first or second bandwidth response ratio pointed to by a random number generated by a random number generation method, the corresponding edge IP is invoked to respond and accelerate the request. In this way, by diverting overloaded traffic to idle edge IPs through bandwidth diversion, the network service quality of edge IPs can be improved, direct replacement of edge IPs can be avoided, network request response efficiency can be improved, bandwidth costs can be saved, and the availability of edge IPs can be improved. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments 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.

[0028] Figure 1A flowchart illustrating a first embodiment of a CDN edge IP network quality scheduling method provided in this application;

[0029] Figure 2 A flowchart illustrating the second embodiment of the CDN edge IP network quality scheduling method provided in this application;

[0030] Figure 3 A schematic diagram illustrating the detection results of edge IP link response time and packet loss rate data within one minute provided in this application;

[0031] Figure 4 A flowchart illustrating the third embodiment of the CDN edge IP network quality scheduling method provided in this application;

[0032] Figure 5 This is a schematic diagram illustrating the complete edge IP bandwidth switching result provided in this application;

[0033] Figure 6 This is a schematic diagram illustrating the edge IP bandwidth switching results provided in this application.

[0034] Figure 7 This is a schematic diagram of the structure of a first embodiment of a CDN edge IP network quality scheduling device provided in this application;

[0035] Figure 8 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0036] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0037] 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.

[0038] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0039] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0040] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of a CDN edge IP network quality scheduling method provided in this application.

[0041] like Figure 1 As shown, the CDN edge IP network quality scheduling method includes steps S101 to S105.

[0042] S101. Upon receiving a content acceleration request for the target edge IP initiated by the client, determine whether a bandwidth pulling operation needs to be performed on the target edge IP based on the network quality score of the target edge IP.

[0043] In one embodiment, when a client requests accelerated content at the edge IP-domain-region granularity, the response edge IP and response bandwidth ratio of the edge IP-domain-region are obtained, and it is determined whether the response is entirely from the target edge IP. If so, the target edge IP is responded to directly. If the bandwidth of the target edge IP cannot meet the requirements for fully responding to the content acceleration request initiated by the client, a bandwidth pulling operation needs to be performed on the target edge IP.

[0044] S102. When it is necessary to perform the bandwidth pulling operation on the target edge IP, obtain the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP.

[0045] In one embodiment, when the target edge IP cannot fully respond to the content acceleration request initiated by the client, it is necessary to use bandwidth diversion to divert some traffic to other edge IPs to share the traffic load.

[0046] In one embodiment, an edge IP whose bandwidth meets the request response requirements is found in the resource pool corresponding to the target edge IP and used as a replacement edge IP. By allocating the bandwidth response ratio of the target edge IP and the replacement edge IP, the request response task is assigned to the target edge IP and the replacement edge IP respectively. In this way, some of the request response tasks of the target edge IP are diverted to the replacement edge IP to complete, reducing the load on the target edge IP, avoiding direct replacement of the edge IP, improving network request response efficiency, and improving the availability of the edge IP.

[0047] S103. Generate random numbers based on the random number generation method;

[0048] In one embodiment, a random number can be generated using a random number generation algorithm. The random number is set according to the first bandwidth response ratio corresponding to the target edge IP and the second bandwidth response ratio corresponding to the replacement edge IP. For example, it can be specified that the sum of the ratios of the first bandwidth response ratio and the second bandwidth response ratio is 100%. Then, the generation range of the random number can be set to 100, and any integer between 0 and 100 can be randomly generated by the random number generation algorithm.

[0049] In one embodiment, the random numbers generated by the random number generation algorithm are pseudo-random numbers. Pseudo-random numbers all have a periodicity, which can be observed given a sufficiently large number of occurrences. That is, the random numbers generated by the random number generation method are distributed according to a first bandwidth response ratio and a second bandwidth response ratio. For example, if the first bandwidth response ratio is 40% and the second bandwidth response ratio is 60%, when the base number of generated random numbers is large enough, the proportion of occurrences pointing to the second response ratio will be approximately 60%, while the proportion pointing to the first bandwidth response ratio will be approximately 40%.

[0050] S104. When the random number corresponds to the first bandwidth response ratio, the target edge IP is invoked to respond to the content acceleration request.

