CN118353850A - Data packet scheduling method and related device - Google Patents

Abstract

The embodiment of the present application discloses a data packet scheduling method and related devices. If a target data packet to be allocated is received, the window parameters of the congestion window of each network link are obtained. The congestion window is a variable of the transmission protocol used in the transmission process, which can adapt to the network state of the network link. The routing probabilities corresponding to each network link are determined according to the window parameters of the congestion window, that is, the probability of each network link being distributed with the target data packet is determined according to the window parameters of the congestion window. The larger the window parameter of the congestion window, the better the network state of the corresponding network link, so the corresponding routing probability is greater, so as to allocate the target data packet to the network link according to the routing probability, so that the network link with a larger routing probability has more opportunities to be distributed with the target data packet, so that the data packet can be distributed to the network link with a better network condition with a greater probability, thereby improving the scheduling effect of the data packet as a whole, and then improving the transmission speed of the data packet.

Description

A data packet scheduling method and related device

Technical Field

The present application relates to the field of computer technology, and in particular to a data packet scheduling method and related devices.

Background technique

With the development of computer technology, access between devices can be achieved through devices including multiple network access methods. For example, mobile Internet users can access various applications in cloud servers through terminal devices including mobile hotspots (Wi-Fi), Long Term Evolution (LTE) and the fifth generation mobile communication technology (5G).

In the process of access, network links are generally used for data transmission. However, the bandwidth of a single network link is limited, and the delay is unstable, resulting in a slow data transmission speed. Based on this, multiple network links are generally used for data transmission in related technologies. For example, a video is packaged into multiple data packets, and the multiple data packets are transmitted separately through multiple network links, thereby improving the data transmission speed.

However, the scheduling method of multi-path data packets in the related art is generally applicable to wired networks. However, due to the heterogeneity and dynamic characteristics of wireless networks or mobile networks, the scheduling method of multi-path data packets in the related art cannot be applied to wireless networks or mobile networks.

Summary of the invention

In order to solve the above technical problems, the present application provides a data packet scheduling method and related devices, which are used to solve the problem that the multi-path data packet scheduling method cannot be applied to wireless networks or mobile networks.

The embodiments of the present application disclose the following technical solutions:

On the one hand, an embodiment of the present application provides a data packet scheduling method, the method comprising:

If a target data packet to be allocated is received, then window parameters of congestion windows corresponding to a plurality of network links are obtained, wherein the network links are channels for transmitting the target data packet, and the window parameters of the congestion windows are used to describe the network status of the corresponding network links;

Determine, according to the window parameters of the multiple congestion windows, the routing probabilities corresponding to the multiple network links respectively, wherein the routing probabilities are used to describe the probabilities of the corresponding network links being distributed with the target data packets; wherein the larger the window parameter of the congestion window is, the larger the corresponding routing probability is;

Distribute the target data packet to the plurality of network links according to the plurality of routing probabilities.

On the other hand, an embodiment of the present application provides a data packet scheduling device, the device comprising: an acquisition unit, a determination unit, and a sending unit;

The acquisition unit is used to acquire window parameters of congestion windows corresponding to a plurality of network links respectively if receiving a target data packet to be allocated, wherein the network link is a channel for transmitting the target data packet, and the window parameters of the congestion window are used to describe the network status of the corresponding network link;

The determining unit is used to determine the routing probabilities corresponding to the plurality of network links respectively according to the window parameters of the plurality of congestion windows, wherein the routing probabilities are used to describe the probabilities of the corresponding network links being distributed with the target data packets; wherein the larger the window parameter of the congestion window is, the larger the corresponding routing probability is;

The sending unit is used to distribute the target data packet to the multiple network links according to the multiple routing probabilities.

On the other hand, an embodiment of the present application provides a computer device, the computer device comprising a processor and a memory:

The memory is used to store a computer program and transmit the computer program to the processor;

The processor is configured to execute the method described in the above aspects according to the instructions in the computer program.

On the other hand, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method described in the above aspects.

On the other hand, an embodiment of the present application provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method described in the above aspects.

It can be seen from the above technical solution that if a target data packet to be allocated is received, the target data packet needs to be distributed to the network link for transmission, and the window parameters of the congestion window of each network link are obtained at this time. Among them, the congestion window is a variable of the transmission protocol used in the transmission process, which can adapt to the network state of the network link, that is, the network state of each network link can be fully reflected through the window parameters of the congestion window. According to the window parameters of the congestion window, the routing probabilities corresponding to each network link are determined, that is, the probability of each network link being distributed with the target data packet is determined according to the window parameters of the congestion window. Among them, the larger the window parameter of the congestion window, the better the network state of the corresponding network link, so the corresponding routing probability is greater, so as to allocate the target data packet to the network link according to the routing probability, so that the network link with a larger routing probability has more opportunities to be distributed with the target data packet. Thus, the routing probability is determined by the window parameters of the congestion window, and the data packet to be allocated is distributed according to the routing probability, so that the data packet can be distributed to the network link with a better network condition with a greater probability, thereby improving the scheduling effect of the data packet as a whole, and then improving the transmission speed of the data packet. In addition, since the window parameters of the congestion window are adjusted in real time according to the network status of the network link, it is possible to more accurately detect changes in the network status of the wireless link, and the routing probability of each network link is calculated separately, effectively adapting to the heterogeneity and dynamics of mobile networks or wireless networks.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

FIG1 is a schematic diagram of an application scenario of a data packet scheduling method provided in an embodiment of the present application;

FIG2 is a schematic diagram of a flow chart of a data packet scheduling method provided in an embodiment of the present application;

FIG3 is a schematic diagram of an application scenario of a data packet scheduling method provided in an embodiment of the present application;

FIG4 is a schematic diagram of a data packet scheduling system provided in an embodiment of the present application;

FIG5 is a flow chart of a method for scheduling data packets provided in an embodiment of the present application;

FIG6 is a flow chart of a method for scheduling data packets provided in an embodiment of the present application;

FIG7 is a schematic diagram of the structure of a data packet scheduling device provided in an embodiment of the present application;

FIG8 is a schematic diagram of the structure of a server provided in an embodiment of the present application;

FIG. 9 is a schematic diagram of the structure of a terminal device provided in an embodiment of the present application.

Detailed ways

The embodiments of the present application are described below in conjunction with the accompanying drawings.

Multiple network links generally have different parameters, such as delay, bandwidth, and jitter, which make multiple network links have different transmission capabilities, or the network status of each network link is different, that is, wireless networks or mobile networks have heterogeneous characteristics, and due to the instability of the network status, the network links will change dynamically at different times, that is, wireless networks or mobile networks have dynamic characteristics. In addition, wireless network links are also affected by the strength of wireless network signals and are susceptible to interference from surrounding media. In a mobile environment, the signal strength has great uncertainty due to changes in the distance from the base station or hotspot, so it is more unstable than the wired link state. Therefore, the scheduling method for multi-path data packets in the related art that is applicable to wired networks cannot be applied to wireless networks or mobile networks.

Based on this, the embodiments of the present application provide a data packet scheduling method and related devices, which determine the routing probability through the window parameters of the congestion window, and distribute the data packets to be allocated according to the routing probability. Since the window parameters of the congestion window are adjusted in real time according to the network status of the network link, it is possible to more accurately detect changes in the network status of the wireless link, and the routing probability of each network link is calculated separately, which effectively adapts to the heterogeneity and dynamics of mobile networks or wireless networks.

The data packet scheduling method provided in the embodiment of the present application can be applied to data packet scheduling devices with data processing capabilities, such as terminal devices and servers. Among them, terminal devices include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, aircraft, etc. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal and the server can be directly or indirectly connected via wired or wireless communication, and this application is not limited here. The embodiments of the present application can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, assisted driving, etc.

