CN116056245A - Data scheduling method, device and computer readable storage medium - Google Patents

Abstract

The application provides a data scheduling method, a data scheduling device and a computer readable storage medium, wherein the method is applied to network equipment and comprises the following steps: identifying a source of each of the at least one data stream; and scheduling a first data stream in the at least one data stream to a target network path according to the source of each data stream in the at least one data stream, wherein the source of the first data stream meets a preset priority range, and the target network path is a network path with the optimal network quality in a plurality of network paths of the network equipment. According to the scheme, the source of the data stream is identified, the data stream meeting the priority condition is scheduled to the target network path with optimal network quality, so that the data stream with relatively important source preferentially uses a relatively good network, and the network utilization is more reasonable.

Description

Data scheduling method, device and computer-readable storage medium

technical field

The present application relates to the communication field, and in particular to a data scheduling method, device and computer-readable storage medium.

Background technique

With the development of wireless router technology, networking schemes such as Mesh networks have emerged. This type of networking is mainly formed through the layout of main routes and sub-routes according to a certain topology. And the current wireless router can support dual networks, that is to say, the wireless router can access the Internet through a mobile network or Ethernet. For example, a wireless router that supports SIM card insertion and network cable connection can be connected to a mobile network or an Ethernet.

In actual scenarios, it is very common that multiple applications of multiple devices or multiple applications of one device access the wireless router at the same time. At this time, the network burden is heavy, and there may be situations such as freezing or disconnection. Although the connected wireless router can be a dual-network wireless router, which can support mobile network and Ethernet, but because the current solution is still using one of the networks, or even using two networks is not smart, so the two networks The use is unreasonable and affects the user's online experience.

That is to say, the current dual-network wireless router only has the ability to access the Internet with two networks, but this ability has not played a good role, that is, the network utilization is unreasonable, resulting in the inability to meet diversified Internet access needs.

Therefore, how to make more rational use of the network of wireless routers to meet diversified Internet access requirements is an urgent technical problem to be solved.

Contents of the invention

The present application provides a data scheduling method, network equipment, and computer-readable storage medium, which can make more reasonable use of the network, thereby satisfying diversified Internet access requirements.

In a first aspect, a data scheduling method is provided, which is applied to a network device, and the method includes: receiving at least one data flow; identifying the source of each data flow in the at least one data flow; according to each data flow in the at least one data flow The source of at least one data flow is dispatched to the target network path, the source of the first data flow satisfies the preset priority range, and the target network path has the highest network quality among the multiple network paths of the network device. Excellent network access.

The technical solution of the present application mainly identifies the source of the received data flow, and dispatches the data flow satisfying the priority condition to the target network path with the best network quality. It can be understood that data streams with relatively important sources preferentially use relatively good networks. Multiple network paths are considered, and network utilization is more reasonable. In addition, data streams with relatively important sources are given priority to use high-quality network channels, which also takes into account the diversified Internet access needs of users.

In combination with the first aspect, in some implementations of the first aspect, when identifying the source corresponding to each data flow in at least one data flow, the following operations may be performed: use the recognition model to process at least one data flow to obtain A source for each of the at least one data stream includes at least one of a business source or an application source. By identifying at least one of the type of business or which application program the data flow comes from, then according to the priority of the source, the data flow of some or a certain source with high priority can be dispatched to the network with the best quality on the target network path. The refinement of source determination can make the scheduling of data flows more in line with Internet access requirements.

Optionally, the recognition model may use a pre-trained neural network model. For example, the neural network model can be trained by using the data stream with the source label (that is, the data stream of known source), and the trained neural network model obtained has the ability to identify the source of the data stream.

With reference to the first aspect, in some implementation manners of the first aspect, when scheduling the first data flow in the at least one data flow to the target network path according to the source of each data flow in the at least one data flow, you may Perform the following operations: determine whether the source of each data flow in at least one data flow meets a preset priority range, and determine the data flow that meets the preset priority range as the first data flow; mark the first data flow; The tagged first data flow is dispatched onto the target network path. Scheduling is made more convenient by marking the first data flow before scheduling. It can be understood that the first data flow is screened out and marked, and the marked data flow can be preferentially dispatched to the target network path during scheduling. In this implementation, after knowing the source of the received data streams and determining which data streams are the first data streams, the first data streams are marked (locked) from the received data streams, and the locking method is to open mark.

With reference to the first aspect, in some implementation manners of the first aspect, when marking the first data flow, the following operations may be performed: parsing the IP information of each data flow in at least one data flow, and converting the IP information into Five-tuple information: determining and marking the first data stream from the at least one data stream according to the five-tuple information. This implementation manner can be understood as an example of a marking manner. Each data flow can be distinguished through the quintuple information, and the quintuple information such as the source address and the destination address of each data flow can be known, so that it can be known whether each data flow is the first data flow and marked.

Optionally, the IP information may be obtained by extracting features of the data stream using the recognition model described above. It can be understood as the output of the process of identifying the source of the data stream. Therefore, it is not necessary to independently design an extraction module or an extraction method of IP information to obtain IP information.

With reference to the first aspect, in some implementations of the first aspect, the above method further includes: determining the network quality of multiple network paths according to the quality parameters of multiple network paths, the quality parameters including signal quality indicators, rate or delay At least one of them, the signal quality index and the rate are positively correlated with the network quality, and the time delay is negatively correlated with the network quality. This implementation mainly determines the network quality through quality parameters.

With reference to the first aspect, in some implementations of the first aspect, when determining the network quality of multiple network paths according to the quality parameters of multiple network paths, the following operations may be performed: according to the multiple network paths in the first The quality parameters of multiple time periods in the time interval, to obtain the network quality metrics of multiple network paths in multiple time periods; the average or weighted average of the network quality metrics of multiple network paths in multiple time periods It is the network quality of multiple network paths in the first time interval. In this implementation manner, by averaging evaluation values (measurement values) of network quality in multiple time periods, the measurement result can be made more accurate, and can be a long-term measurement. The first time interval may be a day, a month, a quarter or even a year. When the first time interval is a day, the multiple time periods may be multiple time periods determined within 24 hours of a day. By calculating the weighted average, for example, increasing the weight of the peak period of Internet access and reducing the weight of the low period of Internet access, the accuracy of the measurement result can be further improved.

With reference to the first aspect, in some implementation manners of the first aspect, the foregoing multiple network paths may include a mobile network path and an Ethernet path. In this implementation mode, the network access includes two types of mobile network and Ethernet, realizing various ways of accessing the Internet. However, it should be understood that in a possible implementation manner of the embodiment of the present application, the multiple network paths may be the same type of network paths, or may be different types of network paths.