[0051] S105. When the random number corresponds to the second bandwidth response ratio, the replacement edge IP is invoked to respond to the content acceleration request.

[0052] In one embodiment, a random number is generated using a random number generation algorithm. The algorithm then randomly determines which edge IP's bandwidth response ratio the random number falls on. If the random number falls on the first bandwidth response ratio corresponding to the target edge IP, the target edge IP is responded to; otherwise, the second bandwidth response ratio corresponding to the replacement edge IP is responded to. By randomly selecting an edge IP, when the number of user requests is sufficient, the bandwidth carried by the target edge IP can be redirected to the replacement edge IP according to the bandwidth redirection ratio.

[0053] For example, suppose there are 1000 customer requests for accelerated content at the edge IP-domain-region granularity within one minute. The first bandwidth response ratio for calling up the target edge IP is 20%, and the second bandwidth response ratio for replacing the edge IP is 80%. For each customer request, a random number between 1 and 100 is randomly generated. When the response is within the range of 1-20, the edge IP is called up; when the random number is between 21-100, the edge IP is replaced. In this way, approximately 200 requests will respond with the edge IP being called up, and approximately 800 requests will respond with the edge IP being replaced, thus achieving the goal of bandwidth redirection according to the specified ratio.

[0054] By using random number generation, bandwidth can be diverted to edge IPs according to user requests. When there are enough requests, bandwidth can be diverted proportionally.

[0055] This embodiment provides a CDN edge IP network quality scheduling method. This method collects the network quality score of the target edge IP to determine whether to perform bandwidth redirection on the target edge IP, avoiding direct replacement of the entire edge IP and improving edge IP availability. When bandwidth redirection is required, corresponding bandwidth response ratios are allocated to the target edge IP and the replacement edge IP. Based on a first or second bandwidth response ratio pointed to by a random number generated by a random number generation method, the corresponding edge IP is invoked to accelerate the request. Thus, by redirecting overloaded traffic to idle edge IPs through bandwidth redirection, the network service quality of the edge IP can be improved, direct replacement of edge IPs is avoided, network request response efficiency is improved, bandwidth costs are saved, and edge IP availability is enhanced.

[0056] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the CDN edge IP network quality scheduling method provided in this application.

[0057] In this embodiment, based on the above... Figure 1 In the illustrated embodiment, step S101 specifically includes:

[0058] S201. Based on preset detection rules, detect the link response quality score and packet loss rate quality score of the target edge IP.

[0059] In one embodiment, the detection rules may include detection methods, detection cycles, and detection durations. For example, the link response quality score of the target edge IP may be evaluated by detecting the link response time of the target edge IP, and the packet loss rate quality score of the target edge IP may be evaluated by detecting the packet loss rate of the target edge IP.

[0060] In one embodiment, in order to ensure detection efficiency and accuracy of detection results, the period of each detection and the total detection time / number of detections can be set, and the detection results corresponding to multiple periods can be obtained for comprehensive evaluation.

[0061] In one embodiment, response time detection and packet loss rate detection can be performed simultaneously.

[0062] Furthermore, based on a preset first detection period, at least one response time of the target edge IP is detected within a first preset duration, wherein the first preset duration includes at least one of the first detection periods; the summation average of each response time is calculated to obtain an average response time; based on a normalization processing algorithm, the average response time is scored to obtain the link response quality score of the target edge IP.

[0063] For example, multiple monitoring machines can be used to detect the total response time of the target edge IP transmitting an 8-bit file every 5 seconds; the multiple response time data detected within 1 minute are summarized according to the target edge IP granularity, such as... Figure 3 As shown, the response time is collected every 5 seconds, and all response time data collected within one minute are statistically analyzed.

[0064] In one embodiment, after detecting multiple cycle response times, the average response time can be calculated, and then normalized according to min-max. The maximum response time is 0.5s with a score of 0, and the minimum response time is 0s with a score of 100. The link response quality score is calculated in the range of 0-100.

[0065] Further, based on a preset second detection period, at least one packet loss rate of the target edge IP is detected within a second preset duration, wherein the second preset duration includes at least one of the first detection periods; the summation average of each packet loss rate is calculated to obtain an average packet loss rate; based on the normalization processing algorithm, the average packet loss rate is scored to obtain the packet loss rate quality score of the target edge IP.

