CN112491573A

CN112491573A - Network parameter configuration method and device

Info

Publication number: CN112491573A
Application number: CN201911353303.0A
Authority: CN
Inventors: 晏思宇; 曲迪; 冀智刚; 夏寅贲
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2019-09-12
Filing date: 2019-12-25
Publication date: 2021-03-12
Anticipated expiration: 2039-12-25
Also published as: CN112491573B

Abstract

The application relates to the technical field of communication, and discloses a method and a device for configuring network parameters (such as ECN (engineering-coupled-network) waterlines), which are used for dynamically configuring the network parameters according to the change of network transmission characteristics, so that the network parameters dynamically adapt to the change of network flow and the network transmission performance is ensured. The method comprises the following steps: acquiring network statistical data corresponding to a first time period, wherein the network statistical data comprise network transmission characteristics of network equipment in the first time period and a first value corresponding to a specified network parameter; determining a second value corresponding to the specified network parameter according to the network statistical data; configuring the specified network parameter of the network device to the second value for a second time period, the second time period being a time period after the first time period.

Description

Network parameter configuration method and device

Cross Reference to Related Applications

The present application claims priority of chinese patent application entitled "a network parameter configuration method and related apparatus" filed by the intellectual property office of the people's republic of china, application No. 201910867097.9, on 12/9/2019, the entire contents of which are incorporated herein by reference.

Technical Field

The embodiment of the application relates to the technical field of communication, in particular to a network parameter configuration method and device.

Background

With the rapid development of the internet, various network services are continuously enriched, and higher requirements are provided for network transmission while higher experience is brought to users. At present, throughput and delay are main indexes of network transmission, wherein network parameters such as an Explicit Congestion Notification (ECN) pipeline for congestion control, a priority-based flow control (PFC) pipeline, and a Random Early Detection (RED) packet loss pipeline play an important role in regulating and controlling the throughput and delay of network transmission on network devices such as switches and routers. For example: in a Transmission Control Protocol (TCP) network, throughput and delay can be regulated by adjusting RED packet loss pipelines of network devices; in a Remote Direct Memory Access (RDMA) network, throughput and delay can be regulated by adjusting an ECN waterline of a network device.

Therefore, reasonable network parameter configuration plays an important role in throughput and time delay of network transmission, and taking an ECN (event-coupled network) waterline as an example, the ECN waterline is high in arrangement, which easily causes buffer (buffer) queues of network equipment to be continuously accumulated, increases data buffer time delay, and causes transmission time delay to be increased, and the ECN waterline is low in arrangement, which easily causes the ECN waterline of the network equipment to be easily triggered, reduces the forwarding amount of data, and causes under-throughput. At present, network parameters are usually configured manually by experts according to experience, but with the increasing complexity of network environments, network services and network traffic become diverse, different network traffic models can be generated, and the manual configuration mode cannot adapt to the increasing complex network environments. Under a dynamic network flow model, how network parameters realize dynamic adjustment and accurate adaptation along with the change of network flow, thereby ensuring network transmission performance becomes a problem to be solved urgently.

Disclosure of Invention

The application provides a network parameter configuration method and a network parameter configuration device, which are used for dynamically configuring network parameters according to the change of network transmission characteristics (network flow models), so that the network parameters dynamically adapt to the change of network flow and the network transmission performance is ensured.

In a first aspect, the present application provides a method for configuring network parameters, including: acquiring network statistical data corresponding to a first time period, wherein the network statistical data comprise network transmission characteristics of network equipment in the first time period and a first value corresponding to a specified network parameter; determining a second value corresponding to the specified network parameter according to the network statistical data; configuring the specified network parameter of the network device to the second value for a second time period, the second time period being a time period after the first time period.

The network parameter configuration method described in this application may be implemented by a network device such as a switch or a router, or by a processing chip, a circuit, or other components in the network device, or by a configuration server connected to the network device via a communication network, or by a processing chip, a circuit, or other components in the configuration server. By adopting the method, the value of the network parameter appointed by the network equipment in the next time period is determined and configured according to the network statistical data of the network equipment in a certain time period, so that the network parameter can realize dynamic adjustment and accurate adaptation along with the change of network flow, and dynamically adapt to the network transmission characteristics, thereby ensuring the network transmission performance.