Optionally, different signal quality indicators may also be set for different types of network paths. For example, the signal quality index of the mobile network path includes a received signal strength indicator RSSI index, and the signal quality index of the Ethernet path includes a negotiated rate. Different types of network paths use different signal quality indicators, which can make the measurement of network quality more accurate.

With reference to the first aspect, in some implementations of the first aspect, the above method further includes: scheduling the second data flow of the at least one data flow on a non-target network path, the second data flow is used to represent the at least one data flow Among the data flows other than the first data flow, the non-target network path is used to represent a network path other than the target network path among the plurality of network paths. In this implementation method, on the basis that data streams with relatively important sources are preferentially dispatched to networks with better quality, other data streams are also dispatched to other network paths, so that all network paths can be fully utilized, improving Improve network utilization and increase Internet speed.

In a second aspect, a data scheduling device is provided, and the device includes a unit composed of software and/or hardware for executing any one of the methods in the first aspect.

A third aspect provides a network device, including a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, any one of the methods in the first aspect can be implemented.

With reference to the third aspect, in some implementation manners of the third aspect, the network device is a wireless router.

In a fourth aspect, a chip is provided, including a processor, the processor is configured to read and execute a computer program stored in a memory, and when the computer program is executed by the processor, any one of the methods in the first aspect can be implemented.

Optionally, the chip further includes a memory, and the memory is electrically connected to the processor.

Optionally, the chip may also include a communication interface.

In a fifth aspect, a computer-readable storage medium is provided, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any one of the methods in the first aspect can be implemented.

In a sixth aspect, a computer program product is provided, the computer program product includes a computer program, and when the computer program is executed by a processor, any one of the methods in the first aspect can be implemented.

Description of drawings

FIG. 1 is a schematic diagram of an applicable network scenario according to an embodiment of the present application.

Fig. 2 is a schematic flowchart of a data scheduling method according to an embodiment of the present application.

Fig. 3 is a schematic flow chart of marking data flow according to the embodiment of the present application.

Fig. 4 is a schematic diagram of a deployment position of an identification module according to an embodiment of the present application.

FIG. 5 is a schematic diagram of an identification process of a data flow source according to an embodiment of the present application.

Fig. 6 is a schematic flowchart of a network quality measurement method for a mobile network according to an embodiment of the present application.

Fig. 7 is a schematic flowchart of a method for measuring network quality of an Ethernet according to an embodiment of the present application.

FIG. 8 is a schematic flowchart of another data scheduling method according to an embodiment of the present application.

Fig. 9 is a schematic diagram of a data scheduling device according to an embodiment of the present application.

Detailed ways

The solutions of the embodiments of the present application will be introduced below with reference to the accompanying drawings. The data scheduling method provided in this application can be applied to routers and other network devices capable of wireless network communication, and can mainly be applied to routing devices with multiple network paths.

FIG. 1 is a schematic diagram of an applicable network scenario according to an embodiment of the present application. As shown in FIG. 1 , this scenario includes at least one network device 110 and at least one downlink device 120 .

The network device 110 can provide a wireless network, and the downlink device 120 can access the Internet through the network device 110 . The downlink device 120 may include a terminal device and a network device, that is, the downlink device 120 only needs to be able to access the wireless network provided by the network device 110 . FIG. 1 is an example where the downlink device 120 includes terminal devices 121, 127-129 and network devices 122-126. It should be understood that when the downlink device is a network device, there may be a terminal device or other network devices downlinked to the network device. As shown in FIG. 1 , the downlink device 122 is a router, and the router also has a downlink device router 125 , and the router 125 also has a downlink device, that is, a terminal device 128 . At this time, the terminal device 128 accesses the network device 110 through the network devices 125 and 122, and then accesses one of the network paths through the network device 110, and finally accesses the Internet.

In Fig. 1, also comprise network equipment 123 and 126, and terminal equipment 129, wherein terminal equipment 129 is the downlink equipment of network equipment 126, network equipment 123 is the downlink equipment of network equipment 126, and network equipment 126 is network equipment 110 The downlink device, the network device 126 and the network device 110 are wired. In the figure, it is taken that the terminal device 129 is a notebook computer as an example.

In Fig. 1, also include network equipment 124 and terminal equipment 127, wherein, terminal equipment 127 is the hanging equipment of network equipment 124, and network equipment 124 is the hanging equipment of network equipment 110, between network equipment 124 and network equipment 110 is a wireless connection.

It can be seen that the downlink device of the network device 110 may be a terminal device or a network device, and the downlink network device connected to the network device 110 may be a wired connection or a wireless connection. That is, the network devices can be wired or network connected. Network devices may or may not be routers of the same type.

It should be understood that no matter the connected device is a network device or a terminal device, and no matter how many layers of network devices are connected, the origin of the final data can be traced back to the terminal device.

Terminal equipment can also be called user equipment, which can be a mobile terminal such as a mobile phone, a computer, a computer, a tablet computer, a smart terminal, a projection device, an Internet TV device, a vehicle terminal, or other terminal equipment capable of connecting to a network. The smart terminal may be, for example, a smart TV, a smart speaker, a smart bracelet, or a smart screen.

The network device 110 can access the Internet through the first network path 130 or the second network path 140 . In the scenario shown in Figure 1, the network path can be either a mobile network or an Ethernet. FIG. 1 takes an example in which the first network path 130 is a mobile network and the second network path 140 is an Ethernet. However, it should be understood that the first network path 130 and the second network path 140 may be network paths of the same type or not, and there is no limitation. In addition, more network paths may be included between the network device 110 and the Internet.

Taking the first network access 130 as a mobile network access and the second network access 140 as an Ethernet access as an example, as shown in Figure 1, the mobile network access includes an access node 131, and Figure 1 takes the access node 131 as a base station as an example, It is used to provide a mobile network signal, and the mobile network may be a long term evolution (long term evolution, LTE) network or a fifth generation (5G) network and other types of mobile networks. The Ethernet path includes a switch 141 and an access server 142 through which the Internet can be accessed. Ethernet can be understood as a wired network, but it does not specifically refer to network access through network cables. For ease of understanding, in some cases, accessing the Internet through a mobile network channel can be regarded as using traffic to access the Internet, while accessing the Internet through an Ethernet channel can be regarded as using broadband to access the Internet.

In the scenario shown in FIG. 1 , when the downlink device 120 of the network device 110 accesses the Internet through the network device 110 , it can access the Internet through the first network path 130 or the second network path 140 . For example, the network device 110 receives the data stream of the downlink device 120, and then uploads it to the Internet through the first network path 130 or the second network path 140; or, the network device 110 receives the Internet data flow, and then send the received data flow to the hanging device 120.