[0066] For example, multiple monitoring machines can be used to ping the target edge IP with 10 packets every 5 seconds, and the response results of the target edge IP can be counted; the packet loss rate data obtained within 1 minute can be summarized at the target edge IP granularity, and so on. Figure 3 As shown, the packet loss rate is collected every 5 seconds, and the packet loss rate data of the target edge IP is statistically analyzed within one minute.

[0067] In one embodiment, after detecting multiple periods of packet loss rate data, the average packet loss rate is calculated, and then normalized according to min-max. The maximum packet loss rate is 20%, with a score of 0, and the minimum packet loss rate is 0%, with a score of 100. The packet loss rate quality score is calculated in the range of 0-100.

[0068] S202. Based on the minimum value MIN function, compare the link response quality score and the packet loss rate quality score to determine the network quality score of the target edge IP.

[0069] In one embodiment, the link response time and packet loss rate of the target edge IP are detected in real time and converted into a network quality score of 0-100 using min-max normalization, which can promptly detect edge IPs with network quality problems.

[0070] The formula for calculating the network quality score of the target edge IP is as follows:

[0071] Network Quality Score = MIN(Link Response Quality Score, Packet Loss Rate Quality Score)

[0072] In one embodiment, the MIN function returns the minimum value among the parameters.

[0073] In one embodiment, if the link response quality score is less than the packet loss rate quality score, then the link response quality score is determined to be the network quality score; if the link response quality score is greater than the packet loss rate quality score, then the packet loss rate quality score is determined to be the network quality score.

[0074] For example, suppose a single monitoring machine collects data such as Figure 3 As shown, the average response time at the one-minute granularity is 40.5ms. Since there are multiple monitoring machines, the average of these multiple machines is used. After min-max normalization, the link response quality score is 91.9. The average packet loss rate at the one-minute granularity is 6%. Since there are multiple monitoring machines, the average of these multiple machines is used. After min-max normalization, the packet loss rate quality score is 70. Therefore, the network quality score for this edge IP is calculated as 70 based on the MIN function.

[0075] S203. Based on the preset network quality score threshold and the comparison result of the network quality score, determine whether it is necessary to perform bandwidth pulling operation on the target edge IP.

[0076] In one embodiment, the current network quality of the target edge IP is determined by comparing its network quality score with a network quality score threshold, thus determining whether it can fully respond to the content acceleration request initiated by the client. If it can fully respond to the request, the target edge IP is responded to directly; if it cannot fully respond to the request, at least one alternative edge IP is selected, and bandwidth diversion is performed on the target edge IP.

[0077] Further, the bandwidth pulling ratio and preset bandwidth pulling step size of the target edge IP are obtained; based on the comparison results of the network quality score and the network quality score threshold, the bandwidth pulling ratio and the bandwidth pulling step size are calculated by stepwise summation or stepwise difference to obtain the first bandwidth pulling ratio of the target edge IP; when the first bandwidth pulling ratio is not zero, it is determined that the bandwidth pulling operation needs to be performed on the target edge IP.

[0078] In one embodiment, the domain name-region information and real-time bearer bandwidth information served by the edge IP are obtained. Based on the domain name-region information, the edge IP resources in the resource pool corresponding to the domain name-region and the real-time redundant bandwidth of the edge IP are obtained, and the bandwidth ratio pull step size is configured as N. The edge IPs are traversed, and it is determined whether the real-time network quality of the edge IP is less than the configured threshold.

[0079] Wherein, the domain name is the domain name of the customer who needs accelerated service; the region is the geographical location of the user plus the carrier's granular planning; the real-time carrying bandwidth is the bandwidth traffic carried by the edge IP in real time at this domain-region granularity.

[0080] Wherein, the edge IP of the resource pool is the edge IP that can be replaced when the network quality of the current domain name-region granular service is poor; the real-time redundant bandwidth of the edge IP is equal to the rated bandwidth of the edge IP minus the real-time bandwidth, and the rated bandwidth is considered to be the maximum bandwidth that the edge IP can provide services.

[0081] In one embodiment, if the network quality score is greater than or equal to the threshold, it indicates that the edge IP network is healthy, and the bandwidth pulling ratio is increased by N, with a maximum bandwidth pulling ratio of 100; if the network quality score is less than the threshold, it indicates that the edge IP network is unhealthy, and the service bandwidth needs to be gradually reduced, and the bandwidth pulling ratio is decreased by N, with a minimum bandwidth pulling ratio of 0.

