Data request processing method, apparatus and device

Abstract

Embodiments of the present application provide a data request processing method, device and equipment, the method can comprise: obtaining a first data request sent by a first client; obtaining a network correspondence relationship between a plurality of network address intervals and a plurality of network types; determining a source network address of the first data request; determining a first network type corresponding to the first data request according to the network address interval where the source network address is located and the network correspondence relationship. The efficiency of processing data requests is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a data request processing method, apparatus and device. Background Technology

[0002] A Content Delivery Network (CDN) network can include multiple CDN nodes. CDN nodes can identify data requests sent by clients, determine the corresponding network type, and then select different network strategies for that data request.

[0003] In related technologies, artificial intelligence (AI) models can be set up in CDN nodes to determine the network type corresponding to a data request. However, in this approach, the AI ​​model determines the network type of the data request after the CDN node has received it, which prolongs the processing time of the CDN node and results in low efficiency in processing data requests. Summary of the Invention

[0004] This application provides a data request processing method, apparatus, and device to improve the efficiency of processing data requests.

[0005] In a first aspect, embodiments of this application provide a data request processing method, applied to a first CDN node in a Content Delivery Network (CDN) network, comprising:

[0006] Get the first data request sent by the first client;

[0007] Obtain the network mapping between multiple network address ranges and multiple network types;

[0008] Determine the source network address of the first data request;

[0009] Based on the network address range where the source network address is located and the network correspondence, the first network type corresponding to the first data request is determined.

[0010] In one possible implementation, determining the first network type corresponding to the first data request based on the network address range where the source network address is located and the network correspondence includes:

[0011] Determine the first network address range in which the source network address is located;

[0012] The first network type is determined based on the first network address range and the network correspondence, wherein the length of the first network address range is M, and M is an integer greater than 1.

[0013] In one possible implementation, determining the first network type based on the first network address range and the network correspondence includes:

[0014] If the network correspondence contains a network type corresponding to the first network address range, then the network type corresponding to the first network address range is determined as the first network type.

[0015] If the network type corresponding to the first network address range does not exist in the network correspondence, then the second network address range where the source network address is located is determined, and the network type corresponding to the second network address range is determined in the network correspondence. The network type corresponding to the second network address range is determined as the first network type. The length of the second network address range is N, where N is an integer greater than 1 and N is an integer less than M.

[0016] In one possible implementation, obtaining the network mapping relationship between multiple network address ranges and multiple network types includes:

[0017] The network correspondence is obtained from the preset storage space of the first CDN node. The network correspondence is obtained in advance and stored in the preset storage space.

[0018] In one possible implementation, before obtaining the network mapping relationship from the preset storage space of the first CDN node, the method further includes:

[0019] The network correspondence is received from a first device, which is either a first CDN node or an AI server in the CDN network, and the network correspondence is generated by the first device.

[0020] The network correspondence is stored in the preset storage space.

[0021] In one possible implementation, the method further includes:

[0022] Obtain multiple transfer logs;

[0023] Determine the multiple source network addresses corresponding to the multiple transmission logs;

[0024] The network types corresponding to the multiple transmission logs are obtained by processing the multiple transmission logs using an AI model;

[0025] The network mapping relationship is generated based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs.

[0026] In one possible implementation, the network mapping relationship is generated based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs, including:

[0027] Based on the N-bit network address intervals where the multiple source network addresses are located, multiple source network address sets are determined from the multiple source network addresses, wherein the N-bit network address intervals where the source network addresses in the multiple source network address sets are the same, and the length of the N-bit network address interval is N;

[0028] Based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs, determine the network type corresponding to each source network address;

[0029] The network mapping relationship is generated based on the multiple sets of source network addresses and the network type corresponding to each source network address in the set of source network addresses.

[0030] In one possible implementation, the network mapping relationship is generated based on the plurality of source network address sets and the network type corresponding to each source network address in the source network address sets, including:

[0031] For any set of source network addresses, a sub-correspondence relationship is determined based on the set of source network addresses and the network type corresponding to each source network address in the set of source network addresses. The sub-correspondence relationship includes at least one network address range and the network type corresponding to each network address range.

[0032] The network correspondence is determined based on the sub-correspondences corresponding to the multiple source network address sets.

[0033] In one possible implementation, determining the sub-correspondence relationship corresponding to the source network address set based on the source network address set and the network type corresponding to each source network address in the source network address set includes:

[0034] Based on the network type corresponding to each source network address in the source network address set, the proportion of each network type in the multiple network types is determined. The multiple network types include at least one of the following: Long Term Evolution (LTE) of general mobile communication technology, 5G of fifth generation mobile communication technology, WiFi of wireless network communication technology, and 4G of fourth generation mobile communication technology.

[0035] Based on the source network address set, the network type corresponding to each source network address in the source network address set, and the type proportion of each network type, the sub-correspondence relationship corresponding to the source network address set is determined.

[0036] In one possible implementation, the sub-correspondence relationship corresponding to the source network address set is determined based on the source network address set, the network type corresponding to each source network address in the source network address set, and the type proportion of each network type, including:

[0037] The target network type is determined from the various network types based on the type proportion of each network type, and the target network type has the largest type proportion.

[0038] Determine the initial network address range corresponding to the source network address set, wherein the length of the initial network address range is N;

[0039] If the type proportion of the target network type is greater than or equal to the first threshold, then the sub-correspondence is determined to include the correspondence between the initial network address range and the target network type;

[0040] If the proportion of the target network type is less than the first threshold, then multiple candidate source network addresses are determined in the source network address set, and multiple source network address subsets are determined in the multiple candidate source network addresses. The sub-correspondence is generated based on the initial network address range, the target network type, the multiple source network address subsets, and the network type corresponding to each source network address in the source network address subset.

[0041] Wherein, the network type corresponding to the candidate source network address is a network type other than the target network type, the M-bit network address intervals of each source network address in the source network address subset are the same, and the length of the M-bit network address interval is M.

[0042] In one possible implementation, the sub-correspondence is generated based on the initial network address range, the target network type, the plurality of source network address subsets, and the network type corresponding to each source network address in the source network address subset, including:

[0043] For any subset of source network addresses, determine the extended network address range corresponding to the subset of source network addresses, and determine the network type corresponding to the extended network address range based on the network type corresponding to each source network address in the subset of source network addresses.

[0044] Determining the sub-correspondence includes: the correspondence between the initial network address range and the target network type, and the correspondence between the extended network address range and the corresponding network type.

[0045] In one possible implementation, the method further includes:

[0046] Based on the first network type corresponding to the first data request, a network strategy is determined, and the network strategy includes at least one of the following: a congestion control strategy, a packet loss recovery strategy, or a sprint strategy.

[0047] The first data request is processed according to the network policy.

[0048] Secondly, embodiments of this application provide a data request processing apparatus applied to a first CDN node in a Content Delivery Network (CDN) network. The apparatus includes: a first acquisition module, a second acquisition module, a first determination module, and a second determination module, wherein...

[0049] The first acquisition module is used to acquire the first data request sent by the first client;

[0050] The second acquisition module is used to acquire the network correspondence between multiple network address ranges and multiple network types;

[0051] The first determining module is used to determine the source network address of the first data request;

[0052] The second determining module is used to determine the first network type corresponding to the first data request based on the network address range where the source network address is located and the network correspondence.

[0053] In one possible implementation, the second determining module is specifically used for:

[0054] Determine the first network address range in which the source network address is located;

[0055] The first network type is determined based on the first network address range, where the length of the first network address range is M, and M is an integer greater than 1.

[0056] In one possible implementation, the second determining module is specifically used for:

[0057] If the network correspondence contains a network type corresponding to the first network address range, then the network type corresponding to the first network address range is determined as the first network type.