In the scenario shown in FIG. 1 , a dotted line indicates a wireless connection mode, and a solid line indicates a wired connection mode. It can be seen that the first network path 130, that is, the mobile network path is in a wireless connection manner, and the second network path 140, that is, an Ethernet path is in a wired connection manner. The connection between the downlink device 120 and the network device 110 may also be wireless or wired. A wireless or wired connection may also be used between the hanging devices, for example, a wireless or wired connection may be used between the hanging routers.

Take the Mesh networking scenario shown in Figure 1 as an example. Then the network device 110 can be the main router (controller) in the Mesh networking, also can be referred to as the mother router, can be used as customer premise equipment (customer premise equipment, CPE), this equipment manages the whole Mesh network in a unified way, such as channels, Topology structure, access of connected devices, access to external network, etc. In Mesh networking, CPE can meet the requirements of mobile network access and Ethernet access in combination with its own network access characteristics. The main router can be used as a CPE device.

The downlink device 120 includes terminal devices 121, 127-129 and network devices 122-126. At this time, the network devices 122-126 are sub-routes (agents), which can also be called sub-routes, and mainly execute commands sent by the main router. , and periodically report network usage and self-capability information to the main router, such as how many terminal devices are connected or which frequency bands and bandwidth are used.

It should be understood that the sub-router can be connected to the parent router in a wired (eg network cable connection) or wireless (eg WIFI) manner.

In the process of data interaction in Figure 1, since the network quality of the two network paths may be different, the quality or rate of data transmission connected to the Internet through different network paths is different, that is, the user's online experience is different. The embodiments of the present application mainly aim at the multi-network scenario (that is, the scenario where the network device supports multiple network paths to access the Internet), and improve network utilization.

Fig. 2 is a schematic flowchart of a data scheduling method according to an embodiment of the present application. The method shown in FIG. 2 can be applied to the network device 110 shown in FIG. 1 . For ease of understanding, the following mainly introduces that the network device is the main router in the Mesh network and the terminal device is a mobile phone.

S201. Receive a data stream.

The data flow received in step S201 may come directly from the terminal device, or may come from a sub-router. When the received data flow comes from the sub-router, the data flow can still be traced back to the specific terminal device. For example at least one data stream can be received. The at least one data stream may come from a terminal device of the main route, or may come from a downstream terminal device of a sub-router.

When a user surfs the Internet through a terminal device, such as using a mobile phone to chat, video, and consult information. That is, the Internet is connected by using one or more application programs (application, APP). Then step S201 can receive the data stream generated by the APP used. It is also possible for multiple terminal devices to surf the Internet at the same time. For example, in a certain scenario, the smart speaker is playing the weather forecast while the Internet TV is playing a program and the mobile phone is making a video call.

S202. Identify the source of the data flow.

That is to say, the source of each data stream in the at least one data stream received in step S201 is identified.

In the embodiment of the present application, the source can be specific to which APP it comes from, and the APP can also be classified according to business type, then the source can refer to which type of business it comes from. Therefore, the source may include at least one of a service source or an application program source (ie, an APP source).

The service source can be regarded as classifying the APP according to the service type, and the source of the data flow from the APP of a certain service type is called the service source of the service type. For example, if the source of business is office, it means that an app related to office can be classified as an office app (that is, the source of business is office). Office-related apps can include e-mail or office software independently developed by the enterprise, for example. That is to say, the data flow from the office-related APP is called the data flow whose business source is office.

The business source may include, for example: at least one of application market, game, audio-visual, live broadcast, news, social networking, shopping, payment, life, online class, video call, download or upload. Audio and video may include interactive audio and video and non-interactive audio and video. Business sources can also be divided in other ways, for example, they can be divided into: consumption, entertainment, life, work, study and others.

The APP source can be indicated by the APP name, such as electronic dictionary, weather, application market or email, and will not be listed one by one.

Each business source includes at least one APP source.

In one example, business sources include: application market, games, short videos, videos, live broadcasts, news, social chat, office, shopping, payment, life, online courses and downloads. The sources of APPs included in the application market are the application markets of various manufacturers, such as the application center and Yongyongbao. The game includes multiple specific game APPs. Short videos include APPs of various short video platforms. Videos include apps that users use to watch audio-visual (audio-visual) programs such as movies and TV. News includes apps of various news platforms, and social chat includes apps with social functions. Office includes office apps such as video calls or conferences. Shopping includes various shopping apps. Payment includes various payment platform APPs or mobile wallets, etc. Life can include various apps such as food, takeaway, and weather. Online courses can include various learning apps. Downloads can include downloading various apps, network disks, or browsers.

It can be seen that business sources can be classified in a variety of ways, and the same APP can sometimes be classified into different business sources. For convenience, the main functions of the APP can be used as the basis for classification. For example, a payment app with a chat function, whose main function is payment, is classified into the payment category.

In an implementation manner, the source includes at least one of a service source or an application program source (that is, an APP source). When identifying the source of the data flow, the data flow can be processed through the identification model, so as to obtain the source of the data flow. By identifying which type of service or application the data flow comes from, it is possible to prioritize some or a certain source of high-priority data flow to the target network path with the best network quality according to the priority of the source. . The determination of the source is more refined, so that the scheduling of the data flow is more in line with the Internet access requirements.

Optionally, at least one data flow may be processed by using an identification model to obtain a source of each data flow in the at least one data flow, and the source includes at least one of a service source or an application program source. By identifying at least one of the type of business or which application program the data flow comes from, then according to the priority of the source, the data flow of some or a certain source with high priority can be dispatched to the network with the best quality on the target network path. The refinement of source determination can make the scheduling of data flows more in line with Internet access requirements.

Optionally, the recognition model may use a pre-trained neural network model or other big data analysis models. For example, the neural network model can be trained by using the data stream with the source label (that is, the data stream of known source), and the trained neural network model obtained has the ability to identify the source of the data stream.

In one example, a recognition model is obtained by training a convolution neural network (CNN) model. The training data includes a data flow and a source label corresponding to the data flow, and the source label includes a business source and a program source. After CNN processes the data stream, it will get the predicted value and confidence of the source. Therefore, identifying the source of the data flow can be regarded as a classification task, that is, to classify the data flow according to the source, and the output result, that is, the category, is the source of the data flow. During the training process, the data stream is input to the CNN, and the obtained source category is compared with the actual source, that is, the above source label, and the weight of the CNN is adjusted, so that the trained CNN has the ability to identify the source of the data stream. Or it can be understood that CNN has the ability to identify the source of data flow by learning the data flow of known sources. The trained CNN model can identify the source of the data stream of unknown origin input into the model.