[0082] In one embodiment, the calculated bandwidth pulling ratio is judged. If the bandwidth pulling ratio is 0, it means that bandwidth pulling is not required. The original edge IP of the bandwidth pulling module is transmitted and the response ratio is 100, indicating that the bandwidth is completely responded by itself. If the bandwidth pulling ratio is not 0, it means that bandwidth pulling is required. The domain name-region served by this edge IP is traversed.

[0083] This embodiment generates a network quality score by real-time detection of the link response quality and packet loss rate of edge IPs, enabling timely detection and response when network quality issues occur at the edge IP. It employs a proportional, step-by-step bandwidth allocation method: when an edge IP experiences network quality problems, bandwidth is gradually shifted to a healthy replacement edge IP; when the edge IP's network quality recovers, bandwidth is gradually shifted back to it. Compared to directly replacing the entire edge IP, this method offers higher bandwidth utilization, stronger availability, and greater bandwidth cost savings.

[0084] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating the third embodiment of the CDN edge IP network quality scheduling method provided in this application.

[0085] In this embodiment, based on the above... Figure 1 In the illustrated embodiment, step S102 specifically includes:

[0086] S301. Obtain the bandwidth pulling ratio and carrying bandwidth of the target edge IP;

[0087] S302. Based on the bandwidth traction ratio and the carrying bandwidth, calculate the traction bandwidth of the domain name-region corresponding to the target edge IP;

[0088] In one embodiment, if the bandwidth pulling ratio of the target edge IP is not 0, it indicates that bandwidth pulling needs to be performed on the target edge IP. This involves traversing the domain-region served by this edge IP; calculating the pulling bandwidth OutBw at the edge IP-domain-region granularity using the following formula:

[0089] OutBw = BearBw * OutRate

[0090] BearBw represents the bandwidth carried by the edge IP at the domain-region granularity, and OutRate represents the bandwidth pull ratio of the edge IP.

[0091] S303. Obtain the comprehensive redundant bandwidth of each candidate edge IP in the resource pool corresponding to the domain name-region;

[0092] S304. When there is a candidate edge IP in the resource pool with a comprehensive redundant bandwidth greater than or equal to the traction bandwidth, the candidate edge IP shall be used as the replacement edge IP for the target edge IP.

[0093] Wherein, the bandwidth pulling ratio is the second bandwidth response ratio corresponding to the replacement edge IP, and the difference between the maximum bandwidth response ratio and the bandwidth pulling ratio is the first bandwidth response ratio corresponding to the target edge IP.

[0094] In one embodiment, a replacement edge IP that can carry the traction bandwidth is found in the domain-zone resource pool based on the traction bandwidth. The real-time comprehensive redundancy bandwidth of the replacement edge IP needs to be greater than or equal to the traction bandwidth OutBw. The first bandwidth response ratio corresponding to the target edge IP is 100 - bandwidth traction ratio, and the second bandwidth response ratio corresponding to the replacement edge IP is the bandwidth traction ratio.

[0095] For example, the acceleration bandwidth ratio pull step size N can be configured to 10, the unhealthy network quality score threshold can be configured to 75, and the network quality score of the edge IP at different times is as follows: Figure 5 As shown.

[0096] When the network quality of the outgoing edge IP becomes unhealthy at score 01, 10% of the bandwidth is redirected to the replacement edge IP with healthy network quality. Subsequently, because the network quality score of the outgoing edge IP remains below the threshold, the bandwidth redirection ratio is gradually increased until all bandwidth is redirected to the replacement edge IP. When the network quality score of the outgoing edge IP exceeds the configured threshold at score 12, bandwidth begins to be gradually restored until it is fully restored at score 21.

[0097] Furthermore, if the edge IP is only experiencing sudden LVS scheduling congestion, such as Figure 6 As shown, within 7 minutes, only the network quality scores of the replaced edge IPs (01 to 03) were below the threshold. At this point, a maximum of 30% of the bandwidth was diverted to the replacement edge IP, and normal operation was restored within 06 minutes. This type of problem does not require a complete edge IP granularity switch; simply diverting a portion of the bandwidth to the replacement edge IP with healthy network quality within a short period can resolve the issue.