[0058] If the network type corresponding to the first network address range does not exist in the network correspondence, then the second network address range where the source network address is located is determined, and the network type corresponding to the second network address range is determined in the network correspondence. The network type corresponding to the second network address range is determined as the first network type. The length of the second network address range is N, where N is an integer greater than 1 and N is an integer less than M.

[0059] In one possible implementation, the second acquisition module is specifically used for:

[0060] The network correspondence is obtained from the preset storage space of the first CDN node. The network correspondence is obtained in advance and stored in the preset storage space.

[0061] In one possible implementation, the device further includes a receiving module and a storage module, wherein,

[0062] The receiving module is used to receive the network correspondence sent by the first device, wherein the first device is a first CDN node or an artificial intelligence (AI) server in the CDN network, and the network correspondence is generated by the first device.

[0063] The storage module is used to store the network correspondence in the preset storage space.

[0064] In one possible implementation,

[0065] The first acquisition module is further configured to acquire multiple transmission logs;

[0066] The second determining module is further configured to determine multiple source network addresses corresponding to the multiple transmission logs;

[0067] In one possible implementation, the apparatus further includes a processing module and a generation module, wherein,

[0068] The processing module is used to process the multiple transmission logs using an AI model to obtain the network type corresponding to the multiple transmission logs;

[0069] The generation module is used to generate the network correspondence based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs.

[0070] In one possible implementation, the generation module is specifically used for:

[0071] Based on the N-bit network address intervals where the multiple source network addresses are located, multiple source network address sets are determined from the multiple source network addresses, wherein the N-bit network address intervals where the source network addresses in the multiple source network address sets are the same, and the length of the N-bit network address interval is N;

[0072] Based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs, determine the network type corresponding to each source network address;

[0073] The network mapping relationship is generated based on the multiple sets of source network addresses and the network type corresponding to each source network address in the set of source network addresses.

[0074] In one possible implementation, the generation module is specifically used for:

[0075] For any set of source network addresses, a sub-correspondence relationship is determined based on the set of source network addresses and the network type corresponding to each source network address in the set of source network addresses. The sub-correspondence relationship includes at least one network address range and the network type corresponding to each network address range.

[0076] The network correspondence is determined based on the sub-correspondences corresponding to the multiple source network address sets.

[0077] In one possible implementation, the generation module is specifically used for:

[0078] Based on the network type corresponding to each source network address in the source network address set, the proportion of each network type in the multiple network types is determined. The multiple network types include at least one of the following: Long Term Evolution (LTE) of general mobile communication technology, 5G of fifth generation mobile communication technology, WiFi of wireless network communication technology, and 4G of fourth generation mobile communication technology.

[0079] Based on the source network address set, the network type corresponding to each source network address in the source network address set, and the type proportion of each network type, the sub-correspondence relationship corresponding to the source network address set is determined.

[0080] In one possible implementation, the generation module is specifically used for:

[0081] The target network type is determined from the various network types based on the type proportion of each network type, and the target network type has the largest type proportion.

[0082] Determine the initial network address range corresponding to the source network address set, wherein the length of the initial network address range is N;

[0083] If the type proportion of the target network type is greater than or equal to the first threshold, then the sub-correspondence is determined to include the correspondence between the initial network address range and the target network type;

[0084] If the proportion of the target network type is less than the first threshold, then multiple candidate source network addresses are determined in the source network address set, and multiple source network address subsets are determined in the multiple candidate source network addresses. The sub-correspondence is generated based on the initial network address range, the target network type, the multiple source network address subsets, and the network type corresponding to each source network address in the source network address subset.

[0085] Wherein, the network type corresponding to the candidate source network address is a network type other than the target network type, the M-bit network address intervals of each source network address in the source network address subset are the same, and the length of the M-bit network address interval is M.

[0086] In one possible implementation, the generation module is specifically used for:

[0087] For any subset of source network addresses, determine the extended network address range corresponding to the subset of source network addresses, and determine the network type corresponding to the extended network address range based on the network type corresponding to each source network address in the subset of source network addresses.

[0088] Determining the sub-correspondence includes: the correspondence between the initial network address range and the target network type, and the correspondence between the extended network address range and the corresponding network type.

[0089] In one possible implementation,

[0090] The determining module is further configured to determine a network strategy based on the first network type corresponding to the first data request, wherein the network strategy includes at least one of the following: a congestion control strategy, a packet loss recovery strategy, or a sprint strategy.

[0091] The processing module is further configured to process the first data request according to the network policy.

[0092] Thirdly, embodiments of this application provide a CDN node, including: a memory and a processor;

[0093] The memory stores computer-executed instructions;

[0094] The processor executes computer execution instructions stored in the memory, causing the processor to perform the data request processing method according to any one of the first aspects.

[0095] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the data request processing method described in any of the first aspects.

[0096] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the data request processing method shown in any of the first aspects.

[0097] This application provides a data request processing method, apparatus, and device, which can acquire a first data request sent by a first client and obtain network correspondences between multiple network address ranges and multiple network types. The source network address of the first data request can be determined, and the first network type corresponding to the first data request can be determined based on the network address range and network correspondences of the source network address. Since the network correspondences are pre-determined based on multiple transmission logs, when a CDN node receives a data request, it can determine the network address range corresponding to the data request and the network type corresponding to the data request based on the network correspondences. This eliminates the need to calculate the network type corresponding to the data request using an AI model, thus separating the AI ​​calculation process from the data request processing process, preventing the AI ​​calculation process from affecting the data request processing process, and improving the efficiency of data request processing. Attached Figure Description

[0098] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0099] Figure 1 A schematic diagram illustrating an application scenario provided for an exemplary embodiment of this application;

[0100] Figure 2 A flowchart illustrating a data request processing method provided for an exemplary embodiment of this application;

[0101] Figure 3 A flowchart illustrating another data request processing method provided for an exemplary embodiment of this application;

[0102] Figure 4 A schematic diagram illustrating a data request processing method provided for an exemplary embodiment of this application;

[0103] Figure 5 A schematic diagram of the structure of a data request processing apparatus provided for an exemplary embodiment of this application;

[0104] Figure 6 A schematic diagram of another data request processing apparatus provided as an exemplary embodiment of this application;

[0105] Figure 7This is a schematic diagram of the structure of a CDN node provided as an exemplary embodiment of this application. Detailed Implementation

[0106] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0107] 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 in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0108] Figure 1 This is a schematic diagram illustrating an application scenario provided for an exemplary embodiment of this application. Please refer to [link / reference]. Figure 1 This includes multiple clients and CDN nodes. The clients can be designated as Client 1, Client 2, ..., Client n, where n is an integer greater than or equal to 1. Any client can communicate with any CDN node.

[0109] For any client, a client can send a data request to a CDN node.

[0110] After receiving a data request, a CDN node can determine the network type corresponding to the data request. Based on the network type, it can determine the network policy corresponding to the data request and process the data request according to the corresponding network policy.

[0111] For example, client 1 can send data request 1 to a CDN node. After receiving data request 1, the CDN node can determine the network type corresponding to data request 1. Assuming network type 1 is 4G, it can determine the network policy corresponding to 4G type and process the data request according to the network policy.

[0112] In related technologies, AI models can be set up in CDN nodes to determine the network type corresponding to a data request. However, in the above method, the AI ​​model determines the network type of the data request after the CDN node has obtained the data request, which prolongs the processing time of the CDN node and results in low efficiency in processing data requests.

[0113] In this embodiment, the network correspondence between multiple network address ranges and multiple network types can be determined in advance based on multiple transmission logs in the CDN network. The CDN nodes can store these network correspondences. When a CDN node receives a data request, it can determine the network address range corresponding to the data request, and then determine the network type corresponding to the data request based on the network correspondence. This eliminates the need to calculate the network type of the data request using an AI model, thus separating the AI ​​calculation process from the data request processing process, preventing the AI ​​calculation process from affecting the data request processing process, and improving the efficiency of data request processing.