The data packet scheduling method provided in the embodiment of the present application is implemented based on cloud computing technology. Cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and network in a wide area network or a local area network to realize data calculation, storage, processing, and sharing.

Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, application technology, etc. based on the cloud computing business model. It can form a resource pool, which is used on demand and is flexible and convenient. Cloud computing technology will become an important support. The backend services of the technical network system require a large amount of computing and storage resources, such as video websites, picture websites and more portal websites. With the rapid development and application of the Internet industry, each item may have its own identification mark in the future, and all need to be transmitted to the backend system for logical processing. Data of different levels will be processed separately. All kinds of industry data need strong system backing support, which can only be achieved through cloud computing.

In an embodiment of the present application, the data packet scheduling method can be applied to scenarios with high requirements for bandwidth, latency and stability, such as online education, e-commerce live broadcast, cloud gaming, etc. Among them, cloud gaming (Cloud gaming), which can also be called gaming on demand, is an online gaming technology based on cloud computing technology. Cloud gaming technology enables thin clients with relatively limited graphics processing and data computing capabilities to run high-quality games. In a cloud gaming scenario, the game is not run on the player's game terminal, but on a cloud server, and the cloud server renders the game scene into a video and audio stream, which is transmitted to the player's game terminal through the network. The player's game terminal does not need to have powerful graphics computing and data processing capabilities, but only needs to have basic streaming media playback capabilities and the ability to obtain player input instructions and send them to the cloud server.

To facilitate understanding of the data packet scheduling method provided in the embodiment of the present application, the application scenario of the data packet scheduling method is executored as a server as an example to executor of the data packet scheduling method.

Referring to FIG. 1 , this figure is a schematic diagram of an application scenario of the data packet scheduling method provided in an embodiment of the present application. As shown in FIG. 1 , the application scenario includes multiple terminal devices 110 and a server 120. Among them, the terminal device 110 can be installed with a video source client, so as to upload a video to the server 120 through the video source client. The server 120, as an ingest server, can receive and store the video uploaded by the terminal device 110.

Taking one of the multiple terminal devices 110 as an example, after receiving the target data packet to be distributed, the terminal device 110 needs to upload the target data packet to the server 120. Since there are two network links that can upload the target data packet, before uploading the target data packet, the terminal device 110 needs to determine which network link to upload through to ensure transmission performance.

Among them, the two network links are network link 1: Wi-Fi path and network link 2: LTE/5G path. After receiving the target data packet, the terminal device 110 obtains the window parameters of the congestion windows corresponding to the two network links, respectively, and the window parameters are used to describe the network status of the corresponding network links. For example, the window parameter of the congestion window of network link 1 is 4096 bytes, and the window parameter of the congestion window of network link 2 is 2048 bytes.

The terminal device 110 determines the routing probabilities corresponding to each network link according to the window parameters of the congestion window. For example, the routing probability of network link one is 0.7, and the routing probability of network link two is 0.3. It should be noted that the congestion window is a variable of the transmission protocol used in the transmission process, which can adapt to the network status of the network link. Therefore, the larger the window parameter of the congestion window, the better the network status of the corresponding network link, and the larger the corresponding routing probability, so the terminal device 110 allocates the target data packet to the network link according to the routing probability, so that the network link with a larger routing probability has more opportunities to distribute the target data packet. For example, in the application scenario of Figure 1, the terminal device 110 sends the target data packet to the server 120 through network link one.

Therefore, the routing probability is determined by the window parameters of the congestion window, and the data packets to be allocated are distributed according to the routing probability, so that the data packets can be distributed to the network links with better network conditions with a greater probability, thereby improving the overall scheduling effect of the data packets and further improving the transmission speed of the data packets. In addition, since the window parameters of the congestion window are adjusted in real time according to the network status of the network link, the network status changes of the wireless link can be detected more accurately, and the routing probability of each network link is calculated separately, which effectively adapts to the heterogeneity and dynamics of the mobile network or wireless network.

The data packet scheduling method provided in the embodiment of the present application can be executed by the server. However, in other embodiments of the present application, the terminal device can also have similar functions as the server, thereby executing the data packet scheduling method provided in the embodiment of the present application, or the terminal device and the server jointly execute the data packet scheduling method provided in the embodiment of the present application, which is not limited in the present embodiment.

The data packet scheduling method provided by the present application is described in detail below through a method embodiment.

See Figure 2, which is a flow chart of the data packet scheduling method provided in the embodiment of the present application. For ease of description, the following embodiment is still introduced by taking the execution subject of the data packet scheduling method as a terminal device as an example. As shown in Figure 2, the data packet scheduling method includes the following steps:

S201: If a target data packet to be allocated is received, then window parameters of congestion windows corresponding to a plurality of network links are obtained.

In actual applications, the terminal device as the sending end generally has multiple ports, each port is connected to a network link. For example, the Wi-Fi port and the Subscriber Identity Module (SIM card) port in the mobile phone are all connected to the corresponding network link. Since multiple network links generally have different parameters, multiple network links have different transmission capabilities, and the network links will change dynamically at different times, that is, wireless networks or mobile networks have the characteristics of heterogeneity and dynamism. In order to ensure data transmission performance, it is necessary to select a network link with a better network status after obtaining the data packet. The network status generally refers to the situation of the network uploading data packets, whether it is blocked, etc.

The target data packet among multiple data packets is taken as an example below. The target data packet is a data packet to be allocated a network link. After the network link is allocated, it is sent to a terminal device or server as a receiving end through the network link. The network link is a channel for transmitting the target data packet.

In the related art, the network status is generally determined based on changes in network characteristics such as bandwidth, round trip time (RTT) and packet loss. However, network characteristics such as bandwidth, RTT and packet loss can only reflect one aspect of the network status. For example, based on RTT, only the overall delay can be minimized, and the bandwidth cannot be adjusted, that is, the network status cannot be fully reflected. Moreover, it is difficult to accurately estimate the situation of a wireless link that fluctuates violently. For example, when the network signal is poor or severely blocked, the changes in various network characteristics are very large, and the current network status cannot be accurately estimated based on only one network characteristic.

Based on this, the embodiment of the present application uses the window parameters of the congestion window of the network link to determine the network state. Among them, the congestion window refers to how many data packets can be sent in a unit of time. It is a congestion avoidance window and a sliding window located at the sending end. The size of the congestion window depends on the degree of congestion of the network and changes dynamically. The principle of the sending end to control the congestion window is: as long as there is no congestion in the network, the congestion window will be increased to send more data packets. However, as long as congestion occurs in the network, the congestion window will be reduced. Therefore, the window parameters of the congestion window can reflect the size of the congestion window and can describe the network state of the corresponding network link.

It should be noted that the congestion window is a variable of the transmission protocol used in the transmission process, which can adapt to the network status of the network link, that is, the window parameters of the congestion window can fully reflect the network status of each network link. The network characteristic is an indicator of the network status and can only reflect one aspect of the network status. Therefore, the network status determined by the window parameters of the congestion window is more accurate.

S202: Determine routing probabilities corresponding to a plurality of network links respectively according to window parameters of a plurality of congestion windows.

As can be seen from the foregoing, the window parameter of the congestion window can reflect the size of the congestion window and can describe the network status of the corresponding network link. For example, the window value of the congestion window adjusted for the last time can be used as the window parameter, and the average value of the window value of the congestion window adjusted multiple times within a period of time can also be used as the window parameter, etc., and the embodiments of the present application do not specifically limit this.

Among them, the window parameter of the congestion window can fully reflect the network status of the corresponding network link. The larger the window parameter of the congestion window, the better the network status of the corresponding network link, making the corresponding routing probability larger, so that the probability of the corresponding network link being distributed with the target data packet can be described through the routing probability. Then, according to the routing probability of each network link, the target data packet is distributed to each network link.