Because the port information and official configuration information of different APPs are different, and the network transmission characteristics of different types of APPs are also different. For example, for a downloading APP, there are more downstream data streams; for a retrieval APP, there are more upstream times, and the number of data streams generated by one interaction is also different. And other characteristic parameters that can distinguish the source of different data streams. Therefore, by extracting such features and classifying based on these features, the APP sources of different data streams can be distinguished.

The processing process of the data stream by the recognition model can be regarded as including two steps of feature extraction and classification. The feature extraction part is to convert the data stream into feature vectors, which include the above and other features that can distinguish different sources. Classification of these eigenvectors yields sources.

It should be understood that the above is just an introduction using CNN as an example, and those skilled in the art may also use other neural networks or other deep learning models or machine learning models to train and obtain the recognition model. For example, a deep neural network, a recurrent neural network, etc. may be used, and will not be listed one by one.

Identifying the business source and/or APP source of the data stream, and then performing subsequent scheduling, can make multiple services or multiple applications use the network at the same time, scheduling and forwarding in a prioritized and secondary manner, so as to meet the diversified Internet access needs of users. For example, in one scenario, the user simultaneously opens three types of services: audio-visual, payment, and office services on one of the connected devices of the router. The data flow of the three types of business, that is, the source of the above three business sources, can be scheduled through subsequent steps, for example, the data flow of the payment service is prioritized as the first data flow described below to be dispatched to a network channel with better quality . In this example, if the office is more important, the data flow of the office business may be prioritized as the first data flow described below to be scheduled on a network path with better quality. For another example, in another scenario, the user opens APP1-APP5, a total of 5 APPs, on a connected device of the router at the same time. After the router executes step S201 to receive the data stream from the connected device, it executes step S202 to identify The data streams of these 5 APPs, that is, the sources of the above 5 APP sources, can be scheduled through subsequent steps, for example, the data streams of one or more APPs with the highest priority among APP1-APP5 are used as the first data described below Flows are preferentially dispatched to network channels with better quality.

In another implementation manner, the above-mentioned source may also include a device source, that is, it is identified which downlink device the data stream originates from. Only the method of identifying the source of the device is different from the method of identifying the source of the business and the source of the APP, that is to say, an identification module capable of identifying the address of the device is required to identify the source of the device. This manner may be to identify which type of device the data stream originates from or which device the data stream originates from, and these two identifications are also different. In this way, when setting the priority, you can set the priority of the connected devices. For example, in the office scene, you can set the priority of the connected devices for The priority of the router is higher than that of the TV, and the office computer and the TV are devices connected to the same router. The above three sources of business source, APP source and device source can be combined with each other, or only one or more of them can be used. In an example, the source may be a device source, a service source, or an APP source. In another example, the source is a combination of a device source and a service source, that is, which service of which device the data stream comes from, or which device under which service the data stream comes from. In yet another example, the source is a combination of a device source and an APP source, that is, which APP of which device the data flow comes from. In yet another example, the source is a combination of a service source and an APP source, that is, which APP under which service the data flow comes from. In yet another example, the source is a combination of a device source, a service source, and an APP source, that is, which APP under which service of which device the data flow comes from.

S203. Schedule the first data flow on the target network path according to the source of the data flow.

That is to say, according to the source of the data flow identified in S202, the first data flow in the at least one data flow received in S201 is scheduled to the target network path. Wherein, the source of the first data stream satisfies the preset priority range. The target network path is a network path with the best network quality among the plurality of network paths.

The aforementioned plurality of network accesses may include mobile network accesses and Ethernet accesses. That is, the network access includes two types of mobile network and Ethernet, so that multiple ways of accessing the Internet can be realized. However, it should be understood that in the embodiment of the present application, the multiple network paths may be the same type of network paths, or may be different types of network paths.

Step S203 can be regarded as first filtering out the data flows whose sources meet the preset priority range, and then dispatching these data flows to the target network path. for example:

Example 1, assuming that the preset priority range is divided according to business sources, and the business sources include: payment, office, consumption, life and others, and the priority is lowered in order, and the preset priority range is the top 2 business sources. Then, the first data flow is the data flow from the payment and office business sources.

Example 2, assuming that the preset priority range is divided according to the APP source, the APP source includes: APP#1-APP#X, X is a positive integer, and the priority is lower in order, and the preset priority range is the top 3 APPs source. Then, the first data stream is the data streams from three APP sources including APP#1-APP#3. For example, the priority of the payment service on the tablet computer is higher than that of the life service on the mobile phone.

Example 3, assuming that the preset priority range is divided according to the combination of business sources and APP sources. Business sources include: payment, office, consumption, life and others. Payment includes 3 APP sources, office includes 2 APP sources, and consumption includes 4 APP sources, Life includes 7 APP sources and Others includes 10 APP sources, and the priority is lowered in order. That is to say, the 3 APP sources in the payment are the top 3 in the priority order, followed by the priority order of the 2 APP sources in the office are 4 and 5, and so on. The preset priority range may be, for example, the top 10 APP sources. Then the first data stream comes from the three businesses of payment, office and consumption, as well as the APP ranked first in life.

Example 4, assuming that the preset priority range is divided according to the source of the device. The source of the device includes: laptop, mobile phone, smart screen, and smart speaker. The priority is lower in order, and the preset priority range is the device source with the highest priority. Then the first data stream is the data stream from the notebook computer, and at this time, the business source and APP source of the data stream need not be considered. However, in this case, an identification module capable of identifying the address of the device needs to be additionally used to identify the source of the device.

Other situations will not be listed one by one. It should be understood that the examples of service sources, APP sources, device sources, and specific values in the above examples are only for the convenience of understanding the solution, and there is no limitation.

In a possible implementation, step S203 in FIG. 2 may include the following operations: judging whether the source of each data stream in at least one data stream satisfies a preset priority range, and the data that satisfies the preset priority range The flow is determined as the first data flow; the first data flow is marked; and the first data flow with the mark is dispatched to the target network path.

Scheduling is made more convenient by marking the first data flow before scheduling. It can be understood that the first data flow is screened out and marked, and the marked data flow can be preferentially dispatched to the target network path during scheduling. In this implementation, after knowing the source of the received data streams and determining which data streams are the first data streams, the first data streams are marked (locked) from the received data streams, and the locking method is to open mark.

In an implementation manner, the network quality of the multiple network paths may be determined according to the quality parameters of the multiple network paths. The quality parameter can be understood as a parameter that can affect network quality. The quality parameter may include at least one of signal quality index, rate or delay. Among them, the signal quality index and the rate are positively correlated with the network quality, and the delay is negatively correlated with the network quality. In other words, the higher the signal quality index, the faster the rate, and the smaller the delay, the higher the network quality. The detailed measurement process will be given in Figure 6 and Figure 7 below, and will not be expanded here.