[0098] Please see Figure 7 , Figure 7 This is a schematic diagram of a first embodiment of a CDN edge IP network quality scheduling device provided in this application. This CDN edge IP network quality scheduling device is used to execute the aforementioned CDN edge IP network quality scheduling method. The CDN edge IP network quality scheduling device can be configured in a server.

[0099] like Figure 7 As shown, the CDN edge IP network quality scheduling device 400 includes: a bandwidth traction execution judgment module 401, a data acquisition module 402, a random number generation module 403, a target edge IP response module 404, and a replacement edge IP response module 405.

[0100] The bandwidth traction execution judgment module 401 is used to determine whether a bandwidth traction operation needs to be performed on the target edge IP based on the network quality score of the target edge IP when a content acceleration request for the target edge IP is received from the client.

[0101] The data acquisition module 402 is used to acquire, when it is necessary to perform the bandwidth pulling operation on the target edge IP, the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP.

[0102] The random number generation module 403 is used to generate random numbers based on the random number generation method;

[0103] The target edge IP response module 404 is used to invoke the target edge IP to respond to the content acceleration request when the random number corresponds to the first bandwidth response ratio;

[0104] The replacement edge IP response module 405 is used to invoke the replacement edge IP response to the content acceleration request when the random number corresponds to the second bandwidth response ratio.

[0105] In one embodiment, the bandwidth traction execution determination module 401 includes:

[0106] The quality score detection unit is used to detect the link response quality score and packet loss rate quality score of the target edge IP based on preset detection rules.

[0107] The network quality score determination unit is used to determine the network quality score of the target edge IP by comparing the link response quality score and the packet loss rate quality score based on the minimum value MIN function.

[0108] The bandwidth traction execution judgment unit is used to determine whether a bandwidth traction operation needs to be performed on the target edge IP based on a preset network quality score threshold and a comparison result of the network quality score.

[0109] In one embodiment, the mass fraction detection unit includes:

[0110] The response time detection subunit is used to detect at least one response time of the target edge IP within a first preset duration based on a preset first detection period, wherein the first preset duration includes at least one of the first detection periods;

[0111] The response time average calculation subunit is used to calculate the summation average of the response times to obtain the response time average.

[0112] The link response quality score acquisition sub-unit is used to score the average response time based on a normalization processing algorithm to obtain the link response quality score of the target edge IP.

[0113] In one embodiment, the mass fraction detection unit further includes:

[0114] A packet loss rate detection subunit is used to detect at least one packet loss rate of the target edge IP within a second preset duration based on a preset second detection period, wherein the second preset duration includes at least one of the first detection periods;

[0115] The average packet loss rate calculation subunit is used to calculate the summation average of the packet loss rates to obtain the average packet loss rate.

[0116] The packet loss rate quality score acquisition subunit is used to score the average packet loss rate based on the normalization processing algorithm to obtain the packet loss rate quality score of the target edge IP.

[0117] In one embodiment, the bandwidth traction execution determination unit includes:

[0118] The bandwidth pulling ratio acquisition subunit is used to acquire the bandwidth pulling ratio and preset bandwidth pulling step size of the target edge IP.

[0119] The bandwidth pulling ratio calculation subunit is used to perform step-by-step summation or step-by-step difference calculation on the bandwidth pulling ratio and the bandwidth pulling step size based on the score comparison result of the network quality score and the network quality score threshold, so as to obtain the first bandwidth pulling ratio of the target edge IP.

[0120] A bandwidth traction execution determination subunit is used to determine that the bandwidth traction operation needs to be performed on the target edge IP when the first bandwidth traction ratio is not zero.

[0121] In one embodiment, the data acquisition module 402 includes:

[0122] The bandwidth acquisition unit is used to acquire the bandwidth pulling ratio and the carrying bandwidth of the target edge IP.

[0123] The traction bandwidth calculation unit is used to calculate the traction bandwidth of the domain name-zone corresponding to the target edge IP based on the bandwidth traction ratio and the carrying bandwidth;

[0124] The comprehensive redundancy bandwidth acquisition unit is used to acquire the comprehensive redundancy bandwidth of each candidate edge IP in the resource pool corresponding to the domain name-region.