The embodiment of the present application does not specifically limit the manner of determining the routing probability according to the window parameter. For example, the routing probability of the jth network link among m network links can be expressed as p _j ^r , where j=1, 2, ..., m. Then the normalization of the routing probability can be expressed as the following formula:

p _j ^r = cwnd _j ^T /∑ _u cwnd _u ^T

Among them, ∑ _j p _j ^r =1, u =1,2,…,m, cwnd _j ^T represents the window parameter of the j-th network link, and ∑ _u cwnd _u ^T represents the sum of the window parameters of m network links.

In addition, since the window parameters of the congestion window are adjusted in real time according to the network status of the network link, it is possible to more accurately detect changes in the network status of the wireless link, and the routing probabilities of each network link are calculated separately without affecting each other, thereby making full use of multiple network links at the sender to send data packets, thereby improving the overall transmission speed of data packets.

S203: Distribute the target data packet to multiple network links according to multiple routing probabilities.

The routing probability is used to describe the probability of the corresponding network link being distributed with the target data packet, so the greater the routing probability, the greater the probability of distributing the target data packet through the corresponding network link. As a possible implementation method, the target data packet can be sent to the network link with the highest routing probability, or the target data packet can be randomly sent to the network links with the top routing probabilities, etc. This application does not make specific limitations on this, and several methods will be used as examples to illustrate it later, which will not be repeated here.

It can be seen from the above technical solution that if a target data packet to be allocated is received, the target data packet needs to be distributed to the network link for transmission, and the window parameters of the congestion window of each network link are obtained at this time. Among them, the congestion window is a variable of the transmission protocol used in the transmission process, which can adapt to the network state of the network link, that is, the network state of each network link can be fully reflected through the window parameters of the congestion window. According to the window parameters of the congestion window, the routing probabilities corresponding to each network link are determined, that is, the probability of each network link being distributed with the target data packet is determined according to the window parameters of the congestion window. Among them, the larger the window parameters of the congestion window, the better the network state of the corresponding network link, so the corresponding routing probability is larger, so as to allocate the target data packet to the network link according to the routing probability, so that the network link with the larger routing probability has more opportunities to be distributed with the target data packet.

Therefore, the routing probability is determined by the window parameters of the congestion window, and the data packets to be allocated are distributed according to the routing probability, so that the data packets can be distributed to the network links with better network conditions with a greater probability, that is, the current more suitable network links are selected to send the target data packets, thereby improving the overall scheduling effect of the data packets, and then improving the transmission speed of the data packets. In addition, since the window parameters of the congestion window are adjusted in real time according to the network status of the network link, it is possible to more accurately detect the changes in the network status of the wireless link, and the routing probability of each network link is calculated separately, which effectively adapts to the heterogeneity and dynamics of the mobile network or wireless network.

As a possible implementation method, the embodiment of the present application provides a specific implementation method of S203, that is, a specific implementation method of distributing target data packets to multiple network links according to multiple routing probabilities, see A1-A4 for details.

A1: Construct m continuous accumulation intervals according to multiple routing probabilities.

Among them, the starting value of the i-th accumulation interval corresponding to the j-th network link is the sum of the routing probabilities of the previous j-1 network links, and the ending value of the i-th accumulation interval is the sum of the routing probabilities of the previous j network links. Both i and j are positive integers less than or equal to m.

For example, if the number of multiple network links is m, m continuous accumulation intervals can be constructed according to the m routing probabilities. If the sum of the routing probabilities of the m network links is 1, the constructed m continuous accumulation intervals can be expressed as [0, p ₁ ^r ), [p ₁ ^r , p ₁ ^r + p ₂ ^r ), …, [∑ _j＝1 ^m-1 p _j ^r , 1] (∑ _j＝1 ^m p _j ^r ＝1). Among them, the starting value of the first accumulation interval corresponding to the first network link is 0, that is, the sum of the routing probabilities of the first 0 network links is 0, and the ending value of the first accumulation interval corresponding to the first network link is p ₁ ^r , that is, the routing probability of the first network link. The starting value of the second accumulation interval corresponding to the second network link is p ₁ ^r , that is, the sum of the routing probabilities of the first network link is p ₁ ^r , and the ending value of the second accumulation interval corresponding to the second network link is p ₁ ^r +p ₂ ^r , that is, the sum of the routing probabilities of the first two network links. And so on, no further details are given here.

A2: Generate random numbers.

Among them, the random number is greater than or equal to the starting value of the first cumulative interval, and the random number is less than or equal to the ending value of the mth cumulative interval, that is, a random number is randomly generated in the interval covered by the m cumulative intervals, such as [0,1].

A3: The network link corresponding to the accumulation interval into which the random number falls is determined as the network link to be sent.

For example, a random number x between [0, 1] is generated and the accumulation interval to which it belongs is observed. For x∈[∑ _j＝1 ^v-1 p _j ^r ,∑ _j＝1 ^v p _j ^r ), the vth network link is selected as the network link to be sent, for sending the target data packet to be allocated. For another example, if the generated random number falls into the third accumulation interval, the third network link is the network link to be sent.

A4: Distribute the target data packets to the network links to be sent.

It can be seen from the above technical solution that by constructing the cumulative interval corresponding to each network link according to the routing probability, the network link corresponding to the cumulative interval in which the generated random number falls is used as the network link to be sent, and then the target data packet is sent through the network link to be sent. Therefore, compared with sending the data packet through the network link with the highest routing probability, the network link with a smaller routing probability can be used to send the target data packet through the random number method, thereby making full use of each network link, avoiding idle network links, and making the data packet sent according to the routing probability distribution of each link as a whole. In addition, since a decision can be made in real time for each data packet to be allocated, and the network link selected as a whole can be made to conform to the routing probability distribution of each network link.

As a possible implementation method, the network link corresponding to the largest routing probability among multiple routing probabilities can also be determined as the network link to be sent, and the network link to be sent is the network link with the best network status among the multiple network links. The target data packet is distributed to the network link to be sent, so that the target data packet is sent through the network link with the best network status, so that the target data packet is sent faster.

As a possible implementation method, the embodiment of the present application also provides two methods for determining window parameters, which are described below respectively.

Method 1: Use the last window value as the window parameter.

B1: If the window value of the congestion window identifying the target network link changes, the updated window value and update time are recorded.

As mentioned above, the window value of the congestion window, that is, the size of the congestion window, will change with the change of the network status. In order to be able to use it later, it is necessary to record each change of the window value, that is, the updated window value and the update time. The following is an example of one network link among multiple network links, that is, the target network link.

It should be noted that the updated window value and update time can be obtained according to the transmission protocol used between the receiving end and the sending end, and this application does not make specific limitations on this.

B2: Determine the updated window value corresponding to the update time closest to the current time as the window parameter of the congestion window corresponding to the target network link.

According to the update time of the target network link, an updated window value corresponding to the update time closest to the current time is determined, that is, the current latest update time and the current latest congestion window value, and the current latest congestion window value is determined as the window parameter of the congestion window corresponding to the target network link.

B3: Taking multiple network links as target network links respectively, and obtaining window parameters of congestion windows corresponding to the multiple network links respectively.

Therefore, by taking each network link as the target network link, the window parameter of the congestion window of each network link can be determined through B1-B2.

It can be seen from the above technical solution that the window value and update time after each update are recorded, the window value of the last update is determined according to the update time, and the window value is determined as the window parameter of the congestion window corresponding to the target network link. Thus, the network status of each network link can be continuously determined by continuously using the latest window value each time as the window parameter of the congestion window, thereby improving the real-time performance of the allocation of the data packets to be allocated.

Method 2: The average value of the window value within a period of time is used as the window parameter.

C1: If the window value of the congestion window identifying the target network link changes, record the updated window value and update time.