[0114] The technical solutions shown in this application will now be described in detail through specific embodiments. It should be noted that the following embodiments may exist independently or in combination with each other; for identical or similar content, the description will not be repeated in different embodiments.

[0115] Figure 2 This is a flowchart illustrating a data request processing method provided for an exemplary embodiment of this application. Please refer to [link / reference]. Figure 2 The method may include:

[0116] S201. Obtain the first data request sent by the first client.

[0117] The execution entity in this application embodiment can be a first CDN node or a data request processing device set in the first CDN node. The data request processing device can be implemented by software or by a combination of software and hardware. The data request processing device can be a processor in the first CDN node. For ease of understanding, the following description will use the first CDN node as the execution entity.

[0118] A CDN network may include multiple CDN nodes, among which a first CDN node may be included.

[0119] For the first CDN node, there can be multiple clients. The first client can be any one of these multiple clients.

[0120] The first client and the first CDN node can communicate with each other. The first client can send a first data request to the first CDN node, and the first CDN node can retrieve the first data request sent by the first client.

[0121] For example, if the multiple clients corresponding to the first CDN node are client 1, client 2, ..., client 10, and if the first client is client 2, then client 2 can send a first data request to the first CDN node, and the first CDN node can obtain the first data request sent by client 2.

[0122] S202. Obtain the network correspondence between multiple network address ranges and multiple network types.

[0123] The network address ranges may include a first network address range and / or a second network address range. The length of the first network address range can be M, and the length of the second network address range can be N. Here, N is an integer greater than 1, and N is an integer less than M.

[0124] For example, if M is 24 and the length of the first network address range is 24, then the first network address range can be a 24-bit network address range, and the first network address range can be 192.167.1; if N is 16 and the length of the second network address range is 16, then the second network address range can be a 16-bit network address range, and the second network address range can be 192.167.

[0125] Multiple network types may include at least one of the following: Long Term Evolution (LTE), 5th Generation Mobile Communication Technology (5G), Wireless Fidelity (WiFi), and 4th Generation Mobile Communication Technology (4G).

[0126] Network mappings can be determined based on multiple transmission logs within the CDN network. For example, network mappings can be shown in Table 1:

[0127] Table 1

[0128] Network address range Network type 192.167 5G 192.168 LTE 192.169 WiFi 192.169.1 4G 192.169.5 4G …… ……

[0129] As shown in Table 1, 192.167, 192.168, and 192.169 can be the second network address range. 192.169.1 and 192.169.5 are the first network address range.

[0130] Specifically, the network type corresponding to the second network address range 192.167 can be 5G, and the network type corresponding to the second network address range 192.168 can be LTE. Since the second network address range 192.169 includes the first network address ranges 192.169.1 and 192.169.5, where the network type corresponding to the first network address range 192.169.1 and 192.169.5 can both be 4G, therefore, in the second network address range 192.169, except for the first network address ranges 192.169.1 and 192.169.5, the network type corresponding to the other first network address ranges is WiFi. For example, in the second network address range 192.169, the first network address ranges such as 192.169.2, 192.169.3, 192.169.4, 192.169.6, ..., 192.169.255 all correspond to WiFi as the network type.

[0131] Optionally, the first CDN node can obtain the network mapping relationship from the preset storage space.

[0132] S203. Determine the source network address of the first data request.

[0133] Optionally, the source network address corresponding to the first data request can be determined in the following ways: the corresponding transmission log can be determined based on the first data request; the network address of the first client can be determined from the transmission log; and the network address of the first client can be determined as the source network address of the first data request.

[0134] Transmission logs can be used for Transmission Control Protocol (TCP) transmission logs.

[0135] After receiving the first data request from the first client, the first CDN node can generate a transmission log corresponding to the first data request. The transmission log may include the network address of the first client that sent the first data request.

[0136] For example, if the first client is client 2, and client 2's network address is 192.169.1.1, after the first CDN node receives the first data request 1 sent by client 2, it can generate a corresponding transmission log 1 based on the first data request 1. The transmission log 1 may include client 2's network address 192.169.1.1. The first CDN node can determine client 2's network address from the transmission log 1 and identify client 2's network address as the source network address corresponding to the first data request 1, that is, the source network address corresponding to the first data request 1 is 192.169.1.1.

[0137] S204. Determine the first network type corresponding to the first data request based on the network address range and network correspondence of the source network address.

[0138] In an optional embodiment, the first network type corresponding to the first data request can be determined by: determining the first network address range where the source network address is located; and determining the first network type based on the first network address range.

[0139] After determining the source network address corresponding to the first data request, the first network address range in which the source network address is located can be determined, and then the first network type can be determined based on the first network address range in which the source network address is located.

[0140] For example, if the source network address corresponding to the first data request is 192.169.1.1, then the first network address range where the source network address is located can be determined to be 192.169.1, and the first network type can be determined based on the first network address range.

[0141] Optionally, determining the first network type can include the following two cases:

[0142] Case 1: The network correspondence contains a network type corresponding to the first network address range.

[0143] In this case, the network type corresponding to the first network address range can be determined as the first network type.

[0144] For example, if the first network address range corresponding to the first data request is 192.169.1, and if the network correspondence is as shown in Table 1, then since the network type corresponding to 192.169.1 in the network correspondence is 4G, it can be determined that the first network type is 4G.

[0145] Case 2: The network type corresponding to the first network address range does not exist in the network correspondence.

[0146] In this case, the second network address range where the network address is located can be determined, and the network type corresponding to the second network address range can be determined in the network correspondence, and the network type corresponding to the second network address range can be determined as the first network type.

[0147] For example, if the source network address corresponding to the first data request is 192.169.2.1, and the first network address range it belongs to is 192.169.2, and if the network correspondence is as shown in Table 1, then since there is no network type corresponding to the first network address range 192.169.2 in the network correspondence, it can be determined that the second network address range it belongs to is 192.169. Since there is a network type corresponding to the second network 192.169 in the network correspondence, which is WiFi, it can be determined that the first network type is WiFi.

[0148] In this embodiment, the first CDN node can obtain a first data request sent by a first client and acquire the network correspondence between multiple network address ranges and multiple network types. The first CDN node determines the source network address of the first data request and can determine the first network type corresponding to the first data request based on the network address range and network correspondence of the source network address. Since the network correspondence is determined in advance based on multiple transmission logs, when the first CDN node receives a data request, it can determine the network address range corresponding to the data request and determine the network type corresponding to the data request based on the network correspondence. This eliminates the need to calculate the network type corresponding to the data request through an AI model, thus separating the AI ​​calculation process from the data request processing process, preventing the AI ​​calculation process from affecting the data request processing process, and improving the efficiency of data request processing.

[0149] Below, in Figure 2 Based on the illustrated embodiments, combined with Figure 3 The above data request processing method will be further explained in detail.

[0150] Figure 3 A flowchart illustrating another data request processing method provided for an exemplary embodiment of this application. Please refer to... Figure 3 The method may include:

[0151] S301, Obtain multiple transmission logs.

[0152] Optionally, the first CDN node can obtain multiple transmission logs corresponding to multiple data requests in a preset storage space.

[0153] For example, the first CDN node can obtain the corresponding transmission logs for 100 data requests, i.e., 100 transmission logs.

[0154] S302. Determine the multiple source network addresses corresponding to multiple transmission logs.

[0155] For any given transmission log, the transmission log may include the network address of the client that sent the data request. The first CDN node can use the client's network address to determine the source network address corresponding to the transmission log.

[0156] For example, if transmission log 1 is the log corresponding to data request 1, and data request 1 is sent by client 1 to the first CDN node, then transmission log 1 may include network address 1 of client 1, and network address 1 can be determined as the source network address corresponding to transmission log 1.