In one possible design, the network transmission characteristics of the network device during the first time period include a throughput of the network device during the first time period, when the throughput of the network device during the first time period increases relative to an initial time period, the initial time period being a time period before the first time period, and the determining the second value corresponding to the specified network parameter according to the network statistics includes: and modifying the first value according to the first adjusting direction and the first step length to obtain the second value.

In the design, the throughput increase is used as positive feedback, and the specified network parameters are continuously adjusted according to the first adjustment direction, so that the throughput of the network equipment is increased, and the network transmission performance is improved.

In one possible design, the method further includes: adjusting the first step size by a first step size increment when the throughput of the network device increases relative to the first time period in the second time period; when the throughput of the network device does not increase in the second time period relative to the first time period, adjusting the first step size by a first step size decrement.

In the design, after the designated network parameters are adjusted, when the throughput of the network equipment rises, the first step length of the designated network parameters is increased according to the first step length increment, so that the current network transmission characteristics of the network equipment can be matched quickly and efficiently, and the network transmission performance is improved; after the designated network parameters are adjusted, when the throughput of the network equipment is not increased, the first step size of the designated network parameters is reduced according to the first step size reduction, so that the problem that the transmission performance of the network is influenced due to the fact that the amplitude of the designated network parameters is adjusted to be large is solved.

In one possible design, the network transmission characteristics of the network device during the first time period include a throughput of the network device during the first time period, and when the throughput of the network device during the first time period does not increase relative to an initial time period, the determining the second value corresponding to the specified network parameter based on the network statistics includes: determining an initial value corresponding to the specified network parameter in an initial time period; modifying the initial value according to a second adjusting direction and a second step length to obtain a second value; or modifying the first value according to a first adjustment direction and a first step length to obtain the second value, wherein the second adjustment direction is opposite to the first adjustment direction, and the initial time period is a time period before the first time period.

In the design, the condition that the throughput does not rise is taken as negative feedback, the specified network parameters are stopped to be continuously adjusted according to the first adjustment direction, the values of the specified network parameters are restored to the values of the network parameters corresponding to the previous time period, and the specified network parameters can be adjusted according to the second adjustment direction, so that the problems that the specified network parameters are unreasonable in setting and the network transmission performance is influenced, such as time delay increase and the like are avoided; in addition, when the throughput is not increased, the specified network parameters can be continuously adjusted according to the first adjustment direction according to a certain probability (such as 10% and the like), so that the problem that the configuration of the specified network parameters falls into the local optimal solution can be effectively avoided.

In one possible design, when the throughput of the network device does not increase relative to the first time period in the second time period, or the data buffering delay is not reduced and the number of data throughput channels is not increased, the specified network parameter of the network device in a third time period is configured to be the first value, and the third time period is a time period after the second time period.

In the design, the designated network parameters are stopped being adjusted according to the second adjusting direction by taking the fact that the throughput is not increased, or the data caching delay is not reduced and the number of data throughput channels is not increased as negative feedback, and the values of the designated network parameters are restored to the values of the network parameters corresponding to the previous time period, so that the problems that the designated network parameters are unreasonable in setting and the network transmission performance is affected, such as underthroughput and the like, are avoided.

In one possible design, the method further includes: when the throughput of the network equipment is not reduced in the second time period relative to the first time period, and the data caching delay is reduced or the number of data throughput channels is increased, adjusting the second step size according to a second step size increment; and when the throughput of the network equipment is reduced relative to the first time period in the second time period, or the data buffering delay is not reduced and the number of data throughput channels is not increased, adjusting the second step size according to a second step size reduction.

In the design, after the designated network parameters are adjusted, when the throughput of the network equipment is not reduced, the data cache delay is reduced or the number of data throughput channels is increased, the second step length of the designated network parameters is increased according to the second step length increment, so that the current network transmission characteristics of the network equipment can be matched quickly and efficiently, and the network transmission performance is improved; after the designated network parameters are adjusted, when the throughput of the network equipment is reduced, or the data caching delay is not reduced and the number of data throughput channels is not increased, the second step length of the designated network parameters is reduced according to the second step length reduction, so that the problem that the transmission performance of the network is influenced due to the fact that the amplitude of the designated network parameters is large in adjustment is avoided.

In one possible design, before obtaining the network statistics corresponding to the first time period, the method further includes: and determining that the difference value of the throughputs calculated for a plurality of times is smaller than the throughput fluctuation threshold value. In the design, the interference of the flow fluctuation and other factors on the network statistical data acquired by the network equipment is avoided, and the accuracy and reliability of the network parameter configuration are ensured.