[0125] The replacement edge IP determination unit is used to select the candidate edge IP as the replacement edge IP of the target edge IP when there is a candidate edge IP in the resource pool with a comprehensive redundant bandwidth greater than or equal to the traction bandwidth.

[0126] Wherein, the bandwidth pulling ratio is the second bandwidth response ratio corresponding to the replacement edge IP, and the difference between the maximum bandwidth response ratio and the bandwidth pulling ratio is the first bandwidth response ratio corresponding to the target edge IP.

[0127] In one embodiment, the network quality score determination unit includes:

[0128] The first sub-unit for determining network quality score is used to determine the link response quality score as the network quality score if the link response quality score is less than the packet loss rate quality score.

[0129] The second sub-unit for determining network quality score is used to determine the packet loss rate quality score as the network quality score if the link response quality score is greater than the packet loss rate quality score.

[0130] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device and each module described above can be referred to the corresponding process in the aforementioned embodiment of the CDN edge IP network quality scheduling method, and will not be repeated here.

[0131] The apparatus provided in the above embodiments can be implemented as a computer program, which can be used in, for example... Figure 8 It runs on the computer device shown.

[0132] Please see Figure 8 , Figure 8 This is a schematic block diagram illustrating the structure of a computer device according to an embodiment of this application. The computer device may be a server.

[0133] See Figure 8 The computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0134] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any CDN edge IP network quality scheduling method.

[0135] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0136] Internal memory provides an environment for the execution of computer programs stored in non-volatile storage media. When these computer programs are executed by a processor, the processor can perform any CDN edge IP network quality scheduling method.

[0137] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0138] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0139] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps:

[0140] When a client sends a request to accelerate content for a target edge IP, the system determines whether bandwidth pulling operation needs to be performed on the target edge IP based on the network quality score of the target edge IP.

[0141] When it is necessary to perform the bandwidth pulling operation on the target edge IP, obtain the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP;

[0142] Generate random numbers based on random number generation methods;

[0143] When the random number corresponds to the first bandwidth response ratio, the target edge IP is invoked to respond to the content acceleration request;

[0144] When the random number corresponds to the second bandwidth response ratio, the replacement edge IP is invoked to respond to the content acceleration request.

[0145] In one embodiment, when the processor determines whether a bandwidth-pushing operation needs to be performed on the target edge IP based on the network quality score of the target edge IP, it is configured to:

[0146] Based on preset detection rules, the link response quality score and packet loss rate quality score of the target edge IP are detected;

[0147] Based on the minimum value MIN function, the network quality score of the target edge IP is determined by comparing the link response quality score and the packet loss rate quality score.

[0148] Based on the preset network quality score threshold and the comparison result of the network quality score, it is determined whether a bandwidth pulling operation needs to be performed on the target edge IP.

[0149] In one embodiment, when the processor implements the detection of the link response quality score and packet loss rate quality score of the target edge IP based on preset detection rules, it is configured to:

[0150] Based on a preset first detection period, at least one response time of the target edge IP is detected within a first preset duration, wherein the first preset duration includes at least one of the first detection periods;

[0151] Calculate the summation average of the various response times to obtain the average response time;

[0152] Based on the normalization processing algorithm, the average response time is scored to obtain the link response quality score of the target edge IP.

[0153] In one embodiment, when the processor implements the detection of the link response quality score and packet loss rate quality score of the target edge IP based on preset detection rules, it is also configured to:

[0154] Based on a preset second detection period, at least one packet loss rate of the target edge IP is detected within a second preset duration, wherein the second preset duration includes at least one of the first detection periods;

[0155] Calculate the summation average of the packet loss rates mentioned above to obtain the average packet loss rate;

[0156] Based on the normalization processing algorithm, the average packet loss rate is scored to obtain the packet loss rate quality score of the target edge IP.

[0157] In one embodiment, when the processor determines whether a bandwidth-pushing operation needs to be performed on the target edge IP based on a comparison of a preset network quality score threshold and the network quality score, it is configured to:

[0158] Obtain the bandwidth pulling ratio and preset bandwidth pulling step size of the target edge IP;

[0159] Based on the comparison results of the network quality score and the network quality score threshold, the bandwidth pulling ratio and the bandwidth pulling step size are calculated by summing or subtracting step by step to obtain the first bandwidth pulling ratio of the target edge IP.