[0157] The first CDN node can identify multiple source network addresses from various transmission logs.

[0158] For example, the multiple source network addresses corresponding to multiple transmission logs can be shown in Table 2:

[0159] Table 2

[0160] Transmission log Source network address Transmission Log 1 192.167.1.1 Transmission Log 2 192.169.2.1 Transmission Log 3 192.168.1.1 Transmission Log 4 192.169.5.1 Transmission Log 5 192.169.1.1 …… ……

[0161] S303. Process multiple transmission logs using an AI model to obtain the network types corresponding to the multiple transmission logs.

[0162] The transmission log may also include the TCP announcement window, connection establishment time, and maximum segment size (MSS).

[0163] The TCP advertised window refers to the amount of data that the first CDN node can receive. For example, the TCP advertised window can be 4095 bytes.

[0164] Connection establishment time refers to the time it takes for the first client to establish a connection with the first CDN node. For example, the connection establishment time between the first client and the first CDN node can be 80 milliseconds (ms).

[0165] The Maximum Segment Size (MSS) is the maximum data length that each segment can carry during communication, negotiated by the sender and receiver when a TCP connection is established. For example, the MSS can be 1370 bytes.

[0166] In one alternative embodiment, the TCP announcement window, connection establishment duration, and MSS can be determined from the transmission log using an AI model, and then the network type corresponding to the transmission log can be determined based on the TCP announcement window, connection establishment duration, and MSS.

[0167] For example, if the AI ​​model determines from transmission log 1 that the TCP advertised window is 4095 bytes, the connection establishment time is 80ms, and the MSS is 1370 bytes, then it can determine the network type 1 corresponding to transmission log 1 based on the TCP advertised window, connection establishment time, and MSS. Assume that the network type 1 corresponding to transmission log 1 is 4G.

[0168] Optionally, multiple transmission logs can be processed using an AI model to obtain the network types corresponding to each transmission log.

[0169] For example, if multiple transmission logs are as shown in Table 2, then the network types corresponding to these multiple transmission logs can be as shown in Table 3:

[0170] Table 3

[0171]

[0172]

[0173] S304. Generate network correspondence based on the multiple source network addresses corresponding to multiple transmission logs and the network types corresponding to multiple transmission logs.

[0174] In an optional embodiment, network correspondence can be generated as follows: based on the N-bit network address range where multiple source network addresses are located, determine multiple sets of source network addresses from among the multiple source network addresses; based on the multiple source network addresses corresponding to multiple transmission logs and the network types corresponding to the multiple transmission logs, determine the network type corresponding to each source network address; and generate network correspondence based on the multiple sets of source network addresses and the network types corresponding to each source network address in the network address set.

[0175] The N-bit network address ranges containing all source network addresses in the source network address set can be the same, and the length of the N-bit network address range is N.

[0176] For example, if N is 16, the source network address set 1 can include 300 source network addresses. These 300 source network addresses can be 192.167.0.1, 192.167.1.2, 192.167.2.1, 192.167.2.3, 192.167.2.3, ... etc., and the 16-bit network address range containing these 300 source network addresses is 192.167, with 192.167 having a length of 16 bits.

[0177] Optionally, the first CDN node can determine multiple source network addresses and, based on the N-bit network address range where the multiple source network addresses are located, determine multiple source network address sets among the multiple source network addresses.

[0178] For example, the first CDN node can identify 2000 source network addresses, and from these 2000 source network addresses, it can identify 500 source network address sets, which can be source network address set 1, source network address set 2, ..., source network address set 500.

[0179] Since multiple source network addresses corresponding to multiple transmission logs are determined in step S302, and the network types corresponding to multiple transmission logs are determined in step S303, the network type corresponding to each source network address can be determined based on the multiple source network addresses and the network types corresponding to multiple transmission logs.

[0180] For example, if multiple source network addresses corresponding to multiple transmission logs are shown in Table 2, and multiple network types corresponding to multiple transmission logs are shown in Table 3, then the network type corresponding to each source network address can be determined, as shown in Table 4:

[0181] Table 4

[0182] Transmission log Source network address Network type Transmission Log 1 192.167.1.1 5G Transmission Log 2 192.169.2.1 WiFi Transmission Log 3 192.168.1.1 LTE Transmission Log 4 192.169.5.1 4G Transmission Log 5 192.169.1.1 4G …… …… ……

[0183] Optionally, for any set of source network addresses, the sub-correspondence relationship corresponding to the set of source network addresses can be determined based on the set of source network addresses and the network type corresponding to each source network address in the set of source network addresses. Then, the network correspondence relationship can be determined based on the sub-correspondence relationship corresponding to multiple sets of source network addresses.

[0184] Since any set of source network addresses can correspond to an N-bit network address range, the sub-correspondences of the source network address sets include at least one network address range and the network type corresponding to each network address range.

[0185] In an optional embodiment, the sub-correspondence relationship corresponding to the source network address set can be determined in the following way: based on the network type corresponding to each source network address in the source network address set, determine the type proportion of each network type among multiple network types; based on the source network address set, the network type corresponding to each source network address in the source network address set, and the type proportion of each network type, determine the sub-correspondence relationship corresponding to the source network address set.

[0186] For example, if source network address set 1 includes 300 source network addresses, of which 290 source network addresses correspond to 5G network type, 6 source network addresses correspond to 4G network type, and 4 source network addresses correspond to WiFi network type, then it can be determined that among the 300 network types corresponding to these 300 source network addresses, the 5G type accounts for approximately 97%, the 4G type accounts for approximately 2%, and the WiFi type accounts for approximately 1%.

[0187] After determining the proportion of each network type, the target network type can be determined from multiple network types based on the proportion of each network type, and the initial network address range corresponding to the source network address set can be determined.

[0188] The target network type can have the highest percentage. For example, if the source network address set 1 includes 300 source network addresses, and among the 300 network types corresponding to these 300 source network addresses, the 5G type accounts for approximately 97%, the 4G type accounts for approximately 2%, and the WiFi type accounts for approximately 1%, then the target network type can be determined to be 5G.

[0189] The length of the initial network address range can be N. For example, if N is 16, and the source network address set 1 contains 300 source network addresses, and the 16-bit network address range of these 300 source network addresses is 192.167, then the initial network address range corresponding to the source network address set 1 can be determined to be 192.167, and the length of the initial network address range can be 16.

[0190] Optionally, determining the sub-correspondence based on the target network type and the initial network can include the following two cases:

[0191] Case 1: The proportion of the target network type is greater than or equal to the first threshold.

[0192] In this case, the sub-correspondence can be determined to include the correspondence between the initial network address range and the target network type.

[0193] The first threshold can be preset by a person. For example, the first threshold could be 95%.

[0194] For example, if the target network type corresponding to the source network address set 1 is 5G, and the 5G type accounts for approximately 97%, if the first threshold is 95%, and if the initial network address range corresponding to the source network address set 1 is 192.167, then since the target network type accounts for 97%, which is greater than the first threshold of 95%, the sub-correspondence can be determined as follows: the initial network 192.167 corresponds to the network type 5G.

[0195] Case 2: The proportion of the target network type is less than the first threshold.

[0196] In this case, multiple candidate source network addresses can be determined from the source network address set, and multiple source network address subsets can be determined from the multiple candidate source network addresses. Then, a sub-correspondence relationship can be generated based on the initial network address range, the target network type, the multiple source network address subsets, and the network type corresponding to each source network address in the source network address subset.

[0197] The network type corresponding to the candidate source network address can be any network type other than the target network type.

[0198] The source network addresses in the subset of source network addresses are all located in the same M-bit network address interval, and the length of the M-bit network address interval is M.