In an example, different signal quality indicators may also be set for different types of network paths. For example, the signal quality index of the mobile network path may include a received signal strength indication (received signal strength indication, RSSI) index, and the signal quality index of the Ethernet path may include a negotiated rate. Different types of network paths use different signal quality indicators, which can make the measurement of network quality more accurate.

In an example, multiple measurements may be averaged to make the measurement result of the network quality more accurate. For example, according to the quality parameters of multiple network paths in multiple time periods in the first time interval, the network quality metric values of multiple network paths in multiple time periods can be obtained; The average or weighted average of the quality metric values is used as the network quality of the multiple network paths in the first time interval. The method can make the measurement result more accurate and can be a long-term measurement by averaging network quality evaluation values (also referred to as measurement values, measurement values) in multiple time periods. The first time interval may be a day, a month, a quarter or even a year. When the first time interval is a day, the multiple time periods may be multiple time periods determined within 24 hours of a day. By calculating the weighted average, for example, increasing the weight of the peak period of Internet access and reducing the weight of the low period of Internet access, the accuracy of the measurement result can be further improved.

Let's take a specific example to illustrate, assuming that the following five time points in a day: 9:00, 12:00, 17:00, 19:00, 21:00 are the measurement start time, and statistics are made every 15 minutes. The time period is calculated 3 times, a total of 15 times in one day, that is, 9:00-9:45, 12:00-12:45, 17:00-17:45, 19:00-19:45, 21 :00-21:45 The network quality was measured 3 times in 5 time intervals, and 3 network quality measurement values were obtained respectively. A total of 15 time periods were measured in one day (an example of the first time interval) (multiple time intervals) One example) network quality, and 15 network quality measurement values are obtained, then the average value or weighted average value of these 15 network quality measurement values can be used as the network quality measurement result of this day. It is also possible to calculate the average value or weighted average value as the average network quality within a month by counting the network quality measurement values of each day in a month. It should be understood that the numerical values in this example are for the convenience of understanding the solution, and there is no limitation. The above five time points can be regarded as a specific example of taking multiple time points during the peak period of surfing the Internet in a day.

The method shown in FIG. 2 mainly identifies the source of the received data flow, and dispatches the data flow satisfying the priority condition to the target network path with the best network quality. It can be understood that data streams with relatively important sources preferentially use relatively good networks. Multiple network paths are considered, and network utilization is more reasonable. In addition, data streams with relatively important sources are given priority to use high-quality network channels, which also takes into account the diversified Internet access needs of users.

In a possible implementation manner, the method shown in FIG. 2 may further include: scheduling a second data flow in at least one data flow to a non-target network path, where the second data flow is used to indicate A data flow other than a data flow, the non-target network access is used to indicate a network access other than the target network access among the plurality of network accesses. In this implementation method, on the basis that data streams with relatively important sources are preferentially dispatched to networks with better quality, other data streams are also dispatched to other network paths, so that all network paths can be fully utilized, improving Improve network utilization and increase Internet speed.

In an implementation manner, the method shown in FIG. 3 may be used to mark the first data stream. Fig. 3 is a schematic flow chart of marking data flow according to the embodiment of the present application. Fig. 3 can be regarded as an example of marking the first data stream in step S203.

S301. Parse IP information of each data flow in at least one data flow, and convert the IP information into quintuple information.

In the embodiment of the present application, an independent module can be set in the core IP protocol stack to identify the corresponding source according to the input data flow. For simplicity, it can be called an identification module or a service aware (SA) module.

In the kernel IP protocol stack, the netfilter framework has five key hook points for data flow, which can be called detection points, key points or hook points. The forwarding data of the connected device will go through three hook points: pre-routing, key node (forward), and post-routing.

Usually, the identification module works at the forward of netfilter, and this module takes the data flow as a unit, parses out the IP information of the data flow, and converts the IP information into five-tuple information. The five-tuple information includes: source IP, destination IP, source port, destination port, protocol TCP/UDP. The device or application corresponding to the flow can thus be identified.

For ease of understanding, the structure of the netfilter framework is introduced below in conjunction with Figure 4. As shown in Figure 4, the netfilter framework includes routing pre-processing nodes (pre-routing), routing decision nodes, key nodes (forward), routing post-processing nodes (post-routing), and entry (local_in) and exit (local_out )node. Among them, the pre-routing processing node is PRE_ROUTING in the figure, which mainly processes destination address conversion and adds specific marks to data packets (ie, data packets or data streams). The routing decision node is the routing decision shown in the figure, which is used to decide whether the data packet is forwarded and processed or used by the router for local services. The protocol stacks above four layers in FIG. 4 may also be referred to as top-layer protocol stacks. In the embodiment of the present application, the identification module is set at the key node, and is used to identify the source of the data message. The post-routing processing node is POST_ROUTING in the figure, which mainly completes source address conversion. The entry of the processing node of the router's local service is LOCAL_IN in the figure, and the exit of the processing node of the router's local service is the LOCAL_OUT in the figure.

Optionally, the IP information may be obtained by extracting features of the data stream using the recognition model described above. It can be understood as the output of the process of identifying the source of the data stream. Therefore, it is not necessary to independently design an extraction module or an extraction method of IP information to obtain IP information.

S302. Determine and mark a first data stream from at least one data stream according to the quintuple information.

Since the five-tuple information includes information such as source IP and source port, the source of the data flow can be deduced.

Optionally, a data stream information table of the identification module can also be established, and the data stream information table includes whether the received data stream has been identified and the identification result.

The method shown in FIG. 3 can be understood as an example of a marking manner. Through the quintuple information, each data flow can be distinguished, and the quintuple information such as the source address and destination address of each data flow can be known, so that it can be known whether each data flow is the first data flow, and then marked. If the process of identifying the first data stream by the identification module is regarded as classifying the received data streams, the source category of each data stream is known. Then, the process of marking data streams is to find out one or some data streams from these data streams, so as to lock these data streams by means of marking. It can be understood that the identification process is a process of obtaining a list of the first data stream, and the marking process is a process of locking the first data stream according to the list.

FIG. 5 is a schematic diagram of an identification process of a data flow source according to an embodiment of the present application. It should be understood that the identification of the source of the data stream shown in Figure 5 is to mark the first data stream, which can be regarded as a process of finding out (locking/marking) the first data stream according to the identification result, and it is not that the identification module identifies each The process of the source of the data stream.

S501. Receive a data stream to be identified through a stream identification entry.

The entry function can be represented by atp_sc_input, and the corresponding interface can be represented by S.

S502. Analyze the IP information of the data stream and convert it into 5-tuple information.

The analysis function can be represented by atp_sc_ct_to_tuple, and the corresponding interface can be represented by S1.

