A CDN Coverage Resource Selection Method Based on Multi-Factor Weighting

Abstract

This invention discloses a CDN coverage resource selection method based on multi-factor weighting scoring, comprising the following steps: Step S1, obtaining the planned bandwidth for generating DNS group-region coverage resources; Step S2, obtaining the upstream traffic IP resources of the DNS group-region; Step S3, sorting the upstream traffic IP resources of the DNS group-region according to distance level. This invention sorts the upstream traffic IPs according to two dimensions: distance level and redundant bandwidth, which can prioritize the selection of the optimal upstream traffic IPs for testing, providing support for the subsequent IP selection pool algorithm. Using the IP selection pool algorithm, the optimal solution combination can be calculated in a single traversal, which is faster than the exhaustive search method and significantly saves system overhead. The multi-factor weighting scoring algorithm calculates scores by controlling the level weight factor and the number factor, which can comprehensively consider both IP distance level and IP number to arrive at the optimal solution.

Description

Technical Field

[0001] This invention relates to the field of Internet technology, specifically to a CDN coverage resource selection method, electronic device, and storage medium based on multi-factor weighting scoring. Background Technology

[0002] A Content Delivery Network (CDN) is a new type of network content service system built on IP networks. It provides content distribution and services based on the efficiency, quality, and order requirements of content access and application. It can avoid bottlenecks and links on the Internet that may affect the speed and stability of data transmission as much as possible, making content transmission faster and more stable.

[0003] CDNs typically deploy a set of uplink IPs (IP addresses of edge node servers exposed to the outside world, allowing users to access content) at the regional granularity of a DNS resolution group (a set of domain names that need CDN acceleration). When a user requests content, they do not need to access the origin server; instead, the content is retrieved from the uplink IPs within the CDN's coverage plan through the CDN scheduling system. This reduces the access pressure on the origin server and allows users to obtain content from the nearest available IP, shortening request times and achieving fast access.

[0004] Resource planning at the DNS resolution group / region level is not simply a matter of deploying a few uplink IPs. Since uplink IPs can serve multiple DNS resolution groups / regions, and multiple uplink IPs can be planned at the DNS resolution group / region level, the available planning options are very limited. When planning coverage resources, the geographical distance between the uplink IPs and the DNS resolution groups / regions must be considered. Longer distances not only increase network transmission time but also consume cross-provincial bandwidth if cross-provincial transmission occurs, increasing the load on provincial links and impacting network service quality. Simultaneously, it's necessary to consider whether the bandwidth redundancy of the planned uplink IPs can meet the bandwidth requirements of the DNS resolution groups / regions, as well as the limitation on the number of planned uplink IPs (too many uplink IPs can lead to caching multiple copies of the same data, wasting memory space, and when there is no cache, returning data to the previous level results in excessive bandwidth consumption and increased bandwidth costs). This makes it very difficult to select a suitable uplink IP solution to serve different DNS resolution groups / regions, achieving both optimal bandwidth cost and good network service quality.

[0005] Therefore, there is a need to provide a CDN coverage resource selection method, electronic device, and storage medium based on multi-factor weighting to improve the above situation.

[0006] Currently, the planning and deployment of traffic IPs at the resolution group-region granularity are generally done manually based on experience. This approach leads to several problems:

[0007] 1) Inefficient: Manual planning requires planning and deployment of each analysis group and region individually, which is time-consuming and labor-intensive;

[0008] 2) Human error: Due to the large number of combination schemes and the many factors that need to be considered, it is not always possible for humans to plan the best coverage, which is prone to human error. If the coverage is not planned reasonably, it will not only increase bandwidth costs but may also affect the quality of network services.

[0009] 3) Increased bandwidth costs: Manual planning cannot take into account bandwidth costs while meeting network service quality requirements through complex algorithms, unlike programs.

[0010] Therefore, there is a need to provide a CDN coverage resource selection method, electronic device, and storage medium based on multi-factor weighting to improve the above situation. Summary of the Invention

[0011] The purpose of this invention is to provide a CDN coverage resource selection method, electronic device and storage medium based on multi-factor weighting scoring, so as to solve the problems of low efficiency, human error and increased bandwidth cost mentioned in the background art.

[0012] To achieve the above objectives, the present invention provides the following technical solution:

[0013] A CDN coverage resource selection method based on multi-factor weighting scoring includes the following steps:

[0014] Step S1: Obtain the planned bandwidth for generating the resolution group - area coverage resources;

[0015] Step S2: Obtain the upstream IP resources of the resolution group-region;

[0016] Step S3: Sort the IP resources of the resolution group-region according to the distance level;

[0017] Step S4: Using the maximum number of IPs to be selected M as the boundary value, initialize and create IP selection pools from 1 to M. The number of IPs that each selection pool can hold is equal to the number of their respective sequence numbers.