[0199] For example, if M is 24, and the subset of source network addresses includes 20 source network addresses, these 20 source network addresses can be 192.169.1.1, 192.169.1.2, 192.169.1.4, 192.169.1.8, 192.169.1.25, ... etc., then the 24-bit network address range containing these 20 source network addresses is 192.169.1, and the length of 192.169.1 is 24 bits.

[0200] For example, if the initial network address range corresponding to the source network address set 2 is 192.169, and the source network address set 2 includes 280 source network addresses, among the 280 network types corresponding to these 280 source network addresses, WiFi accounts for approximately 80%, 4G accounts for approximately 9%, and 5G accounts for approximately 11%. Since WiFi accounts for the largest proportion, the target network type can be determined to be WiFi. Therefore, the sub-correspondence can be determined to include WiFi corresponding to the initial network address range 192.169.

[0201] If the first threshold is 95%, and since the target network type (WiFi) accounts for 80%, which is less than the first threshold of 95%, then in the source network address set 2, the source network addresses corresponding to network types 4G and 5G can be determined as candidate source network addresses. Therefore, there are 280 * (9% + 11%) = 56 candidate source network addresses. Of these 56 candidate source network addresses, 31 correspond to the 5G network type, and 25 correspond to the 4G network type. Assuming that two source network address subsets can be determined from these 56 candidate source network addresses, they can be designated as source network address subset 1 and source network address subset 2.

[0202] Optionally, for any subset of source network addresses, the extended network address range corresponding to the subset of source network addresses can be determined, and the network type corresponding to the extended network address range can be determined based on the network type corresponding to each source network address in the subset of source network addresses. Thus, the sub-correspondence can be determined, including: the correspondence between the initial network address range and the target network type, and the correspondence between the extended network address range and the corresponding network type.

[0203] The length of the extended network address range can be M. For example, if M is 24, and the source network address subset 1 contains 20 source network addresses, and the 24-bit network address range for these 20 source network addresses is 192.169.1, then the extended network address range corresponding to the source network address subset 1 can be determined to be 192.169.1, and the length of 192.169.1 is 24 bits.

[0204] Optionally, the proportion of each network type can be determined based on the network type corresponding to each source network address in the source network address subset, and the network type with the largest proportion can be determined as the network type corresponding to the extended network address range.

[0205] For example, if the extended network address range corresponding to source network address subset 1 is 192.169.1, and source network address subset 1 includes 20 source network addresses, of which 16 source network addresses correspond to the 4G network type and 4 source network addresses correspond to the 5G network type, then it can be determined that among the 20 network types corresponding to these 20 source network addresses, the 4G type accounts for 80% and the 5G type accounts for 20%. Since the 4G type accounts for the largest proportion, it can be determined that the network type corresponding to the extended network address range 192.169.1 is 4G.

[0206] If the initial network address range corresponding to source network address set 2 is 192.169, and the target network type is WiFi, then there are two source network address subsets in source network address set 2. Source network address subset 1 corresponds to the extended network address range of 192.169.1, which corresponds to the network type 4G; and source network address subset 2 corresponds to the extended network address range of 192.169.5, which also corresponds to the network type 4G. Therefore, the sub-correspondence relationships are: the initial network address range 192.169 corresponds to the network type WiFi; the extended network address range 192.169.1 corresponds to the network type 4G; and the extended network address range 192.169.5 corresponds to the network type 4G.

[0207] Optionally, after determining the sub-correspondences between multiple source network address sets, the network correspondences can be determined. The network correspondences include the sub-correspondences between multiple source network address sets.

[0208] For example, if the sub-correspondences corresponding to source network address set 1 are: network address range 192.167 corresponds to network type 5G; the sub-correspondences corresponding to source network address set 2 are: network address range 192.168 corresponds to network type LTE; the sub-correspondences corresponding to source network address set 3 are: network address range 192.169 corresponds to network type WiFi; the sub-correspondences corresponding to source network address subset 1 are: network address range 192.169.1 corresponds to network type 4G; the sub-correspondences corresponding to source network address subset 2 are: network address range 192.169.5 corresponds to network type 4G; ..., then the network correspondences can be determined based on the sub-correspondences corresponding to these multiple source network address sets, as shown in Table 1.

[0209] It should be noted that steps S301 to S304 constitute the process of generating network mapping relationships, which can be generated in the first device. The first device can be either the first CDN node or the AI ​​server. The AI ​​server can be a server other than the multiple CDN nodes in the CDN network.

[0210] Network correspondences can be generated periodically. For example, network correspondences can be generated every 30 minutes (min).

[0211] S305. Store the network correspondence in the preset storage space.

[0212] If the first device is the first CDN node, then after the first CDN node generates the network mapping relationship, it can pre-store the network mapping relationship in the preset storage space.

[0213] If the first device is an AI server different from the first CDN node, it can periodically send network correspondence to the first CDN node so that the first CDN node can receive the network correspondence sent by the first device and pre-store the network correspondence in a preset storage space.

[0214] Since the network mapping can be generated through the first device and synchronized to any CDN node, there is no need to set up an AI model for AI calculation in each CDN node, which greatly reduces the computing resources occupied and reduces computing costs.

[0215] S306. Obtain the first data request sent by the first client.

[0216] When a customer has a need, the first client can send a first data request to the first CDN node, and the first CDN node can obtain the first data request sent by the first client.

[0217] S307. Obtain the network correspondence from the preset storage space of the first CDN node.

[0218] For example, if the preset storage space contains network correspondence 1, then the first CDN node can obtain network correspondence 1 from the preset storage space.

[0219] S308. Determine the source network address of the first data request.

[0220] It should be noted that the specific execution process of step S308 can be found in the specific execution process of step S203, and will not be repeated here.

[0221] S309. Determine the first network type corresponding to the first data request based on the network address range and network correspondence of the source network address.

[0222] It should be noted that the specific execution process of step S309 can be found in the specific execution process of step S204, and will not be repeated here.

[0223] S310. Determine the network policy based on the first network type corresponding to the first data request.

[0224] Network strategies may include at least one of the following: congestion control strategies, packet loss recovery strategies, or sprint strategies.

[0225] For example, if the first network type corresponding to the first data request is 4G, the first CDN node can determine the corresponding network policy based on the first network type 4G.

[0226] S311. Process the first data request according to the network policy.

[0227] For example, if the first network type corresponding to the first data request is 4G, after the first CDN node determines the network policy corresponding to the first network type 4G, it can process the first data request according to the corresponding network policy.

[0228] In this embodiment, multiple transmission logs can be acquired, and multiple source network addresses corresponding to these logs can be determined. The transmission logs can be processed using an AI model to obtain the network type corresponding to each log. Then, based on the multiple source network addresses and network types corresponding to the multiple transmission logs, a network mapping relationship can be generated, and this relationship can be stored in a pre-defined storage space. The first CDN node can acquire a first data request sent by a first client and retrieve the network mapping relationship from the pre-defined storage space. The first CDN node can determine the first network type corresponding to the first data request based on the network address range and network mapping relationship, and determine a network policy based on the first network type. Then, the first data request can be processed according to the network policy. Since the first device can periodically generate network mapping relationships in advance, the CDN node can pre-acquire these relationships and store them in the pre-defined storage space. Once the first CDN node receives a data request, it can determine the network address range corresponding to the data request and determine the network type corresponding to the data request based on the network correspondence. This eliminates the need to calculate the network type corresponding to the data request through an AI model, thus separating the AI ​​calculation process from the data request processing process. This avoids the AI ​​calculation process affecting the data request processing process and improves the efficiency of processing data requests.

[0229] Below, based on any of the above embodiments, specific examples will be used to illustrate... Figure 4 The above data request processing procedure will be explained in further detail.

[0230] Figure 4 This is a schematic diagram illustrating a data request processing method provided for an exemplary embodiment of this application. Please refer to [link / reference]. Figure 4 This includes model training equipment, a CDN network, and multiple clients. The CDN network can include multiple CDN nodes, which can be CDN node 1, CDN node 2, ..., CDN node k, where k is an integer greater than or equal to 1. The multiple clients can be client 1, client 2, ..., client m, where m is an integer greater than or equal to 1.