As mentioned above, the window value of the congestion window, that is, the size of the congestion window, will change with the change of the network status. In order to be able to use it later, it is necessary to record each change of the window value, that is, the updated window value and the update time. The following is an example of one network link among multiple network links, that is, the target network link.

As a possible implementation, since some network links are updated slowly, the window value of the initial congestion window corresponding to each network link can be recorded, which can be expressed as cwnd _j ⁰ , where j = 1, 2, ..., m, and m is the number of multiple network links. When the window value of the congestion window of the j-th network link changes, if the window value received by the receiving end is different from the window value used when sending, it is considered that the window value of the congestion window has changed. The i-th updated congestion window of the j-th network link can be expressed as cwnd _j ⁱ , and correspondingly, the update time of the i-th updated j-th network link can be expressed as t _j ⁱ .

C2: Determine the window value of the congestion window of the target network link during the link influence period according to the update time.

As a possible implementation manner, the window parameters may be determined in advance or when needed. For example, when a target data packet is waiting to be allocated as a target data packet at time t _c , the window value of the congestion window of each network link is searched within the link impact period _Ta , which can be expressed as cwnd _j ^k , where t _j ^k ∈[t _c -T _a ,t _c ], i.e., the window value of the congestion window updated within the link impact period is searched.

The link impact period is a pre-set time period. Through a large number of experiments, it is found that when the link impact period is 100 milliseconds, the aggregate throughput of multiple network links can be maximized.

C3: Determine a window parameter of the congestion window of the target network link according to the window value of the congestion window of the target network link during the link influence period.

It can be seen from the above technical solution that the window parameters of the congestion window of each network link are determined based on the window value of the congestion window within a period of time, that is, the link impact period. In this way, the influence of emergencies can be avoided, so as to more accurately determine the overall status of each network link within a short period of time (that is, within the link impact period), and can also be used to compare the network status between different network links, and then determine the network link used to distribute the target data packet.

The embodiment of the present application does not specifically limit the specific method of determining the window parameter of the congestion window of the target network link according to the window value of the congestion window of the target network link during the link influence period. Two methods are used as examples for explanation below.

The first method is to determine the window parameters by arithmetic mean.

D1: Determine the arithmetic mean of the window values of the congestion window of the target network link during the link impact period.

D2: Determine the arithmetic mean as the window parameter of the congestion window of the target network link.

Therefore, by taking the average value of the window value of the congestion window within a period of time, that is, the link impact period, as the window parameter of the congestion window, emergencies can be avoided, thereby improving the stability of the network state.

The second method is to determine the window parameters by weighted harmonic mean.

E1: Determine the number of window adjustments of the target network link during the link influence period according to the window value of the congestion window of the target network link during the link influence period.

Since each network link has different transmission capabilities, or the network states of each network link are different, the window value of each network link may be different, and the number of window adjustments of each network link during the link influence period may also be different.

It should be noted that although multiple updated window values are within the link impact period, the degree of influence of each window value on the network state is different. For example, the window value whose update time is closer to the current one has a greater influence on the network state, and the window value whose update time is farther from the current one has a smaller influence on the network state. Therefore, in order to improve the accuracy of the window parameters, different weights are assigned to window values with different update times by means of weighted harmonic mean.

E2: Get the difference between the target update time and the time when the target data packet is waiting for distribution.

The target update time is one of the update times of the target network link during the link influence period.

E3: Obtain the ratio of the difference to the window value of the congestion window corresponding to the target update time.

E4: Taking the update time of the target network link during the link influence period as the target update time, and obtaining the sum of multiple ratios.

E5: The ratio of the window adjustment times and the sum of the ratios is determined as the window parameter of the congestion window of the target network link.

E2-E5 can be expressed as cwnd _j ^T =N _a /∑ _k [(t _c -t _j ^k )/cwnd _j ^k ].

Wherein, cwnd _j ^T is the window parameter of the congestion window of the j-th network link. _{N a} is the number of window adjustments of the j-th network link during the link influence period. t _c is the time when the target data packet waits for allocation. _{t j} ^k is the update time of the k-th update of the congestion window of the j-th network link. cwnd _j ^k is the window value of the k-th update of the congestion window of the j-th network link.

Therefore, using the harmonic mean of the window values of the congestion window during the link impact period as the window parameter of the congestion window can not only avoid emergencies and improve the stability of the network state, but also take into account the different degrees of influence of congestion windows at different update times on the network state during the link impact period, thereby improving the accuracy of the window parameters and further improving the accuracy of the network state evaluation, so as to determine the comparison of the network states between various network links.

As a possible implementation, the embodiment of the present application also provides two methods for adjusting the congestion window, so that the adjustment of the congestion window is more suitable for the network state change of the network link.

Congestion window adjustment method 1: based on packet loss.

F1: Get the response signal returned by the receiving end.

In actual applications, the sender sends a data packet to the receiver through a network link, and the receiver sends a response signal to the sender to inform the sender of its own situation, such as whether the data packet is received, through the response signal.

It should be noted that the timing of adjusting the congestion window can be after the sender sends the target data packet through the network link to be sent, or before the sender sends the target data packet through the network link to be sent, and this application does not make specific restrictions on this. The following is an example of adjusting the congestion window of the network link to be sent after the sender sends the target data packet through the network link to be sent.

F2: If the response signal does not include a confirmation character, determine a first number of data packets in a buffer queue of the network link to be sent.

If the response signal does not include an ACK character, it indicates that there is currently a packet loss situation, that is, after the sender sends the target data packet through the to-be-sent network, the receiver may not receive the target data packet. At this time, the first number of data packets in the cache queue of the to-be-sent network link is determined to determine two situations in which packet loss exists, which are respectively described below through F3 and F4.

Among them, in data communication, ACK is a transmission control character sent by the receiving end to the sending end. It indicates that the sent data has been confirmed to be received correctly. For example, in the Transmission Control Protocol/Internet Protocol (TCP/IP), if the receiving end successfully receives the data, it will reply with a response signal including ACK data. Usually, ACK data has its own fixed format, length, etc.

As a possible implementation manner, the first quantity may be determined by the following formula:

N _cg =(cwnd _j ^T /RTT-cwnd _j ^T /RTT _cg )*RTT _min

Wherein, N _cg is the first number, cwnd _j ^T is the window parameter of the congestion window of the j-th network link, and in the embodiment of the present application, it can be the window parameter of the congestion window of the network link to be sent. RTT is the round-trip delay of the network link to be sent. RTT _min is the minimum round-trip delay among the round-trip delays of multiple network links. RTT _cg can be expressed as RTT _cg = (1-β)*RTT _cg +β*SRTT, SRTT is the smooth round-trip delay of the network link transmission, and β is the attenuation factor. As a possible implementation method, β can be taken as 0.2.

F3: If the first number is greater than the first data packet threshold, the congestion window is reduced by a first decreasing weight.

If the first number is greater than the first packet threshold, it means that the number of packets cached in the cache queue of the network link to be sent is small. At this time, the reason for packet loss may be due to the network instability of the network link to be sent, which causes packet loss for a moment, that is, a low packet loss rate. After that, the network state may instantly return to a normal state, or it may continue to be in the current unstable state, so the congestion window can be reduced, that is, the congestion window of the network to be sent is reduced by the first descending weight to avoid the network state of the network link to be sent being always bad.

The embodiment of the present application does not specifically limit the first data packet threshold, and those skilled in the art can set it according to actual needs. As a possible implementation, it is found through experiments that when the first data packet threshold is 2, the aggregate throughput can be maximized.

F4: If the first number is less than or equal to the first data packet threshold, reducing the congestion window by a second decreasing weight.