[0018] Step S5: Iterate through the sorted IPs with the highest traffic and determine whether the redundant bandwidth of each IP can be filled into the M selection pools.

[0019] Step S6: When all IP resources at the resolution group-region granularity have been traversed, calculate the score for each selection pool.

[0020] As a further technical solution of the present invention: the up-traffic IP resources include the distance level between the geographical location of the IP and the coverage area and the IP redundancy bandwidth.

[0021] As a further technical solution of the present invention: the distance levels are divided into: the same province planning level, the same region planning level, and the same operator planning level.

[0022] As a further technical solution of the present invention: the sorting order in step 3 is that the planning level of the same province is greater than the planning level of the same region, which is greater than the planning level of the same operator. Within the same level, the IP redundancy bandwidth is sorted in descending order, with the one with the larger redundancy bandwidth placed first.

[0023] As a further technical solution of the present invention: the number of available traffic IPs pushed by the CDN domain name resolution server to the client is limited by the packet length.

[0024] As a further technical solution of the present invention: step 5 specifically includes:

[0025] Step S51: Round up the result of dividing the planned bandwidth of the resolution group-region by the redundant bandwidth of the IP. The resulting number N represents the number of IPs that can be placed into the selection pool of N and above.

[0026] Step S52: Iterate through the selection pools from N onwards. If the number of IPs in the selection pool is less than the selection pool number, then put the IP into the selection pool; otherwise, do not put it into the selection pool.

[0027] As a further technical solution of the present invention: step 6 specifically includes:

[0028] Step S61: Traverse each selection pool, determine the number of IPs in the selection pool, skip if the number of IPs is not equal to the selection pool number, otherwise calculate the score;

[0029] Step S62: Calculate the score for all IPs in the selection pool using the following formula:

[0030]

[0031] In the above formula, x represents the number of IPs in the same planning level in the selection pool, y represents the number of IPs in the regional planning level in the selection pool, z represents the number of IPs in the operator planning level in the selection pool, S represents the number weight, X represents the weight of the same planning level, Y represents the weight of the regional planning level, and Z represents the weight of the operator planning level.

[0032] As a further technical solution of the present invention: by controlling the value range of X, Y, Z and S, the weight of IP distance level and number of IPs can be controlled.

[0033] An electronic device includes a memory and a processor, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the aforementioned CDN coverage resource selection method based on multi-factor weighting.

[0034] A storage medium storing a computer program, which, when executed by a processor, implements the aforementioned CDN coverage resource selection method based on multi-factor weighting.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. By sorting upstream IPs according to two dimensions—distance level and redundant bandwidth—the system can prioritize selecting the best upstream IPs for testing, thus supporting the subsequent IP selection pool algorithm.

[0037] 2. By adopting the IP selection pool algorithm, the optimal combination of solutions can be calculated in one traversal, which is faster than the exhaustive search method and saves a lot of system overhead.

[0038] 3. A multi-factor weighted scoring algorithm is adopted. By controlling the hierarchical weight factor and the number factor, the score is calculated, which can comprehensively consider the two factors of IP distance level and IP number to obtain the optimal solution. Attached Figure Description

[0039] Figure 1 A flowchart illustrating the CDN coverage resource selection method based on multi-factor weighting.

[0040] Figure 2 This is a diagram illustrating the IP selection pool.

[0041] Figure 3 A diagram illustrating the IP selection results. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] like Figure 1-3 As shown, a CDN coverage resource selection method based on multi-factor weighting scoring is proposed. Figure 1 As shown, it includes the following steps:

[0044] Step S1: Obtain the planned bandwidth for generating the resolution group - area coverage resources;

[0045] In implementation, the resolution group is a set of domain names that require CDN acceleration services; the region is the user's geographical location plus carrier-level planning, such as: China Telecom Beijing, China Unicom Hangzhou, China Mobile Fujian, etc.; the planned bandwidth is the amount of bandwidth required by the resolution group-region.

[0046] Step S2: Obtain the IP resources of the resolution group-region, including the distance level between the geographical location of the IP and the coverage area and the IP redundancy bandwidth. The distance level is divided into: same province planning level, same region planning level, and same operator planning level.

[0047] In implementation, for IP resource planning within a resolution group-region, generally speaking, the closer the geographical location of the IP is to the area it serves, the better, without considering other factors. Closer geographical proximity not only reduces the overall request link duration and improves network service quality, but also saves inter-provincial bandwidth for operators, reducing bandwidth costs. The "same province planning level" indicates that the coverage area and the IP address are geographically located within the same province; the "same region planning level" indicates that the coverage area and the IP address are geographically located within the same region; and the "same operator planning level" indicates that the coverage area and the IP address are geographically located within the same operator.