[0231] The model training device can communicate with the first CDN node. For example, the first CDN node can be CDN node 1, and the model training device can communicate with CDN node 1.

[0232] The first CDN node can have multiple corresponding clients. Each client can communicate with the first CDN node. The client can send a first data request to the first CDN node. When a client sends a first data request, it generates a client log corresponding to that request. Upon receiving the first data request, the first CDN node generates a transmission log corresponding to that request.

[0233] Model training devices can be devices outside of the CDN network and can be used to train AI models. For example, model training devices can be servers or computing devices.

[0234] Before training the AI ​​model, multiple client logs can be sent to the model training device from multiple clients, allowing the device to receive these logs. The client logs can include the timestamp of the first data request, the client's network address and port number, the target CDN node's network address and port number, and the network type.

[0235] Multiple transmission logs can be sent to the model training device via multiple CDN nodes, allowing the model training device to receive multiple transmission logs. The transmission logs may include the timestamp of receiving the first data request, the client's network address and port number, and the network address and port number of the CDN node.

[0236] After receiving multiple client logs and multiple transmission logs, the model training device can determine the timestamp, client network address and port number, target CDN node network address and port number in the client logs, and the timestamp, client network address and port number, CDN node network address and port number in the transmission logs, and perform correlation processing on the client logs and transmission logs.

[0237] Optionally, the correlation processing of client logs and transmission logs can include the following two cases:

[0238] Scenario 1: The timestamp, client network address and port number, target CDN node network address and port number in the client log, and the timestamp, client network address and port number, and CDN node network address and port number in the transmission log are consistent.

[0239] In this case, client logs and transmission logs can be correlated.

[0240] For example, if the model training device includes Client Log 1 and Transmission Log 1, and if the model training device determines that Client Log 1 contains a timestamp of 1679386411, a client network address of 192.168.1.1, a port number of 3000, and a target CDN node network address of 192.168.2.6, a port number of 2080; and if the model training device determines that Transmission Log 1 contains a timestamp of 1679386411, a client network address of 192.168.1.1, a port number of 3000, and a CDN node network address of 192.168.2.6, a port number of 2080, then since the timestamp, client network address, port number, and CDN node network address and port number are consistent in Client Log 1 and Transmission Log 1, it can be determined that Client Log 1 and Transmission Log 1 are logs corresponding to the same first data request, and therefore Client Log 1 and Transmission Log 1 can be associated.

[0241] Scenario 2: The timestamp, client network address and port number, target CDN node network address and port number in the client log, and the timestamp, client network address and port number, and CDN node network address and port number in the transmission log are inconsistent.

[0242] In this case, it is not necessary to correlate the client logs and the transmission logs.

[0243] For example, if the model training device includes Client Log 1 and Transmission Log 1, and if the model training device determines that Client Log 1 contains a timestamp of 1679386309, a client network address of 192.168.1.1, and a port number of 3000, and the target CDN node network address of 192.168.2.6 and a port number of 2080; and if the model training device determines that Transmission Log 1 contains a timestamp of 1679386309, a client network address of 192.168.2.1, and a port number of 2095, and the CDN node network address of 192.168.2.6 and a port number of 2080, then since the client network address and port number in Client Log 1 and Transmission Log 1 are inconsistent, it can be determined that Client Log 1 and Transmission Log 1 are logs corresponding to different first data requests, and therefore it is not necessary to associate Client Log 1 and Transmission Log 1.

[0244] Optionally, multiple client logs and multiple transport logs are correlated to determine the transport log associated with each client log. Since the client logs may include the network type corresponding to the first data request, the network type can be determined in the client, and thus the network type corresponding to the transport log can be correlated.

[0245] For example, if client log 1 is associated with transmission log 1, and if it is determined in client log 1 that the network type corresponding to the first data request is 4G, then the network type corresponding to transmission log 1 can be associated with 4G.

[0246] Optionally, after associating the transmission logs with the corresponding network type, multiple transmission logs associated with the network type can be used as training data to train the AI ​​model.

[0247] Since the transmission log can also include TCP's advertised window, connection establishment duration, and MSS, the AI ​​model can identify and process the transmission log during the training process. That is, it can determine the TCP advertised window, connection establishment duration, and MSS from the transmission log, and then determine the network type corresponding to the TCP advertised window, connection establishment duration, and MSS based on the network type corresponding to the transmission log.

[0248] For example, if the network type corresponding to transmission log 1 is 4G, and if the AI ​​model determines in transmission log 1 that the TCP advertised window is 4095 bytes, the connection establishment time is 80ms, and the MSS is 1370 bytes, then since the network type corresponding to transmission log 1 is 4G, the AI ​​model can determine that when the TCP advertised window is 4095 bytes, the connection establishment time is 80ms, and the MSS is 1370 bytes, the corresponding network type is 4G.

[0249] The AI ​​model can be trained several times so that it can accurately determine the corresponding network type based on the TCP announcement window, connection establishment duration, and MSS in the transmission log.

[0250] After the model training device obtains the AI ​​model, it can save the AI ​​model in the preset storage space.

[0251] In step ①, the model training device can periodically send AI models to the first device. The first device can be the first CDN node in the CDN network. For example, the first CDN node can be CDN node 1, and the model training device can send AI models to CDN node 1. The update cycle of the AI ​​model can be one year.

[0252] It should be noted that a CDN network can include a first CDN node and multiple second CDN nodes, only requiring the AI ​​model to be configured on the first CDN node. For example, if the CDN network includes k CDN nodes, the first CDN node can be CDN node 1, the second CDN nodes can be CDN node 2, ..., CDN node k, then the AI ​​model can be configured on CDN node 1, but not on CDN nodes 2, ..., CDN node k.

[0253] In step ②, multiple second CDN nodes can send multiple transmission logs to the first CDN node in real time, so that the first CDN node can obtain multiple transmission logs. These multiple transmission logs are generated when multiple second CDN nodes receive data requests. For example, CDN node 2, ..., CDN node k can each send multiple transmission logs to CDN node 1.

[0254] In step ③, for any transmission log, the first CDN node can identify and process the transmission log through an AI model. That is, it can determine the TCP announcement window, connection establishment duration and MSS in the transmission log, and then determine the network type corresponding to the transmission log based on the TCP announcement window, connection establishment duration and MSS.

[0255] For example, if the AI ​​model determines from transmission log 1 that the TCP advertised window is 4095 bytes, the connection establishment time is 80ms, and the MSS is 1370 bytes, then it can determine the network type 1 corresponding to transmission log 1 based on the TCP advertised window, connection establishment time, and MSS. Assume that the network type 1 corresponding to transmission log 1 is 4G.

[0256] Optionally, multiple transmission logs can be processed using an AI model to obtain the network types corresponding to each transmission log. These multiple transmission logs may include transmission logs sent by multiple second CDN nodes, as well as transmission logs generated by the first CDN node based on data requests. For example, the network types corresponding to the multiple transmission logs may be as shown in Table 3.

[0257] For any given transport log, the first CDN node can determine the source network address corresponding to the transport log by using the client's network address. For example, multiple source network addresses corresponding to multiple transport logs can be shown in Table 2.

[0258] The first CDN node can generate network mapping relationships using an AI model, based on multiple source network addresses corresponding to multiple transmission logs and the network types corresponding to multiple transmission logs. The execution process for generating network mapping relationships can be found in step S304, and will not be repeated here. For example, the network mapping relationships can be shown in Table 1. The first CDN node can store the network mapping relationships in a preset storage space.