If the first number is less than or equal to the first data packet threshold, it means that the number of data packets cached in the cache queue of the network link to be sent is large, resulting in the inability of the cache queue to continue to receive new data packets, and the sending end will be unable to transmit data packets to the receiving end, resulting in a high packet loss rate. At this time, it is necessary to quickly reduce the window value of the congestion window of the network link to be sent, that is, to reduce the congestion window of the network to be sent through the second descending weight with a larger value, so that fewer data packets are sent through the network link to be sent, thereby reducing the number of packet losses.

It should be noted that the first drop weight is less than the second drop weight. For example, if the first number N _cg > the first data packet threshold θ _c , the first drop weight can be expressed as w _d ^r , and the adjusted congestion window can be expressed as cwnd _j ^T+1 =(1-w _d ^r )*cwnd _j ^T . Wherein, cwnd _j ^T is the window parameter of the congestion window of the j-th network link, and cwnd _j ^T+1 is the window parameter of the updated congestion window of the j-th network link. If the first number N _cg <= the first data packet threshold θ _c , the second drop weight can be expressed as w _d ^c , and the adjusted congestion window can be expressed as cwnd _j ^T+1 =(1-w _d ^c )*cwnd _j ^T . As a possible implementation, the value of the first drop weight w _d ^r can be 0.2, and the value of the second drop weight w _d ^c can be 0.3.

From the above technical solution, it can be seen that the congestion window of the network link with high packet loss rate needs to be reduced quickly, and the congestion window of the network link with low packet loss rate can be reduced slowly. Therefore, the window value of the congestion window is adaptively adjusted according to different packet loss situations, so that the adjustment of the congestion window is closer to the network state change of the wireless network link or the mobile network link.

Congestion window adjustment method 2: based on delay conditions.

F1: Get the response signal returned by the receiving end.

F5: If the response signal includes a confirmation character, then the round-trip delays corresponding to the multiple network links are obtained, and a reference delay is determined from the round-trip delays corresponding to the multiple network links.

If the response signal includes a confirmation character, it means that the receiving end has received the target data packet. At this time, the round-trip delays corresponding to multiple network links can be obtained. For example, the round-trip delays of each network link in the past period of time are counted. From the round-trip delays corresponding to multiple network links, any round-trip delay is determined as the benchmark delay, such as the minimum round-trip delay or the maximum round-trip delay.

F6: Determine the update rate of the congestion window of the network link to be sent according to the round-trip delay of the network link to be sent and the benchmark delay.

For example, if the reference delay is the minimum round-trip delay among the round-trip delays of multiple network links, that is, the minimum round-trip delay, the update rate of the congestion window of the network link to be sent can be expressed as γ _j . Wherein, γ _j =RTT _j /RTT _min . RTT _min is the minimum round-trip delay among the round-trip delays of multiple network links, and RTT _j is the round-trip delay of the jth network link.

F7: If the update rate is greater than or equal to the update rate threshold, reduce the congestion window of the network link to be sent.

If the update rate is greater than or equal to the update rate threshold, it means that the network link to be sent is in a high-latency situation. In order to avoid the high-latency congestion window updating speed too fast, which makes the delay difference between different network links greater and affects the stability of throughput, the congestion window can be reduced at this time, that is, the update rate of the congestion window can be limited to make it enter the attenuation stage.

As a possible implementation method, an exponentially weighted moving average method can be used to reduce the congestion window of the network link to be sent, which can be expressed as follows:

cwnd _j ^T+1 = β*cwnd _j ^T +(1-β)*(cwnd _j ^T /γ _j )

Wherein, cwnd _j ^T is the window parameter of the congestion window of the j-th network link, and cwnd _j ^T+1 is the window parameter of the updated congestion window of the j-th network link. γ _j is the update rate of the congestion window of the network link to be sent. Β is the attenuation factor. As a possible implementation, β can be 0.2.

The embodiment of the present application does not specifically limit the value of the update rate threshold, and those skilled in the art can set it according to actual needs. As a possible implementation, when the update rate threshold is 3, the aggregate throughput is maximum.

F8: If the update rate is less than the update rate threshold, increase the congestion window of the network link to be sent.

If the update rate is less than the update rate threshold, it means that the network link to be sent is in a low latency situation. At this time, the congestion window of the network link to be sent can be increased, such as by using traditional congestion control algorithms such as opportunistic linked increases algorithm (OLIA).

As a possible implementation method, determine the second number of data packets in the cache queue of the network link to be sent; if the second number exceeds the second data packet threshold, wait for another confirmation character to be received, and then increase the congestion window of the network link to be sent; if the second number is less than the second data packet threshold, increase the congestion window of the network link to be sent. The second data packet threshold can be equal to the first data packet threshold, and the present application does not make specific restrictions on this, and those skilled in the art can make social according to actual needs. Therefore, by increasing the window value of the congestion window after receiving an ACK on a network link with a low packet loss rate, and increasing the congestion window size after receiving two ACK returns on a network link with a high packet loss rate, it can be ensured that when the network link is blocked, the increase in the congestion window is more conservative, which slows down the introduction of more data packet transmissions.

As a possible implementation manner, if the congestion window is still in the slow start phase, the slow start threshold is additionally set to the original congestion window size cwnd _j ^t .

It can be seen from the above technical solution that the operation of reducing the congestion window is performed in the congestion window of the network link where the high latency is located, and the operation of increasing the congestion window is performed in the congestion window of the network link where the low latency is located, so as to avoid the high-latency congestion window from being updated too quickly, causing the delay difference between different network links to increase, and affecting the stability of throughput.

In order to facilitate further understanding of the technical solution provided by the embodiment of the present application, the data packet scheduling method provided by the embodiment of the present application is taken as an example in which the execution subject is a terminal device, and the data packet scheduling method is introduced as an overall example.

First, the application scenario of the data packet scheduling method is described.

The data packet scheduling method provided in the embodiment of the present application can be applied to various multi-access mobile applications, that is, scenarios where data packets are sent to mobile applications through multiple or multiple network links, especially streaming media applications that have high requirements for data transmission bandwidth, latency and stability, such as online education, e-commerce live broadcast, VR/AR cloud games, etc. Among them, streaming media is a new media transmission method, which refers to a technology and process that compresses a series of multimedia data and sends the data in segments through the Internet, which can instantly transmit audio and video for viewing; the most important technical feature of streaming media is streaming transmission, which allows data to be transmitted like flowing water.

The following takes the process of multi-channel video transmission from the video source to the ingestion server deployed in the cloud or edge during video live broadcast as an example.

See Figure 3, which is a schematic diagram of an application scenario of a data packet scheduling method provided in an embodiment of the present application. In Figure 3, the video source is the sender, and a mobile phone or a webcam can be used as a video source to continuously capture video frames. After encoding, the sender sends data packets to the ingest server through multiple network links such as WiFi, LTE/5G, and a network router. The ingest server is connected to multiple content delivery network (CDN) servers, thereby sending the received data packets to the CDN server.

Then, a data packet scheduling system is described. The data packet scheduling system applies the data packet scheduling method provided in an embodiment of the present application, which is described below in conjunction with Figure 4.

Referring to FIG. 4 , which is a schematic diagram of a data packet scheduling system provided in an embodiment of the present application. FIG. 4 can also be viewed as an end-to-end framework running on both sides of a video source and an ingestion server. The data packet scheduling system includes a video source 410 and an ingestion server 420. Data transmission is achieved between the video source 410 and the ingestion server 420 through a transmission control protocol based on UDP Quick UDP Internet Connection (QUIC), which is described below.

The video source 410 includes a video capture module 411, a frame conversion module 412, a packet scheduling module 413 and a congestion control module 414. The video capture module 411 continuously captures video frames, and caches the encoded video frames in a sending queue of the video source. The frame conversion module 412 extracts video frames from the sending queue in sequence and encapsulates them into QUIC data packets. The packet scheduling module 413 selects a suitable network link for each encapsulated data packet, such as the target data packet described above, based on the window value of the congestion window of the network link, that is, executes the aforementioned S201-S203. The congestion control module 414 adjusts the window value of the congestion window of each network link according to the round-trip delay and/or packet loss of each network link, that is, executes the aforementioned F1-F8.