In a second aspect, the present application provides a communication device having the functionality to implement the method according to the first aspect or any one of the possible designs of the first aspect, where the functionality may be implemented by hardware, or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions, such as a communication unit and a processing unit.

In one possible design, the device may be a chip or an integrated circuit.

In one possible design, the apparatus includes a memory and a processor, the memory is used for storing a program executed by the processor, and when the program is executed by the processor, the apparatus may perform the functions of the first aspect or any of the possible designs of the first aspect.

In one possible design, the apparatus may be a network device such as a switch, router, or the like.

In a third aspect, the present application further provides a network device, where the network device has a function of implementing the behavior of the network device in the first aspect or any one of the methods of the first aspect. The functions can be realized by hardware, and the functions can also be realized by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the above-described functions.

In one possible design, the network device includes a communication interface configured to receive and transmit data, a processor configured to support the network device to perform the corresponding functions in the first aspect or any one of the possible designs of the first aspect, and a memory. The memory is coupled to the processor and retains program instructions and data necessary for the network device.

In a fourth aspect, the present application provides a computer-readable storage medium storing computer instructions that, when executed, implement the method of the first aspect or any one of the possible designs of the first aspect.

In a fifth aspect, the present application also provides a computer program product comprising a computer program or instructions which, when executed, may implement the method as described in the first aspect above or in any one of the possible designs of the first aspect.

In a sixth aspect, the present application further provides a chip for implementing the method according to the first aspect or any one of the possible designs of the first aspect.

The technical effects achieved by the second to sixth aspects are described with reference to the technical effects achieved by the first aspect.

Drawings

Fig. 1 is a schematic diagram of a communication architecture according to an embodiment of the present application;

fig. 2 is a schematic diagram of an ECN mechanism provided in an embodiment of the present application;

fig. 3 is a schematic diagram of a network parameter configuration algorithm according to an embodiment of the present application;

fig. 4 is a schematic diagram of a finite state machine implementation of a network parameter configuration algorithm according to an embodiment of the present application;

fig. 5 is a schematic diagram of an implementation of a throughput detection phase according to an embodiment of the present application;

fig. 6 is a schematic diagram of an implementation of a delay detection stage according to an embodiment of the present application;

fig. 7 is a schematic diagram of a network parameter configuration process according to an embodiment of the present application;

fig. 8 is a schematic diagram illustrating an implementation of network parameter configuration according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a communication device according to an embodiment of the present application.

Detailed Description

Fig. 1 exemplarily shows a possible communication system architecture diagram provided in the embodiment of the present application, which includes a plurality of terminal devices and a plurality of network devices, where the plurality of terminal devices may implement communication in a Transmission Control Protocol (TCP), Remote Direct Memory Access (RDMA), and other manners through one or more network devices. The network device can continuously adjust (configure) the network parameters for congestion control according to network statistical data (such as network throughput) in one or more time periods, so that the network parameters in the network device can realize dynamic adjustment and accurate adaptation along with the change of network flow, and the network transmission performance is guaranteed.

Optionally, in the communication system architecture shown in fig. 1, a configuration server may be further included, and the configuration server may be interconnected with one or more network devices through a communication network, where the communication network may be a local area network, a wide area network switched by a relay (relay) device, or a local area network and a wide area network. The network device can also send the network statistical data of the network device itself in one or more time periods to the configuration server, and the configuration server continuously adjusts (configures) the network parameters of the network device for congestion control based on the network statistical data of the network device in one or more time periods and sends the network parameters to the network device, so that the network parameters in the network device can realize dynamic adjustment and accurate adaptation along with the change of network flow, and the network transmission performance is guaranteed.

Before describing the embodiments of the present application, some terms in the present application will be explained to facilitate understanding for those skilled in the art.

1) Remote Direct Memory Access (RDMA), is a direct memory access technology that quickly moves data directly from the memory of one terminal device to the memory of another terminal device over a network to eliminate the impact on processor performance. The main processes of RDMA are: the first terminal device directly sends the data in the memory to network devices such as switches and routers in the network through a network card of the first terminal device, the network devices forward the data to the second terminal device, and the network card of the second terminal device directly stores the received data in the memory of the second terminal device, so that processors of the two terminal devices are not required to participate in the data access process, and the influence on the performance of the processors of the terminal devices is reduced. This allows high-throughput, low-latency network communications, particularly suitable for use in massively parallel computer clusters.