S503. Search for flow table information corresponding to the service identification module according to the quintuple information.

The search function can be represented by atp_sc_data_resove, and the corresponding interface can be represented by S2.

The flow table information can include whether the recognition is completed and the result of the recognition, such as business sources such as games and videos, or specific APP sources such as Glory of Kings, Peace Elite, and Alipay.

S504. Judging whether the flow identification is completed, if the judgment result is yes, execute step S505, and when the judgment result is no, execute step S506.

S505. Traffic statistics and counting.

Statistical functions can be represented by atp_sc_serv_stats_add.

S506. Identify the data flow.

That is to say, if it is found from the flow table information that it has not been identified, then the data flow is identified and the flow table information is filled.

The execution function of this step can be represented by atp_sc_eng_input, and the corresponding interface can be represented by S3.

This step may also include identifying pre-recognition configuration parameters and a recognition engine for specific recognition.

The identified data streams can then be marked accordingly.

Fig. 5 can be regarded as a specific example of the method shown in Fig. 3, S502 can be regarded as an example of S301, and S503-S506 can be regarded as an example of S302.

As mentioned above, the network quality can be determined according to the quality parameters of the network path. The following introduces the measurement methods for the network quality of the mobile network and the Ethernet respectively with reference to FIG. 6 and FIG. 7 .

Fig. 6 is a schematic flowchart of a network quality measurement method for a mobile network according to an embodiment of the present application. Each step in Fig. 6 will be introduced below.

S601. The AT module collects the signal quality index of the modem module.

FIG. 6 takes RSSI as an example for the signal quality indicator. The modem module is mainly responsible for real-time communication with the transmission equipment in the mobile network path such as the base station, receiving the command information of the base station in real time, and also actively querying the information from the base station. It belongs to the mobile network wireless core module of the CPE, and the interface providing the interface for the AT module can be express.

S602. The AT module sends a signal quality indicator to the quality measurement module.

The AT module collects the signal data of the modem module in real time through the interface M, and reports the RSSI and other indicators of the modem signal to the quality measurement module at the same time. For example, in scenarios such as location movement and signal interference of the CPE, the signal quality will be reported in time for the quality measurement module to update the network quality in a timely manner. Here, it is assumed that the interface reported to the quality measurement module is interface A.

S603. The rate statistics module collects rate and delay information.

The rate statistics module collects the data rate and delay sent by the network interface layer.

S604. The rate statistics module sends rate and delay information to the quality measurement module.

The rate statistics module sends the collected rate and delay information to the quality measurement module. For example, they can be sent via interface T1 and interface T2 respectively.

S605. The quality measurement module evaluates the network quality according to the received quality parameter information.

Optionally, the network quality (Q) may be calculated according to the rate (V), time delay (T) and signal quality index (R).

In an implementation manner, Q may satisfy the following formula: Q=(aV+bR)/(cT ² +d), where a, b, c, and d are calculation factors and may be determined through commissioning. That is to say, the value of the calculation factor can be determined through mathematical modeling by obtaining sufficient historical data of V, T, R, and Q.

It can be seen from the above implementation manners that V and R are positively correlated with Q, and T is negatively correlated with Q. It should be understood that the formula of this implementation is just an example, and those skilled in the art can also design other formulas, for example, a multiplicative relationship is used between V and R, or T does not perform square operations, etc., as long as the requirements are met, there is no limited.

In an example, N times are measured in one day, N is a positive real number, and Q _i is used to represent the measurement value of the i-th network quality, that is, the measurement result, 1≤i≤N, then the signal quality of the Q _d-th day Q _d =∑Q _i /N. Through the signal quality of each day, the signal quality of the month can also be calculated as Q _m =∑Q _d /M, where M is the number of measurement days in the current month.

Fig. 7 is a schematic flowchart of a method for measuring network quality of an Ethernet according to an embodiment of the present application. Each step in Fig. 7 will be introduced below.

S701. The signal quality acquisition module sends the negotiation rate to the quality measurement module.

FIG. 7 takes the signal quality index as the negotiation rate as an example.

S702. The rate statistics module collects rate and delay information.

The rate statistics module collects the data rate and delay sent by the network interface layer.

S703. The rate statistics module sends rate and delay information to the quality measurement module.

The rate statistics module sends the collected rate and delay information to the quality measurement module. For example, they can be sent via interface T1 and interface T2 respectively.

S704. The quality measurement module evaluates the network quality according to the received quality parameter information.

Optionally, the network quality (Q) may be calculated according to the rate (V), time delay (T) and signal quality index (N).

In an implementation manner, Q may satisfy the following formula: Q=(aV+bR)/(cT ² +d), where a, b, c, and d are calculation factors and may be determined through commissioning. That is to say, the value of the calculation factor can be determined through mathematical modeling by obtaining sufficient historical data of V, T, R, and Q.

It can be seen from the above implementation manners that V and R are positively correlated with Q, and T is negatively correlated with Q. It should be understood that the formula of this implementation is just an example, and those skilled in the art can also design other formulas, for example, a multiplicative relationship is used between V and R, or T does not perform square operations, etc., as long as the requirements are met, there is no limited.

In an example, N times are measured in one day, N is a positive real number, and Q _i is used to represent the measurement value of the i-th network quality, that is, the measurement result, 1≤i≤N, then the signal quality of the Q _d-th day Q _d =∑Q _i /N. Through the signal quality of each day, the signal quality of the month can also be calculated as Q _m =∑Q _d /M, where M is the number of measurement days in the current month.

It can be seen that the measurement method of Ethernet is similar to that of mobile network. And through experimental statistics, it is found that in some scenarios, the network quality of the mobile network and the network quality of the Ethernet have a certain correlation, and it is found through experiments that under certain circumstances, a, c, and d can inherit the corresponding values of the mobile network , b can be multiplied by a weight coefficient in the mobile network, and the weight coefficient can be, for example, 0.2 or 0.3.

FIG. 8 is a schematic flowchart of another data scheduling method according to an embodiment of the present application. FIG. 8 can be regarded as a specific example of the method shown in FIG. 2 . The steps in Figure 8 are described below.

S801. The receiving unit sends the received data stream to the identification module.

That is to say, after receiving the data stream from the downlink device, the receiving unit sends it to the identification module.

This step can be regarded as an example of S201.

S802. The identifying module identifies the source of the data stream.

This step can be regarded as an example of S202.

S803. The identifying module sends the data stream whose source has been identified to the marking module.

S804. The marking module marks the high-priority data flow.

That is to say, the marking module marks the data flow satisfying the preset priority range.

For example, according to the identification result of the identification module, the marking module adds a characteristic mark X to the data streams of high-priority services such as game services and payment services.

S805. The marking module sends the marked data stream to the scheduling module.