The QUIC data packets are sent to the ingestion server 420 through multiple network links, and the ingestion server 420 caches them in the receiving queue. The ingestion server 420 includes a frame conversion module 421 and a video cache module 422. Among them, the frame conversion module 421 extracts the QUIC data packets from the receiving queue in sequence, reorders them according to the packet numbers of the QUIC data packets, reconstructs them into video frames and sends them to the video cache module 422. The video cache module 422 performs operations such as content distribution on the video frames.

In addition, the data packet scheduling system in the related art establishes an application layer in the user state and a transport layer in the system kernel, so that the application layer and the transport layer cannot be directly accessed. In the data packet scheduling system provided in the embodiment of the present application, a cross-layer network architecture is designed to integrate the transport layer with the application layer, and both the application layer and the transport layer are established in the user state, and the video stream generated by the video source is transmitted online to the server, reducing the redundant configuration and operation of the two layers, and can also reduce the transmission time.

Finally, the method for scheduling data packets is described below with reference to FIG5 and FIG6.

See Figure 5, which is a flow chart of a method for scheduling data packets provided in an embodiment of the present application. The method for scheduling data packets can be applied to the packet scheduling module 413 in Figure 4, which is described in detail below.

S501: Start.

S502: Record the initial window value of the congestion window of each network link.

S503: Determine whether the congestion window of each network link needs to be adjusted. If so, execute S504; if not, execute S505.

S504: If it is identified that the window value of the congestion window of a certain network link has changed, the updated window value and the update time are recorded, and S503 is executed.

S505: Determine whether there is a target data packet to be allocated on each network link. If yes, execute S506; if not, execute S509.

S506: Obtain window parameters of the congestion windows of each network link.

For details, please refer to B2-B3, which will not be repeated here.

S507: Determine routing probabilities corresponding to the multiple network links respectively according to the window parameters of the multiple congestion windows.

For details, please refer to A1-A4, which will not be repeated here.

S508: Distribute the target data packet to multiple network links according to the multiple routing probabilities, and execute S505.

S509: Determine whether to stop sending the target data packet, if so, execute S510; if not, execute S503.

S510: End.

See Figure 6, which is a flow chart of a method for scheduling data packets provided in an embodiment of the present application. The method for scheduling data packets can be applied to the congestion control module 414 in Figure 4, which is described in detail below.

S601: Start.

S602: Obtain a response signal returned by the receiving end.

S603: Determine whether the response signal includes a certain character, if so, execute S604; if not, execute S609.

S604: Determine a first number of data packets in a cache queue of a network link to be sent.

S605: Determine whether the first number is greater than a first data packet threshold, if so, execute S606; if not, execute S607.

S606: Reduce the congestion window by first decreasing the weight, and execute S608.

S607: Reduce the congestion window by a second weight reduction, and execute S608.

S608: Determine whether to continue to obtain the response signal returned by the receiving end, if so, execute S602; if not, execute S616.

S609: Determine an update rate of the congestion window of the network link to be sent.

For details, please refer to F5-F6, which will not be repeated here.

S610: Determine whether the update rate threshold is exceeded, if so, execute S611; if not, execute S612.

S611: Use an exponentially weighted moving average method to reduce the congestion window of the network link to be sent, and execute S608.

S612: Determine a second number of data packets in the buffer queue of the network link to be sent.

S613: Determine whether the second number exceeds the second data packet threshold, if so, execute S614; if not, execute S615.

S614: Waiting to receive another confirmation character.

S615: Increase the congestion window of the network link to be sent.

S616: End.

It can be seen from the above technical solution that the routing probability is determined by the window value based on the congestion window, and the target data packet is probabilistically allocated through random numbers, and the congestion control mechanism is optimized to perceive the delay and packet loss in real time, accurately detect the changes in the wireless link network state, and reasonably select the transmission path to make full use of the overall available bandwidth, effectively avoid inaccurate link state prediction and head-of-line blocking and additional network overhead. Various multi-access mobile applications with high requirements for data transmission bandwidth, delay and stability can provide efficient data transmission support, effectively adapt to the heterogeneity and dynamics of mobile networks, make full use of the overall available bandwidth and avoid additional network overhead. Experiments have found that compared with the existing mainstream multi-path scheduling methods, there is a significant improvement of more than 20% in aggregate throughput and video transmission bit rate. Among them, the video transmission bit rate refers to the number of bits corresponding to the transmission of the video stream per unit time in streaming media applications.

With respect to the data packet scheduling method described above, the present application also provides a corresponding data packet scheduling device so that the above-mentioned data packet scheduling method can be applied and implemented in practice.

Refer to Figure 7, which is a schematic diagram of the structure of a data packet scheduling device provided in an embodiment of the present application. As shown in Figure 7, the data packet scheduling device 700 includes: an acquisition unit 701, a determination unit 702 and a sending unit 703;

The acquisition unit 701 is used to acquire window parameters of congestion windows corresponding to multiple network links respectively if a target data packet to be allocated is received, wherein the network link is a channel for transmitting the target data packet, and the window parameters of the congestion window are used to describe the network status of the corresponding network link;

The determining unit 702 is used to determine the routing probabilities corresponding to the multiple network links respectively according to the window parameters of the multiple congestion windows, and the routing probabilities are used to describe the probability that the corresponding network links are distributed with the target data packets; wherein the larger the window parameter of the congestion window is, the larger the corresponding routing probability is;

The sending unit 703 is used to distribute the target data packet to the multiple network links according to the multiple routing probabilities.

It can be seen from the above technical scheme that if a target data packet to be allocated is received, the target data packet needs to be distributed to the network link for transmission, and the window parameters of the congestion window of each network link are obtained at this time. Among them, the congestion window is a variable of the transmission protocol used in the transmission process, which can adapt to the network state of the network link, that is, the network state of each network link can be fully reflected through the window parameters of the congestion window. According to the window parameters of the congestion window, the routing probabilities corresponding to each network link are determined, that is, the probability of each network link being distributed with the target data packet is determined according to the window parameters of the congestion window. Among them, the larger the window parameter of the congestion window, the better the network state of the corresponding network link, so the corresponding routing probability is larger, so as to allocate the target data packet to the network link according to the routing probability, so that the network link with the larger routing probability has more opportunities to be distributed with the target data packet. Thus, the routing probability is determined by the window parameters of the congestion window, and the data packet to be allocated is distributed according to the routing probability, so that the data packet can be distributed to the network link with better network condition with a greater probability, thereby improving the scheduling effect of the data packet as a whole, and then improving the transmission speed of the data packet. In addition, since the window parameters of the congestion window are adjusted in real time according to the network status of the network link, it is possible to more accurately detect changes in the network status of the wireless link, and the routing probability of each network link is calculated separately, effectively adapting to the heterogeneity and dynamics of mobile networks or wireless networks.

As a possible implementation manner, the number of the multiple network links is m, where m is an integer greater than 1, and the sending unit 703 is specifically configured to:

Constructing m continuous accumulation intervals according to the multiple routing probabilities; wherein the starting value of the jth accumulation interval corresponding to the jth network link is the sum of the routing probabilities of the previous j-1 network links, and the ending value of the jth accumulation interval is the sum of the routing probabilities of the previous j network links, and j is a positive integer less than or equal to m;

Generate a random number, the random number is greater than or equal to the starting value of the first accumulation interval, and the random number is less than or equal to the ending value of the mth accumulation interval;

Determine the network link corresponding to the accumulation interval in which the random number falls as the network link to be sent;

Distribute the target data packet to the network link to be sent.