[0160] When the first bandwidth pulling ratio is not zero, it is determined that the bandwidth pulling operation needs to be performed on the target edge IP.

[0161] In one embodiment, when the processor implements the steps of obtaining the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP, it is configured to:

[0162] Obtain the bandwidth pull ratio and carrying bandwidth of the target edge IP;

[0163] Based on the bandwidth traction ratio and the carrying bandwidth, calculate the traction bandwidth of the domain name-zone corresponding to the target edge IP;

[0164] Obtain the comprehensive redundant bandwidth of each candidate edge IP in the resource pool corresponding to the domain name-region;

[0165] When there is a candidate edge IP in the resource pool with a comprehensive redundant bandwidth greater than or equal to the traction bandwidth, the candidate edge IP is used as the replacement edge IP for the target edge IP;

[0166] Wherein, the bandwidth pulling ratio is the second bandwidth response ratio corresponding to the replacement edge IP, and the difference between the maximum bandwidth response ratio and the bandwidth pulling ratio is the first bandwidth response ratio corresponding to the target edge IP.

[0167] In one embodiment, when the processor implements the method of determining the network quality score of the target edge IP by comparing the link response quality score and the packet loss rate quality score based on the minimum value MIN function, it is configured to:

[0168] If the link response quality score is less than the packet loss rate quality score, then the link response quality score is determined to be the network quality score.

[0169] If the link response quality score is greater than the packet loss rate quality score, then the packet loss rate quality score is determined to be the network quality score.

[0170] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement any of the CDN edge IP network quality scheduling methods provided in the embodiments of this application.

[0171] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMediaCard (SMC), SecureDigital (SD) card, or FlashCard equipped on the computer device.

[0172] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A CDN edge IP network quality scheduling method, characterized in that, The method includes: When a client sends a request to accelerate content for a target edge IP, the system determines whether bandwidth pulling operation needs to be performed on the target edge IP based on the network quality score of the target edge IP. When it is necessary to perform the bandwidth pulling operation on the target edge IP, obtain the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP; Generate random numbers based on random number generation methods; When the random number corresponds to the first bandwidth response ratio, the target edge IP is invoked to respond to the content acceleration request; When the random number corresponds to the second bandwidth response ratio, the replacement edge IP is invoked to respond to the content acceleration request; The step of determining whether bandwidth redirection is needed for the target edge IP based on its network quality score includes: Based on preset detection rules, the link response quality score and packet loss rate quality score of the target edge IP are detected; Based on the minimum value MIN function, the network quality score of the target edge IP is determined by comparing the link response quality score and the packet loss rate quality score. Based on the preset network quality score threshold and the comparison result of the network quality score, it is determined whether a bandwidth pulling operation needs to be performed on the target edge IP. The step of determining whether to perform bandwidth redirection on the target edge IP based on a comparison of a preset network quality score threshold and the network quality score includes: Obtain the bandwidth pulling ratio and preset bandwidth pulling step size of the target edge IP; Based on the comparison results of the network quality score and the network quality score threshold, the bandwidth pulling ratio and the bandwidth pulling step size are calculated by summing or subtracting step by step to obtain the first bandwidth pulling ratio of the target edge IP. When the first bandwidth pulling ratio is not zero, it is determined that the bandwidth pulling operation needs to be performed on the target edge IP. The step of obtaining the replacement edge IP corresponding to the target edge IP includes: Obtain the bandwidth pull ratio and carrying bandwidth of the target edge IP; Based on the bandwidth traction ratio and the carrying bandwidth, calculate the traction bandwidth of the domain name-zone corresponding to the target edge IP; Obtain the comprehensive redundant bandwidth of each candidate edge IP in the resource pool corresponding to the domain name-region; When there is a candidate edge IP in the resource pool with a comprehensive redundant bandwidth greater than or equal to the traction bandwidth, the candidate edge IP is used as a replacement edge IP for the target edge IP.

2. The CDN edge IP network quality scheduling method according to claim 1, characterized in that, The method of detecting the link response quality score and packet loss rate quality score of the target edge IP based on preset detection rules includes: Based on a preset first detection period, at least one response time of the target edge IP is detected within a first preset duration, wherein the first preset duration includes at least one of the first detection periods; Calculate the summation average of the various response times to obtain the average response time; Based on the normalization processing algorithm, the average response time is scored to obtain the link response quality score of the target edge IP.