[0259] In step ④, the first CDN node can periodically send the network mapping relationship to multiple second CDN nodes, so that the multiple second CDN nodes can receive the network mapping relationship and store it in a preset storage space. For example, CDN node 1 can send the network mapping relationship to CDN node 2, ..., CDN node k every 30 minutes, and CDN node 2, ..., CDN node k can store the network mapping relationship.

[0260] Since network mappings can be stored in multiple secondary CDN nodes, there is no need to deploy AI models in multiple secondary CDN nodes, which greatly reduces the computing resources required and lowers computing costs.

[0261] In step ⑤, when a customer has a request, the first client can send a first data request to the first CDN node, and the first CDN node can obtain the first data request sent by the first client. For example, the first client can be client 2, and client 2 can send a first data request to CDN node 1.

[0262] In step ⑥, after the first CDN node obtains the first data request, it can retrieve the network mapping relationship from the preset storage space. Based on the network address range and network mapping relationship corresponding to the first data request, it determines the first network type corresponding to the first data request. Then, based on the first network type, it determines the network policy and processes the first data request accordingly. For example, if the first CDN node determines that the first network type corresponding to the first data request is 4G based on the network mapping relationship, it can determine the corresponding network policy based on the 4G network type and process the first data request accordingly.

[0263] It should be noted that since the second CDN node also stores network correspondences, similarly, after the second CDN node obtains the first data request, it can obtain the network correspondences in the preset storage space, and determine the first network type corresponding to the first data request based on the network address range and network correspondences corresponding to the first data request. Then, it can determine the network strategy based on the first network type corresponding to the first data request, and process the first data request according to the network strategy.

[0264] The first or second CDN node can determine the network type corresponding to the data request based on the network correspondence. This enables the determination of the network type corresponding to each first data request with a very small amount of computing resources (compared to 1 / 10 to 1 / 20 of existing technologies), with an accuracy of over 90%. Furthermore, since the network correspondence is predetermined, it will not affect the data request processing process, resulting in very low task intrusion.

[0265] In this embodiment, the model training device can acquire multiple client logs and multiple transmission logs, and can perform correlation processing on the multiple client logs and multiple transmission logs to determine the network types associated with each of the multiple transmission logs. The multiple transmission logs associated with network types are then used as training data to train the AI ​​model. After obtaining the AI ​​model, it can be sent to the first CDN node. The first CDN node can acquire multiple transmission logs and determine the multiple source network addresses corresponding to the multiple transmission logs. It can use the AI ​​model to identify the transmission logs to determine the network types corresponding to the transmission logs. Furthermore, it can use the AI ​​model to generate network mapping relationships based on the multiple source network addresses and network types corresponding to the multiple transmission logs, and store the network mapping relationships in a preset storage space. The first CDN node can acquire a first data request sent by a first client and obtain the network mapping relationship from the preset storage space. The first CDN node can determine the first network type corresponding to the first data request based on the network address range and network mapping relationship corresponding to the first data request, and determine a network strategy based on the first network type corresponding to the first data request. It can then process the first data request according to the network strategy. Because network mappings can be generated periodically in advance, they can be sent to multiple CDN nodes, allowing all CDN nodes to use them. When a CDN node receives a data request, it can determine the network address range corresponding to the request and, based on the network mapping, the network type. This eliminates the need for AI models to calculate the network type, separating the AI ​​computation from the data request processing. This prevents AI computation from interfering with data request processing and improves the efficiency of data request handling.

[0266] Figure 5 This is a schematic diagram of a data request processing apparatus provided for an exemplary embodiment of this application. Please refer to... Figure 5 The device can be applied to a first CDN node in a Content Delivery Network (CDN) network. The device may include: a first acquisition module 11, a second acquisition module 12, a first determination module 13, and a second determination module 14, wherein...

[0267] The first acquisition module 11 is used to acquire the first data request sent by the first client;

[0268] The second acquisition module 12 is used to acquire the network correspondence between multiple network address ranges and multiple network types, wherein the network correspondence is determined based on multiple transmission logs in the CDN network;

[0269] The first determining module 13 is used to determine the source network address of the first data request;

[0270] The second determining module 14 is used to determine the first network type corresponding to the first data request based on the network address range where the source network address is located and the network correspondence.

[0271] The data request processing apparatus provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0272] In one possible implementation, the second determining module 14 is specifically used for:

[0273] Determine the first network address range in which the source network address is located;

[0274] The first network type is determined based on the first network address range, where the length of the first network address range is M, and M is an integer greater than 1.

[0275] In one possible implementation, the second determining module 14 is specifically used for:

[0276] If the network correspondence contains a network type corresponding to the first network address range, then the network type corresponding to the first network address range is determined as the first network type.

[0277] If the network type corresponding to the first network address range does not exist in the network correspondence, then the second network address range where the source network address is located is determined, and the network type corresponding to the second network address range is determined in the network correspondence. The network type corresponding to the second network address range is determined as the first network type. The length of the second network address range is N, where N is an integer greater than 1 and N is an integer less than M.

[0278] In one possible implementation, the second acquisition module 12 is specifically used for:

[0279] The network correspondence is obtained from the preset storage space of the first CDN node. The network correspondence is obtained in advance and stored in the preset storage space.

[0280] The data request processing apparatus provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0281] Figure 6 A schematic diagram of another data request processing apparatus provided for an exemplary embodiment of this application. Please refer to... Figure 6 ,exist Figure 5 Based on the illustrated device, the device may further include: a receiving module 15, a storage module 16, a processing module 17, and a generating module 18, wherein,

[0282] The receiving module 15 is used to receive the network correspondence sent by the first device, wherein the first device is a first CDN node or an artificial intelligence (AI) server in the CDN network, and the network correspondence is generated by the first device.

[0283] The storage module 16 is used to store the network correspondence in the preset storage space.

[0284] The data request processing apparatus provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0285] In one possible implementation,

[0286] The first acquisition module 11 is further configured to acquire multiple transmission logs;

[0287] The second determining module 14 is further configured to determine multiple source network addresses corresponding to the multiple transmission logs;

[0288] In one possible implementation,

[0289] The processing module 17 is used to process the multiple transmission logs through an AI model to obtain the network type corresponding to the multiple transmission logs;

[0290] The generation module 18 is used to generate the network correspondence based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs.

[0291] In one possible implementation, the generation module 18 is specifically used for:

[0292] Based on the N-bit network address intervals where the multiple source network addresses are located, multiple source network address sets are determined from the multiple source network addresses, wherein the N-bit network address intervals where the source network addresses in the multiple source network address sets are the same, and the length of the N-bit network address interval is N;

[0293] Based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs, determine the network type corresponding to each source network address;

[0294] The network mapping relationship is generated based on the multiple sets of source network addresses and the network type corresponding to each source network address in the set of source network addresses.

[0295] In one possible implementation, the generation module 18 is specifically used for:

[0296] For any set of source network addresses, a sub-correspondence relationship is determined based on the set of source network addresses and the network type corresponding to each source network address in the set of source network addresses. The sub-correspondence relationship includes at least one network address range and the network type corresponding to each network address range.

[0297] The network correspondence is determined based on the sub-correspondences corresponding to the multiple source network address sets.

[0298] In one possible implementation, the generation module 18 is specifically used for:

[0299] Based on the network type corresponding to each source network address in the source network address set, the proportion of each network type in the multiple network types is determined. The multiple network types include at least one of the following: Long Term Evolution (LTE) of general mobile communication technology, 5G of fifth generation mobile communication technology, WiFi of wireless network communication technology, and 4G of fourth generation mobile communication technology.

[0300] Based on the source network address set, the network type corresponding to each source network address in the source network address set, and the type proportion of each network type, the sub-correspondence relationship corresponding to the source network address set is determined.

[0301] In one possible implementation, the generation module 18 is specifically used for:

[0302] The target network type is determined from the various network types based on the type proportion of each network type, and the target network type has the largest type proportion.