As a possible implementation manner, the sending unit 703 is specifically configured to:

Determine the network link corresponding to the largest routing probability among the plurality of routing probabilities as the network link to be sent;

Distribute the target data packet to the network link to be sent.

As a possible implementation manner, the device further includes a window parameter determining unit, configured to:

If a window value of a congestion window identifying a target network link changes, recording an updated window value and an update time, the target network link being one of the plurality of network links;

Determine the updated window value corresponding to the update time closest to the current time as the window parameter of the congestion window corresponding to the target network link;

The multiple network links are respectively used as the target network links, and window parameters of congestion windows corresponding to the multiple network links are obtained.

As a possible implementation manner, the device further includes a window parameter determining unit, configured to:

If a window value of a congestion window identifying a target network link changes, recording an updated window value and an update time, the target network link being one of the plurality of network links;

Determine, according to the update time, a window value of a congestion window of the target network link within a link impact period, wherein the link impact period is a preset time period;

A window parameter of the congestion window of the target network link is determined according to a window value of the congestion window of the target network link during the link influence period.

As a possible implementation manner, the device further includes a window parameter determining unit, configured to:

Determining the number of window adjustments of the target network link during the link impact period according to the window value of the congestion window of the target network link during the link impact period;

Obtaining a difference between a target update time and a time when the target data packet is waiting for allocation, wherein the target update time is one of the update times of the target network link during the link influence period;

Obtaining a ratio of the difference to a window value of a congestion window corresponding to the target update time;

Taking the update time of the target network link during the link influence period as the target update time, and obtaining the sum of multiple ratios;

The ratio of the window adjustment times to the sum of the ratios is determined as the window parameter of the congestion window of the target network link.

As a possible implementation manner, the device further includes a window parameter determining unit, configured to:

Determine an arithmetic mean of window values of the congestion window of the target network link during the link influence period;

The arithmetic mean is determined as a window parameter of a congestion window of the target network link.

As a possible implementation manner, the device further includes a congestion window adjustment unit, configured to:

After sending the target data packet through the network link to be sent, obtaining a response signal returned by the receiving end;

If the response signal does not include a confirmation character, determining a first number of data packets in a cache queue of the to-be-sent network link;

If the first number is greater than a first packet threshold, reducing the congestion window by a first decreasing weight;

If the first number is less than or equal to the first data packet threshold, the congestion window is reduced by a second decreasing weight, and the first decreasing weight is less than the second decreasing weight.

As a possible implementation manner, the device further includes a congestion window adjustment unit, configured to:

If the response signal includes a confirmation character, obtaining round trip delays corresponding to the plurality of network links respectively, and determining a reference delay from the round trip delays corresponding to the plurality of network links respectively;

Determining an update rate of a congestion window of the network link to be sent according to the round-trip delay of the network link to be sent and the reference delay;

If the update rate is greater than or equal to the update rate threshold, reducing the congestion window of the network link to be sent;

If the update rate is less than the update rate threshold, the congestion window of the network link to be sent is increased.

As a possible implementation manner, the device further includes a congestion window adjustment unit, configured to:

An exponentially weighted moving average method is used to reduce the congestion window of the network link to be sent.

As a possible implementation manner, the device further includes a congestion window adjustment unit, configured to:

Determine a second number of data packets in the cache queue of the to-be-sent network link;

If the second number exceeds the second data packet threshold, then wait for receiving another confirmation character and then increase the congestion window of the network link to be sent;

If the second number is less than the second data packet threshold, the congestion window of the network link to be sent is increased.

The embodiment of the present application also provides a computer device, which is the computer device described above, and the computer device can be a server or a terminal device. The data packet scheduling device described above can be built into the server or the terminal device. The computer device provided by the embodiment of the present application will be introduced from the perspective of hardware entity. Among them, FIG8 shows a schematic diagram of the structure of the server, and FIG9 shows a schematic diagram of the structure of the terminal device.

Referring to FIG8 , which is a schematic diagram of a server structure provided in an embodiment of the present application, the server 1400 may have relatively large differences due to different configurations or performances, and may include one or more processors 1422, such as central processing units (CPU), memory 1432, one or more application programs 1442 or storage media 1430 (e.g., one or more mass storage devices) of data 1444. Among them, the memory 1432 and the storage medium 1430 may be short-term storage or permanent storage. The program stored in the storage medium 1430 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Furthermore, the processor 1422 may be configured to communicate with the storage medium 1430 and execute a series of instruction operations in the storage medium 1430 on the server 1400.

The server 1400 may also include one or more power supplies 1426, one or more wired or wireless network interfaces 1450, one or more input and output interfaces 1458, and/or one or more operating systems 1441, such as Windows Server ^™ , Mac OS X ^™ , Unix ^™ , Linux ^™ , FreeBSD ^™ , etc.

The steps executed by the server in the above embodiment may be based on the server structure shown in FIG. 8 .

The CPU 1422 is used to perform the following steps:

If a target data packet to be allocated is received, then window parameters of congestion windows corresponding to a plurality of network links are obtained, wherein the network links are channels for transmitting the target data packet, and the window parameters of the congestion windows are used to describe the network status of the corresponding network links;

Determine, according to the window parameters of the multiple congestion windows, the routing probabilities corresponding to the multiple network links respectively, wherein the routing probabilities are used to describe the probabilities of the corresponding network links being distributed with the target data packets; wherein the larger the window parameter of the congestion window is, the larger the corresponding routing probability is;

Distribute the target data packet to the plurality of network links according to the plurality of routing probabilities.

Optionally, the CPU 1422 may also execute the method steps of any specific implementation of the data packet scheduling method in the embodiments of the present application.

See Figure 9, which is a schematic diagram of the structure of a terminal device provided in an embodiment of the present application. Figure 9 shows a block diagram of a partial structure of a smart phone related to the terminal device provided in an embodiment of the present application, and the smart phone includes: a radio frequency (RF) circuit 1510, a memory 1520, an input unit 1530, a display unit 1540, a sensor 1550, an audio circuit 1560, a wireless fidelity (WiFi) module 1570, a processor 1580, and a power supply 1590 and other components. Those skilled in the art will understand that the smart phone structure shown in Figure 9 does not constitute a limitation on the smart phone, and may include more or fewer components than shown, or combine certain components, or arrange components differently.

The following is a detailed introduction to the various components of the smart phone in conjunction with FIG9:

The RF circuit 1510 may be used for receiving and sending signals during information transmission or communication or during a call. In particular, after receiving the downlink information from the base station, it is sent to the processor 1580 for processing. In addition, the designed uplink data is sent to the base station.

The memory 1520 may be used to store software programs and modules. The processor 1580 implements various functional applications and data processing of the smartphone by running the software programs and modules stored in the memory 1520 .

The input unit 1530 can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the smartphone. Specifically, the input unit 1530 may include a touch panel 1531 and other input devices 1532. The touch panel 1531, also known as a touch screen, can collect user touch operations on or near it and drive the corresponding connection device according to a pre-set program. In addition to the touch panel 1531, the input unit 1530 may also include other input devices 1532. Specifically, other input devices 1532 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.

The display unit 1540 may be used to display information input by the user or information provided to the user and various menus of the smartphone. The display unit 1540 may include a display panel 1541, and the display panel 1541 may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

The smartphone may also include at least one sensor 1550, such as a light sensor, a motion sensor, and other sensors. As for other sensors that may be configured in the smartphone, such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., they will not be described in detail here.

The audio circuit 1560, the speaker 1561, and the microphone 1562 can provide an audio interface between the user and the smartphone. The audio circuit 1560 can transmit the received audio data to the speaker 1561 after converting the received audio data into an electrical signal, which is converted into a sound signal for output; on the other hand, the microphone 1562 converts the collected sound signal into an electrical signal, which is received by the audio circuit 1560 and converted into audio data, and then the audio data is output to the processor 1580 for processing, and then sent to another smartphone through the RF circuit 1510, or the audio data is output to the memory 1520 for further processing.