3. The CDN edge IP network quality scheduling method according to claim 2, characterized in that, The method of detecting the link response quality score and packet loss rate quality score of the target edge IP based on preset detection rules also includes: Based on a preset second detection period, at least one packet loss rate of the target edge IP is detected within a second preset duration, wherein the second preset duration includes at least one of the first detection periods; Calculate the summation average of the packet loss rates mentioned above to obtain the average packet loss rate; Based on the normalization processing algorithm, the average packet loss rate is scored to obtain the packet loss rate quality score of the target edge IP.

4. The CDN edge IP network quality scheduling method according to claim 1, characterized in that, The bandwidth pulling ratio is the second bandwidth response ratio corresponding to the replacement edge IP, and the difference between the maximum bandwidth response ratio and the bandwidth pulling ratio is the first bandwidth response ratio corresponding to the target edge IP.

5. The CDN edge IP network quality scheduling method according to claim 1, characterized in that, The determination of the network quality score of the target edge IP by comparing the link response quality score and the packet loss rate quality score based on the minimum value MIN function includes: If the link response quality score is less than the packet loss rate quality score, then the link response quality score is determined to be the network quality score. If the link response quality score is greater than the packet loss rate quality score, then the packet loss rate quality score is determined to be the network quality score.

6. A CDN edge IP network quality scheduling device, characterized in that, The CDN edge IP network quality scheduling device includes: The bandwidth traction execution judgment module is used to determine whether a bandwidth traction operation needs to be performed on the target edge IP based on the network quality score of the target edge IP when a content acceleration request for the target edge IP is received from the client. The data acquisition module is used to acquire, when it is necessary to perform the bandwidth pulling operation on the target edge IP, the replacement edge IP corresponding to the target edge IP, the first bandwidth response ratio corresponding to the target edge IP, and the second bandwidth response ratio corresponding to the replacement edge IP; The random number generation module is used to generate random numbers based on random number generation methods. The target edge IP response module is used to invoke the target edge IP to respond to the content acceleration request when the random number corresponds to the first bandwidth response ratio; The replacement edge IP response module is used to invoke the replacement edge IP to respond to the content acceleration request when the random number corresponds to the second bandwidth response ratio; The bandwidth traction execution judgment module includes: The quality score detection unit is used to detect the link response quality score and packet loss rate quality score of the target edge IP based on preset detection rules. The network quality score determination unit is used to determine the network quality score of the target edge IP by comparing the link response quality score and the packet loss rate quality score based on the minimum value MIN function. The bandwidth pulling execution judgment unit is used to determine whether a bandwidth pulling operation needs to be performed on the target edge IP based on a preset network quality score threshold and a comparison result of the network quality score. The bandwidth traction execution judgment unit includes: The bandwidth pulling ratio acquisition subunit is used to acquire the bandwidth pulling ratio and preset bandwidth pulling step size of the target edge IP. The bandwidth pulling ratio calculation subunit is used to perform step-by-step summation or step-by-step difference calculation on the bandwidth pulling ratio and the bandwidth pulling step size based on the score comparison result of the network quality score and the network quality score threshold, so as to obtain the first bandwidth pulling ratio of the target edge IP. A bandwidth traction execution determination subunit is used to determine that the bandwidth traction operation needs to be performed on the target edge IP when the first bandwidth traction ratio is not zero. The data acquisition module includes: The bandwidth acquisition unit is used to acquire the bandwidth pulling ratio and the carrying bandwidth of the target edge IP. The traction bandwidth calculation unit is used to calculate the traction bandwidth of the domain name-zone corresponding to the target edge IP based on the bandwidth traction ratio and the carrying bandwidth; The comprehensive redundancy bandwidth acquisition unit is used to acquire the comprehensive redundancy bandwidth of each candidate edge IP in the resource pool corresponding to the domain name-region. The replacement edge IP determination unit is used to select a candidate edge IP as a replacement edge IP for the target edge IP when there is a candidate edge IP in the resource pool with a comprehensive redundant bandwidth greater than or equal to the traction bandwidth.

7. A computer device, characterized in that, The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the CDN edge IP network quality scheduling method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the CDN edge IP network quality scheduling method as described in any one of claims 1 to 5.

Images