It may be to send only the marked data flow, or it may be to send all the data flow, but it contains the marked data flow.

S806. The measurement module measures the network quality of the network path.

Take network access including mobile networks and Ethernet as an example. The measurement module can use the method shown in FIG. 6 to obtain the network quality of the mobile network path, and can also use the method shown in FIG. 7 to obtain the network quality of the Ethernet path. When multiple network paths are all mobile networks or Ethernets, the network quality of multiple network paths is measured by using the method shown in FIG. 6 or FIG. 7 . When multiple network paths include mobile network and Ethernet at the same time, the method in Figure 6 is used to measure the network quality of the mobile network path, and the method in Figure 7 is used to measure the network quality of the Ethernet path.

S807. The scheduling module receives the network quality sent by the measurement module.

Take network access including mobile networks and Ethernet as an example. In one example, the measurement module comprehensively compares the network quality of the mobile network and the Ethernet according to the network measurement results, selects the target network path, that is, the path with better network quality, and reports it to the scheduling module, that is to say, the measurement module What is sent is related information such as the network quality that has been determined to be the target network path. In another example, the measurement module sends the quality of all network paths for the scheduling module to determine the target network path according to the network quality of all network paths.

S808. The scheduling module updates the target network path.

That is to say, the dispatching module can delete the old routing path configuration information and create new routing path configuration information according to the reporting result of the measuring module, and the new routing path is the target network path.

It can be seen that S806-S808 is not affected by whether any step in steps S801-S805 is executed.

S809. The scheduling module schedules the marked data flow to the target network path.

For example, the data flow marked with X is dispatched to the target network path.

S803-S809 can be regarded as an example of S203.

The above mainly introduces the data scheduling method in the embodiment of the present application with reference to the accompanying drawings. It should be understood that although the various steps in the flowcharts involved in the above-mentioned embodiments are shown sequentially, these steps are not necessarily executed sequentially in the order shown in the figure. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flow charts involved in the above-mentioned embodiments may include multiple steps or stages, and these steps or stages are not necessarily executed at the same time, but may be performed at different times For execution, the execution order of these steps or stages is not necessarily performed sequentially, but may be executed in turn or alternately with other steps or at least a part of steps or stages in other steps. The data scheduling device according to the embodiment of the present application will be introduced below with reference to the accompanying drawings.

Fig. 9 is a schematic diagram of a data scheduling device according to an embodiment of the present application. As shown in FIG. 9 , the apparatus 1000 includes a receiving unit 1100 and a processing unit 1200 . The apparatus 1000 may be integrated in a network device, or may be a network device. The network device may be, for example, a wireless router supporting multiple networks.

In an implementation manner, the apparatus 1000 may be the network device 110 in FIG. 1 , or be set in the network device 110 .

The apparatus 1000 can be used to implement any one of the above data scheduling methods. For example, the receiving unit 1100 can be used to perform step S201, and the processing unit 1200 can be used to perform steps S202 and S203. For another example, the processing unit 1200 may also be configured to execute each step in FIG. 3 , FIG. 5 , FIG. 6 or FIG. 7 . For another example, the receiving unit 1100 may be the receiving unit in FIG. 8 and may be configured to execute step S801, and the processing unit 1200 may be configured to execute steps S802-S809.

As shown in FIG. 9 , the processing unit 1200 may include an identification module 1210 and a scheduling module 1220 . The identifying module 1210 may be the aforementioned identifying module (SA module), configured to identify the source of the data flow. The identification module 1210 can be used to execute step S202, and S802, S803.

The scheduling module 1220 is configured to schedule the data flow according to the source of the data flow, for example, steps S203, S807-S809 may be executed.

In an implementation manner, the processing unit 1200 further includes a measurement module 1230, which is connected to the scheduling module 1220 and used to measure the network quality of each network path. The measurement module 1230 may be the measurement module shown in FIG. 8 , and may be used to execute the steps in FIG. 6 and FIG. 7 , and may also be used to execute S806 and S807.

In one example, the measurement module 1230 includes a mobile network measurement module 1231 and an Ethernet measurement module 1232, wherein the mobile network measurement module 1231 is used to measure the network quality of the mobile network path, and the Ethernet measurement module 1232 is used to measure the network quality of the Ethernet path. quality.

The mobile network measurement module 1231 may be used to execute various steps shown in FIG. 6 .

As shown in FIG. 9 , the mobile network measurement module 1231 may include four submodules from submodule 1 to submodule 4 . The four sub-modules may respectively correspond to the modules in FIG. 6 : modem module, AT module, rate statistics module and quality measurement module.

The Ethernet metric module 1232 may be used to execute various steps shown in FIG. 7 .

As shown in FIG. 9 , the Ethernet measurement module 1232 may include three submodules from a submodule 5 to a submodule 7 . The three sub-modules may respectively correspond to the modules in FIG. 7 : a signal quality collection module, a rate statistics module and a quality measurement module.

In one implementation manner, the processing unit 1200 further includes a marking module 1240 connected to the scheduling module 1220 and configured to mark the data flow, for example, may be used to execute steps S803, S804 and S805. The marking module 1240 may be the marking module in FIG. 8 . The marking module 1240 can also be used to execute various steps shown in FIG. 3 and FIG. 5 .

In an implementation manner, the processing unit 1200 may also be configured to perform the following operation: dispatch the second data flow in the at least one data flow to a non-target network path, and the second data flow is used to indicate that the at least one data flow except the first A data flow other than a data flow, the non-target network access is used to indicate a network access other than the target network access among the plurality of network accesses.

In an implementation manner, the apparatus 1000 may further include a storage unit, configured to store data such as a data stream. The storage unit may be integrated in the processing unit 1200 , or may be a unit independent of the receiving unit 1100 and the processing unit 1200 .

It should be noted that the information interaction and execution process between the above-mentioned devices/units are based on the same concept as the method embodiment of the present application, and its specific functions and technical effects can be found in the method embodiment section. I won't repeat them here.

Those skilled in the art can clearly understand that for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used for illustration. In practical applications, the above-mentioned functions can be assigned to different functional units, Completion of modules means that the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment may be integrated into one processing unit, or each unit may exist separately physically, or two or more units may be integrated into one unit, and the above-mentioned integrated units may adopt hardware It can also be implemented in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the present application. For the specific working processes of the units and modules in the above system, reference may be made to the corresponding processes in the aforementioned method embodiments, and details will not be repeated here.

The embodiment of the present application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and operable on the at least one processor, and the processor executes The computer program implements the steps in any of the above method embodiments.

The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in each of the foregoing method embodiments can be realized.