The processor 1580 is the control center of the smartphone, and uses various interfaces and lines to connect various parts of the entire smartphone, and executes various functions of the smartphone and processes data by running or executing software programs and/or modules stored in the memory 1520, and calling data stored in the memory 1520. Optionally, the processor 1580 may include one or more processing units.

The smart phone also includes a power supply 1590 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 1580 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption management.

Although not shown, the smartphone may also include a camera, a Bluetooth module, etc., which will not be described in detail here.

In the embodiment of the present application, the memory 1520 included in the smart phone can store program codes and transmit the program codes to the processor.

The processor 1580 included in the smart phone can execute the data packet scheduling method provided in the above embodiment according to the instructions in the program code.

An embodiment of the present application also provides a computer-readable storage medium for storing a computer program, wherein the computer program is used to execute the data packet scheduling method provided in the above embodiment.

The embodiment of the present application also provides a computer program product or a computer program, which includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the data packet scheduling method provided in various optional implementations of the above aspects.

A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the above program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the above storage medium can be at least one of the following media: read-only memory (English: Read-Only Memory, abbreviated: ROM), RAM, magnetic disk or optical disk, etc. Various media that can store program codes.

It should be noted that each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments. The device and system embodiments described above are merely schematic, in which the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a technician familiar with the technical field within the technical scope disclosed in the present application should be included in the protection scope of the present application. Based on the implementation methods provided in the above aspects, the present application can also be further combined to provide more implementation methods. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims (15)

1. A data packet scheduling method, characterized in that the method comprises: If a target data packet to be allocated is received, then window parameters of congestion windows corresponding to a plurality of network links are obtained, wherein the network links are channels for transmitting the target data packet, and the window parameters of the congestion windows are used to describe the network status of the corresponding network links; Determine, according to the window parameters of the multiple congestion windows, the routing probabilities corresponding to the multiple network links respectively, wherein the routing probabilities are used to describe the probabilities of the corresponding network links being distributed with the target data packets; wherein the larger the window parameter of the congestion window is, the larger the corresponding routing probability is; Distribute the target data packet to the plurality of network links according to the plurality of routing probabilities. 2. The method according to claim 1, wherein the number of the plurality of network links is m, where m is an integer greater than 1, and the step of distributing the target data packet to the plurality of network links according to the plurality of routing probabilities comprises: Constructing m continuous accumulation intervals according to the multiple routing probabilities; wherein the starting value of the jth accumulation interval corresponding to the jth network link is the sum of the routing probabilities of the previous j-1 network links, and the ending value of the jth accumulation interval is the sum of the routing probabilities of the previous j network links, and j is a positive integer less than or equal to m; Generate a random number, the random number is greater than or equal to the starting value of the first accumulation interval, and the random number is less than or equal to the ending value of the mth accumulation interval; Determine the network link corresponding to the accumulation interval in which the random number falls as the network link to be sent; Distribute the target data packet to the network link to be sent. 3. The method according to claim 1, characterized in that distributing the target data packet to the multiple network links according to the multiple routing probabilities comprises: Determine the network link corresponding to the largest routing probability among the plurality of routing probabilities as the network link to be sent; Distribute the target data packet to the network link to be sent. 4. The method according to claim 1, characterized in that the method further comprises: If a window value of a congestion window identifying a target network link changes, recording an updated window value and an update time, the target network link being one of the plurality of network links; Determine the updated window value corresponding to the update time closest to the current time as the window parameter of the congestion window corresponding to the target network link; The multiple network links are respectively used as the target network links, and window parameters of congestion windows corresponding to the multiple network links are obtained. 5. The method according to claim 1, characterized in that the method further comprises: If a window value of a congestion window identifying a target network link changes, recording an updated window value and an update time, the target network link being one of the plurality of network links; Determine, according to the update time, a window value of a congestion window of the target network link within a link impact period, wherein the link impact period is a preset time period; A window parameter of the congestion window of the target network link is determined according to a window value of the congestion window of the target network link during the link influence period. 6. The method according to claim 5, characterized in that the determining the window parameter of the congestion window of the target network link according to the window value of the congestion window of the target network link during the link influence period comprises: Determining the number of window adjustments of the target network link during the link impact period according to the window value of the congestion window of the target network link during the link impact period; Obtaining a difference between a target update time and a time when the target data packet is waiting for allocation, wherein the target update time is one of the update times of the target network link during the link influence period; Obtaining a ratio of the difference to a window value of a congestion window corresponding to the target update time; Taking the update time of the target network link during the link influence period as the target update time, and obtaining the sum of multiple ratios; The ratio of the window adjustment times to the sum of the ratios is determined as the window parameter of the congestion window of the target network link. 7. The method according to claim 5, characterized in that the determining the window parameter of the congestion window of the target network link according to the window value of the congestion window of the target network link during the link influence period comprises: Determine an arithmetic mean of window values of the congestion window of the target network link during the link influence period; The arithmetic mean is determined as a window parameter of a congestion window of the target network link. 8. The method according to claim 1, characterized in that after sending the target data packet through the network link to be sent, the method further comprises: Get the response signal returned by the receiving end; If the response signal does not include a confirmation character, determining a first number of data packets in a cache queue of the to-be-sent network link; If the first number is greater than a first packet threshold, reducing the congestion window by a first decreasing weight; If the first number is less than or equal to the first data packet threshold, the congestion window is reduced by a second decreasing weight, and the first decreasing weight is less than the second decreasing weight. 9. The method according to claim 8, characterized in that the method further comprises: If the response signal includes a confirmation character, obtaining round trip delays corresponding to the plurality of network links respectively, and determining a reference delay from the round trip delays corresponding to the plurality of network links respectively; Determining an update rate of a congestion window of the network link to be sent according to the round-trip delay of the network link to be sent and the reference delay; If the update rate is greater than or equal to the update rate threshold, reducing the congestion window of the network link to be sent; If the update rate is less than the update rate threshold, the congestion window of the network link to be sent is increased. 10. The method according to claim 9, characterized in that the step of reducing the congestion window of the network link to be sent comprises: An exponentially weighted moving average method is used to reduce the congestion window of the network link to be sent. 11. The method according to claim 9, wherein increasing the congestion window of the network link to be sent comprises: Determine a second number of data packets in the cache queue of the to-be-sent network link; If the second number exceeds the second data packet threshold, then wait for receiving another confirmation character and then increase the congestion window of the network link to be sent; If the second number is less than the second data packet threshold, the congestion window of the network link to be sent is increased. 12. A data packet scheduling device, characterized in that the device comprises: an acquisition unit, a determination unit and a sending unit; The acquisition unit is used to acquire window parameters of congestion windows corresponding to a plurality of network links respectively if receiving a target data packet to be allocated, wherein the network link is a channel for transmitting the target data packet, and the window parameters of the congestion window are used to describe the network status of the corresponding network link; The determining unit is used to determine the routing probabilities corresponding to the plurality of network links respectively according to the window parameters of the plurality of congestion windows, wherein the routing probabilities are used to describe the probabilities of the corresponding network links being distributed with the target data packets; wherein the larger the window parameter of the congestion window is, the larger the corresponding routing probability is; The sending unit is used to distribute the target data packet to the multiple network links according to the multiple routing probabilities. 13. A computer device, characterized in that the computer device comprises a processor and a memory: The memory is used to store a computer program and transmit the computer program to the processor; The processor is configured to execute the method according to any one of claims 1 to 11 according to the instructions in the computer program. 14. A computer-readable storage medium, characterized in that the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 11. 15. A computer program product comprising a computer program, characterized in that when the computer program product is run on a computer device, the computer device is enabled to execute the method according to any one of claims 1 to 11.