[0303] Determine the initial network address range corresponding to the source network address set, wherein the length of the initial network address range is N;

[0304] If the type proportion of the target network type is greater than or equal to the first threshold, then the sub-correspondence is determined to include the correspondence between the initial network address range and the target network type;

[0305] If the proportion of the target network type is less than the first threshold, then multiple candidate source network addresses are determined in the source network address set, and multiple source network address subsets are determined in the multiple candidate source network addresses. The sub-correspondence is generated based on the initial network address range, the target network type, the multiple source network address subsets, and the network type corresponding to each source network address in the source network address subset.

[0306] Wherein, the network type corresponding to the candidate source network address is a network type other than the target network type, the M-bit network address intervals of each source network address in the source network address subset are the same, and the length of the M-bit network address interval is M.

[0307] In one possible implementation, the generation module 18 is specifically used for:

[0308] For any subset of source network addresses, determine the extended network address range corresponding to the subset of source network addresses, and determine the network type corresponding to the extended network address range based on the network type corresponding to each source network address in the subset of source network addresses.

[0309] Determining the sub-correspondence includes: the correspondence between the initial network address range and the target network type, and the correspondence between the extended network address range and the corresponding network type.

[0310] In one possible implementation,

[0311] The second determining module 14 is further configured to determine a network strategy based on the first network type corresponding to the first data request, wherein the network strategy includes at least one of the following: a congestion control strategy, a packet loss recovery strategy, or a sprint strategy.

[0312] The processing module 17 is further configured to process the first data request according to the network policy.

[0313] The data request processing apparatus provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0314] This application provides an exemplary embodiment of a CDN node structure diagram. Please refer to [link / reference]. Figure 7 The CDN node 20 may include a processor 21 and a memory 22. For example, the processor 21 and the memory 22 are interconnected via a bus 23.

[0315] The memory 22 stores computer-executed instructions;

[0316] The processor 21 executes the computer execution instructions stored in the memory 22, causing the processor 21 to perform the data processing method as shown in the above method embodiment.

[0317] Accordingly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the data request processing method described in any of the above method embodiments.

[0318] Accordingly, embodiments of this application may also provide a computer program product, including a computer program, which, when executed by a processor, can implement the data request processing method shown in any of the above method embodiments.

[0319] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0320] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0321] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0322] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0323] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0324] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0325] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0326] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0327] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A data request processing method, characterized in that, The first CDN node applied in a Content Delivery Network (CDN) network includes: Get the first data request sent by the first client; Obtain the network correspondence between multiple network address ranges and multiple network types; the network correspondence is determined based on the sub-correspondences corresponding to multiple source network address sets, wherein the N-bit network address ranges of the source network addresses in the source network address sets are the same, and the length of the N-bit network address range is N; if the type proportion of the target network type is less than a first threshold, then multiple candidate source network addresses are determined from the source network address sets, and multiple source network address subsets are determined from the multiple candidate source network addresses. For any source network address subset, the extended network address range corresponding to the source network address subset is determined. Based on the network type corresponding to each source network address in the source network address subset, the extended network address range is determined. Network type; the sub-correspondence includes: the correspondence between the initial network address range and the target network type, and the correspondence between the extended network address range and the corresponding network type; the target network type refers to the network type that has the largest proportion among various network types for each source network address in the source network address set, and the network type corresponding to each source network address is obtained by processing multiple transmission logs through an AI model; the length of the initial network address range is N; the network type corresponding to the candidate source network address is a network type other than the target network type, and the M-bit network address ranges in the source network address subset are the same, and the length of the M-bit network address range is M; Determine the source network address of the first data request; Determine the first network address range in which the source network address is located; Based on the first network address range and the network correspondence, the first network type corresponding to the first data request is determined, where the length of the first network address range is M, and M is an integer greater than 1. If the network type corresponding to the first network address interval does not exist in the network correspondence, then the second network address interval where the source network address is located is determined, and the network type corresponding to the second network address interval is determined in the network correspondence. The network type corresponding to the second network address interval is determined as the first network type. The length of the second network address interval is N, where N is an integer greater than 1 and N is an integer less than M. In the second network address interval, all other first network address intervals besides the first network address interval correspond to the same network type.

2. The method according to claim 1, characterized in that, in, If the network type corresponding to the first network address range exists in the network correspondence, then the network type corresponding to the first network address range is determined as the first network type.

3. The method according to claim 1 or 2, characterized in that, Retrieve the network mapping between multiple network address ranges and multiple network types, including: The network correspondence is obtained from the preset storage space of the first CDN node. The network correspondence is obtained in advance and stored in the preset storage space.

4. The method according to claim 3, characterized in that, Before obtaining the network mapping relationship from the preset storage space of the first CDN node, the process further includes: The network correspondence is received from a first device, which is either a first CDN node or an AI server in the CDN network, and the network correspondence is generated by the first device. The network correspondence is stored in the preset storage space.

5. The method according to any one of claims 1-2 and 4, characterized in that, The method further includes: Obtain multiple transfer logs; Determine the multiple source network addresses corresponding to the multiple transmission logs; The network types corresponding to the multiple transmission logs are obtained by processing the multiple transmission logs using an AI model; The network mapping relationship is generated based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs.

6. The method according to claim 5, characterized in that, Based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs, the network mapping relationship is generated, including: Based on the N-bit network address range where the multiple source network addresses are located, determine multiple sets of source network addresses from the multiple source network addresses; Based on the multiple source network addresses corresponding to the multiple transmission logs and the network types corresponding to the multiple transmission logs, determine the network type corresponding to each source network address; The network mapping relationship is generated based on the multiple sets of source network addresses and the network type corresponding to each source network address in the set of source network addresses.

7. The method according to claim 6, characterized in that, Based on the multiple sets of source network addresses and the network type corresponding to each source network address in the set of source network addresses, the network mapping relationship is generated, including: For any set of source network addresses, a sub-correspondence relationship is determined based on the set of source network addresses and the network type corresponding to each source network address in the set of source network addresses. The sub-correspondence relationship includes at least one network address range and the network type corresponding to each network address range. The network correspondence is determined based on the sub-correspondences corresponding to the multiple source network address sets.

8. The method according to claim 7, characterized in that, Based on the source network address set and the network type corresponding to each source network address in the source network address set, determine the sub-correspondence relationship corresponding to the source network address set, including: Based on the network type corresponding to each source network address in the source network address set, the proportion of each network type in the multiple network types is determined. The multiple network types include at least one of the following: Long Term Evolution (LTE) of general mobile communication technology, 5G of fifth generation mobile communication technology, WiFi of wireless network communication technology, and 4G of fourth generation mobile communication technology. Based on the source network address set, the network type corresponding to each source network address in the source network address set, and the type proportion of each network type, the sub-correspondence relationship corresponding to the source network address set is determined.

9. The method according to claim 8, characterized in that, Based on the source network address set, the network type corresponding to each source network address in the source network address set, and the proportion of each network type, the sub-correspondence relationship corresponding to the source network address set is determined, including: The target network type is determined from among the various network types based on the proportion of each type. Determine the initial network address range corresponding to the source network address set; If the type proportion of the target network type is greater than or equal to the first threshold, then the sub-correspondence is determined to include the correspondence between the initial network address range and the target network type.

10. The method according to any one of claims 1-2, 4, 6-9, characterized in that, The method further includes: Based on the first network type corresponding to the first data request, a network strategy is determined, and the network strategy includes at least one of the following: a congestion control strategy, a packet loss recovery strategy, or a sprint strategy. The first data request is processed according to the network policy.

11. A CDN node, characterized in that, include: Memory and processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the data request processing method as described in any one of claims 1-10.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the data request processing method according to any one of claims 1 to 10.

13. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the data request processing method according to any one of claims 1 to 10.

Images