An embodiment of the present application provides a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal can implement the steps in the foregoing method embodiments when executed.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the procedures in the methods of the above embodiments in the present application can be completed by instructing related hardware through computer programs, and the computer programs can be stored in a computer-readable storage medium. The computer program When executed by a processor, the steps in the above-mentioned various method embodiments can be realized. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device capable of carrying computer program codes to a photographing device/terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (random access memory, RAM), electrical carrier signals, telecommunication signals, and software distribution media. Such as U disk, mobile hard disk, magnetic disk or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunication signals under legislation and patent practice.

In the above-mentioned embodiments, the descriptions of each embodiment have their own emphases, and for parts that are not detailed or recorded in a certain embodiment, refer to the relevant descriptions of other embodiments.

Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present application.

In the embodiments provided in this application, it should be understood that the disclosed device/network device and method may be implemented in other ways. For example, the device/network device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units Or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and/or components, but does not exclude one or more other Presence or addition of features, wholes, steps, operations, elements, components and/or collections thereof.

It should also be understood that the term "and/or" used in the description of the present application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

As used in this specification and the appended claims, the term "if" may be construed, depending on the context, as "when" or "once" or "in response to determining" or "in response to detecting ". Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be construed, depending on the context, to mean "once determined" or "in response to the determination" or "once detected [the described condition or event] ]” or “in response to detection of [described condition or event]”.

In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third" and so on are only used to distinguish descriptions, and should not be understood as indicating or implying relative importance.

Reference to "one embodiment" or "some embodiments" or the like in the specification of the present application means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in other embodiments," etc. in various places in this specification are not necessarily All refer to the same embodiment, but mean "one or more but not all embodiments" unless specifically stated otherwise. The terms "including", "comprising", "having" and variations thereof mean "including but not limited to", unless specifically stated otherwise.

The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still implement the foregoing embodiments Modifications to the technical solutions described in the examples, or equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the application, and should be included in the Within the protection scope of this application.

Claims (19)

1. A data scheduling method, applied to network equipment, is characterized in that, comprising: receiving at least one data stream; identifying a source of each of the at least one data stream; According to the source of each data flow in the at least one data flow, dispatching the first data flow in the at least one data flow to the target network path, the source of the first data flow satisfies a preset priority range, The target network path is a network path with the best network quality among the multiple network paths of the network device. 2. The method according to claim 1, wherein the identifying the source of each data flow in the at least one data flow comprises: The at least one data flow is processed by using the identification model to obtain the source of each data flow in the at least one data flow, and the source includes at least one of a service source or an application program source. 3. The method according to claim 1, wherein, according to the source of each data flow in the at least one data flow, scheduling the first data flow in the at least one data flow to the target network path on, including: judging whether the source of each data flow in the at least one data flow meets the preset priority range, and determining the data flow that meets the preset priority range as the first data flow; marking the first data stream; Scheduling the tagged first data flow onto the target network path. 4. The method according to claim 3, wherein the marking the first data stream comprises: parsing the IP information of each data flow in the at least one data flow, and converting the IP information into five-tuple information; According to the five-tuple information, the first data stream is determined from the at least one data stream and marked. 5. The method according to any one of claims 1 to 4, characterized in that the method further comprises: Determine the network quality of the multiple network paths according to the quality parameters of the multiple network paths, the quality parameters include at least one of signal quality indicators, rate or delay, the signal quality indicators and the rate are respectively positively correlated with the network quality, and the time delay is negatively correlated with the network quality. 6. The method according to claim 5, wherein the determining the network quality of the plurality of network paths according to the quality parameters of the plurality of network paths comprises: Obtaining network quality metric values of the multiple network paths in the multiple time periods according to the quality parameters of the multiple network paths in the multiple time periods in the first time interval; Taking the average or weighted average of the network quality metric values of the multiple network paths in the multiple time periods as the network quality of the multiple network paths in the first time interval. 7. The method according to any one of claims 1 to 4, wherein the plurality of network accesses include mobile network accesses and Ethernet accesses. 8. The method according to any one of claims 1 to 4, wherein the method further comprises: scheduling a second data flow in the at least one data flow on a non-target network path, the second data flow is used to represent a data flow in the at least one data flow except the first data flow, the The non-target network accesses are used to represent network accesses in the plurality of network accesses except the target network accesses. 9. A data scheduling device, set in a network device, characterized in that it comprises: a receiving unit, configured to receive at least one data stream; a processing unit configured to identify a source of each of the at least one data stream; The processing unit is further configured to, according to the source of each data flow in the at least one data flow, schedule a first data flow in the at least one data flow to a target network path, and the first data flow of the first data flow The source satisfies a preset priority range, and the target network path is a network path with the best network quality among the multiple network paths of the network device. 10. The device according to claim 9, wherein the processing unit is specifically configured to: The at least one data flow is processed by using the identification model to obtain the source of each data flow in the at least one data flow, and the source includes at least one of a service source or an application program source. 11. The device according to claim 9, wherein the processing unit is specifically configured to: determine whether the source of each data stream in the at least one data stream satisfies the preset priority range, and will satisfy The data flow in the preset priority range is determined as the first data flow; the first data flow is marked; and the first data flow with the mark is scheduled on the target network path. 12. The device according to claim 11, wherein the processing unit is specifically configured to: parsing the IP information of each data flow in the at least one data flow, and converting the IP information into five-tuple information; According to the five-tuple information, the first data stream is determined from the at least one data stream and marked. 13. The device according to any one of claims 9 to 12, wherein the processing unit is further configured to: Determine the network quality of the multiple network paths according to the quality parameters of the multiple network paths, the quality parameters include at least one of signal quality indicators, rate or delay, the signal quality indicators and the rate are respectively positively correlated with the network quality, and the time delay is negatively correlated with the network quality. 14. The device according to claim 13, wherein the processing unit is specifically configured to: Obtaining network quality metric values of the multiple network paths in the multiple time periods according to the quality parameters of the multiple network paths in the multiple time periods in the first time interval; Taking the average or weighted average of the network quality metric values of the multiple network paths in the multiple time periods as the network quality of the multiple network paths in the first time interval. 15. The apparatus according to any one of claims 9 to 12, wherein the plurality of network accesses comprises a mobile network access and an Ethernet access. 16. The device according to any one of claims 9 to 12, wherein the processing unit is further configured to: scheduling a second data flow in the at least one data flow on a non-target network path, the second data flow is used to represent a data flow in the at least one data flow except the first data flow, the The non-target network accesses are used to represent network accesses in the plurality of network accesses except the target network accesses. 17. A network device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, characterized in that, when the processor executes the computer program, the computer program according to claim The method described in any one of 1 to 8. 18. The network device according to claim 17, wherein the network device is a wireless router. 19. A computer-readable storage medium, the computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented .

Images