[0048] Step S3: Sort the IP resources of the resolution group-region according to the distance level, in the order of the same province planning level > the same region planning level > the same operator planning level. Within the same level, sort them in descending order according to the IP redundancy bandwidth, with the one with the larger redundancy bandwidth at the front.

[0049] In implementation, assuming the coverage area is Beijing Telecom, IP1 and IP2 belong to Beijing Telecom with redundant bandwidths of 80M and 100M respectively, IP3 belongs to Tianjin Telecom with a redundant bandwidth of 120M, and IP4 belongs to Fujian Telecom with a redundant bandwidth of 140M. The sorted order is: IP2, IP1, IP3, IP4.

[0050] Step S4: Since the number of available IPs for requesting traffic pushed to customers by the CDN domain name resolution server is limited by the packet length, an IP selection pool of 1 to M (maximum number of IPs to select) is initialized with the maximum number M as the boundary value. Each selection pool can hold an IP number equal to its corresponding sequence number.

[0051] In implementation, such as Figure 2 As shown, assuming the maximum number M equals 7, then 7 IP selection pools, numbered 1 to 7, are created initially. Each selection pool can hold a maximum of the number of IPs corresponding to the selection pool number.

[0052] Step S5: Iterate through the sorted IPs with increasing traffic and determine whether the redundant bandwidth of each IP can be filled into the M selection pools. The main steps are as follows:

[0053] Step S51: Round up the result of dividing the planned bandwidth of the resolution group-region by the redundant bandwidth of the IP. The resulting number N represents the number of IPs that can be placed into the selection pool of N and above.

[0054] Step S52: Iterate through the selection pools from N onwards. If the number of IPs in the selection pool is less than the selection pool number, the IP can be added to the selection pool; otherwise, it is not added to the selection pool.

[0055] In implementation, it is assumed that the current resolution group-region's regional granularity is China Telecom Fujian, with a planned bandwidth of 700M; IP1, IP2, IP3, IP4, IP5, IP6, and IP7 all belong to China Telecom Fujian, with redundancies of 240M, 240M, 100M, 100M, 100M, 100M, and 100M respectively; IP8 and IP9 both belong to China Telecom Guangdong, with redundancies of 240M and 240M respectively; IP10 belongs to China Telecom Beijing, with a redundancy of 700M. Iterate through IPs 1 through 10. For IP1, at least three IPs with the same redundant bandwidth are needed to satisfy the planned bandwidth of the resolution group-zone, i.e., N=3. Place IP1 in selection pools 3 through 7. For IP2, N=3, place it in selection pools 3 through 7. For IP3, N=7, place it in selection pool 7. For IP4, N=7, place it in selection pool 7. For IP5, N=7, place it in selection pool 7. For IP6, N=7, place it in selection pool 7. For IP7, N=7, place it in selection pool 7. For IP8, N=3, place it in selection pools 3 and 6. Selection pool 7 already has 7 IPs, so no more are added. For IP9, N=3, place it in selection pools 4 through 6. Selection pool 3 also already has 3 IPs, so no more are added. For IP10, N=1, place it in selection pools 1, 2, 5, and 6. The final result is as follows. Figure 3 As shown.

[0056] Step S6: After all IP resources at the resolution group-region granularity have been traversed, score each selection pool. The main steps are as follows:

[0057] Step S61: Traverse each selection pool, determine the number of IPs in the selection pool, skip if the number of IPs is not equal to the selection pool number, otherwise calculate the score;

[0058] Step S62: Calculate the weights of all IPs in the selection pool using the following formula:

[0059]

[0060] In the above formula, x represents the number of IPs in the same province planning level in the selection pool, y represents the number of IPs in the same region planning level in the selection pool, z represents the number of IPs in the same operator planning level in the selection pool, S represents the number weight, X represents the weight of the same province planning level, Y represents the weight of the same region planning level, and Z represents the weight of the same operator planning level. By controlling the value range of X, Y, Z, and S, the weight of IP distance level and IP number can be controlled.

[0061] Step S64: Compare the scores and select the IP from the pool corresponding to the highest score as the best solution for deployment.

[0062] In implementation, the weights are set as follows: S = 0.04 for the number of items; X = 1 for the planning level within the same province; Y = 0.9 for the planning level within the same region; and Z = 0.8 for the planning level within the same operator. For example... Figure 3 The selection results shown are used to calculate the scores for each selection pool. Selection pool 1 has only one IP from the same carrier's planning level, scoring 80 points. Selection pool 2 has only one IP, so no points are calculated. Selection pool 3 has two IPs from the same province's planning level and one IP from the same region's planning level, scoring 88.9 points. Selection pool 4 has two IPs from the same province's planning level and two IPs from the same region's planning level, scoring 83.6 points. Selection pool 5 has two IPs from the same province's planning level, two IPs from the same region's planning level, and one IP from the same carrier's planning level, scoring 77.3 points. Selection pool 6 has only five IPs, so no points are calculated. Selection pool 7 has seven IPs from the same province's planning level, scoring 76 points. Comparing all scores, selection pool 3 has the highest score, so the solution from selection pool 3 is selected as the optimal solution and deployed. It can be seen that the optimal solution takes into account both the distance of IP resources and the number of IPs based on the multi-factor weighting scoring method. Not only does it ensure that most of the bandwidth is allocated to high-traffic IPs within the same province, reducing the bandwidth load between provinces and improving the quality of network services, but it also reduces the number of IPs, resulting in less cache space and less internal bandwidth consumption, thus reducing costs.

[0063] An embodiment of the present invention also provides a server, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described cache node bandwidth scheduling method.

[0064] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for scheduling cache node bandwidth.

[0065] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0066] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A CDN coverage resource selection method based on multi-factor weighting scoring, characterized in that, Includes the following steps: Step S1: Obtain the planned bandwidth for generating the resolution group - area coverage resources; Step S2: Obtain the upstream IP resources of the resolution group-region; Step S3: Sort the IP resources of the resolution group-region according to the distance level; Step S4: Using the maximum number of IPs to be selected M as the boundary value, initialize and create IP selection pools from 1 to M. The number of IPs that each selection pool can hold is equal to the number of their respective sequence numbers. Step S5: Iterate through the sorted IPs with the highest traffic and determine whether the redundant bandwidth of each IP can be filled into the M selection pools. Step S6: When all IP resources at the resolution group-region granularity have been traversed, calculate the score for each selection pool. Step 5 specifically includes: Step S51: Round up the result of dividing the planned bandwidth of the resolution group-region by the redundant bandwidth of the IP. The resulting number N represents the number of IPs that can be placed into the selection pool of N and above. Step S52: Iterate through the selection pools from N onwards. If the number of IPs in the selection pool is less than the selection pool number, then put the IP into the selection pool; otherwise, do not put it into the selection pool.

2. The CDN coverage resource selection method based on multi-factor weighting scoring according to claim 1, characterized in that, The upstream IP resources include the distance level between the geographical location of the IP and the coverage area, and the IP redundancy bandwidth.

3. The CDN coverage resource selection method based on multi-factor weighting scoring according to claim 2, characterized in that, The distance levels are divided into: same province planning level, same region planning level, and same operator planning level.

4. The CDN coverage resource selection method based on multi-factor weighting scoring according to claim 1, characterized in that, The sorting order in step 3 is as follows: planning level of the same province > planning level of the same region > planning level of the same operator. Within the same level, the IP redundancy bandwidth is sorted in descending order, with the one with the larger redundancy bandwidth listed first.

5. The CDN coverage resource selection method based on multi-factor weighting scoring according to claim 1, characterized in that, The number of available IP addresses for traffic requests that a CDN domain name resolution server pushes to a customer is limited by the packet length.

6. The CDN coverage resource selection method based on multi-factor weighting scoring according to claim 1, characterized in that, Step 6 specifically includes: Step S61: Traverse each selection pool, determine the number of IPs in the selection pool, skip if the number of IPs is not equal to the selection pool number, otherwise calculate the score; Step S62: Calculate the score for all IPs in the selection pool using the following formula: In the above formula, x represents the number of IPs in the same planning level in the selection pool, y represents the number of IPs in the regional planning level in the selection pool, z represents the number of IPs in the operator planning level in the selection pool, S represents the number weight, X represents the weight of the same planning level, Y represents the weight of the regional planning level, and Z represents the weight of the operator planning level.

7. The CDN coverage resource selection method based on multi-factor weighting scoring according to claim 6, characterized in that, By controlling the range of values ​​for X, Y, Z, and S, the weights of IP distance levels and the number of IPs can be controlled.

8. An electronic device comprising a memory and a processor, characterized in that, include: At least one processor; And a memory communicatively connected to at least one processor; wherein the memory stores instructions executable by at least one processor, the instructions being executed by at least one processor to enable at least one processor to perform the CDN coverage resource selection method based on multi-factor weighting as described in any one of claims 1-7.

9. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the CDN coverage resource selection method based on multi-factor weighting scoring as described in any one of claims 1-7.

Images