2) The network device may be a switch, a router, a gateway, or other devices or devices having a data exchange function, and the present invention is not limited thereto.

3) A terminal device, a device with transceiving function, for example, a handheld device with wireless/wired connection function, a vehicle-mounted device, a wearable device, a computing device, a service server, a Mobile Station (MS) or other processing device connected to a wireless modem, and the like, and a mobile terminal communicating with one or more core networks via an access network, and the like.

4) Throughput, which refers to the amount of data (measured in bits, bytes, packets, etc.) successfully transmitted per unit of time to a network, device, port, virtual circuit, or other facility.

5) Data buffering delay refers to delay caused by data transmission by a buffer of a network device, and generally takes an average time required for data to be transmitted from one end (a buffer queue inlet) to the other end (a buffer queue outlet) of a buffer queue of the network device.

6) The network device may provide one or more buffer queues for data transmission (forwarding), and each buffer queue provided by the network device for data transmission is a data throughput channel of the network device. In general, under a certain condition of a network device buffer, the number of data throughput channels of the network device decreases with the increase of the average buffer occupancy of the data buffer queue, and conversely, the number of data throughput channels of the network device increases with the decrease of the average buffer occupancy of the data buffer queue.

7) And an Explicit Congestion Notification (ECN) mechanism, as shown in fig. 2, the network device may configure an ECN upper pipeline (Kmax), an ECN lower pipeline (Kmin), and an ECN marking probability (PropPcnt), where the ECN marking probability may also be referred to as a maximum marking probability. Marking probabilities are defined by an ECN (equal to the actual marking probability is 0) upper waterline and an ECN (equal to the actual marking probability is 0) of a data packet passing through the network equipment when the length of the buffer queue is lower than the ECN lower waterline, the ECN (equal to the actual marking probability is 1) of the data packet passing through the network equipment when the length of the buffer queue is higher than the ECN upper waterline, the ECN (equal to the actual marking probability is 1) of the data packet passing through the network equipment when the length of the buffer queue is not lower than the ECN lower waterline and not higher than the ECN upper waterline, the probability that the data packet passing through the network equipment linearly increases with the actual length of the buffer queue of the network equipment is adopted by the marking probabilities, and the marking probabilities are generally adopted by

Wherein K is the actual length of the network device buffer queue.

When receiving the CNP, the sending end device may correspondingly reduce the sending rate of the data packet according to information such as ECN marking probability of the data packet included in the CNP, and alleviate network congestion of the network device, thereby avoiding data packet loss. Generally, increasing an ECN upper waterline, increasing an ECN lower waterline, or decreasing an ECN marking probability may improve throughput of network equipment, but may cause queue congestion and increase delay; conversely, decreasing the ECN waterline, or increasing the ECN marking probability will alleviate queue congestion and reduce latency, but will tend to cause throughput loss.

8) The system comprises a priority-based flow control (PFC) mechanism, and is a queue-based backpressure mechanism, wherein a PFC waterline is configured on a network device, when the length of a cache queue in the network device reaches the PFC waterline, a pause (pause) frame is sent to an upstream device to inform the upstream device to pause sending a data packet, and when the length of the cache queue in the network device is lower than the PFC waterline, a PFC cancellation message is sent to the upstream device to inform the upstream device to resume sending the data packet, so that data packet loss is avoided.

The method aims to adjust network parameters such as an ECN (engineering center network) water-feeding line, an ECN water-dropping line, an ECN marking probability, a PFC (power factor correction) water line and the like through dynamic adaptation of network transmission characteristics, so that the network parameters are dynamically adjusted and accurately adapted along with the change of network flow, the throughput is maximized, and the time delay is reduced, so that the network transmission performance is guaranteed.

The following describes embodiments of the present application in detail with reference to the drawings. In addition, it should be understood that in the embodiments of the present application, at least one may also be described as one or more, and a plurality may be two, three, four or more, and the present application is not limited thereto.

In the embodiments of the present application, "/" may indicate a relationship in which the objects associated before and after are "or", for example, a/B may indicate a or B; "and/or" may be used to describe that there are three relationships for the associated object, e.g., A and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. For convenience in describing the technical solutions of the embodiments of the present application, in the embodiments of the present application, terms such as "first" and "second" may be used to distinguish technical features having the same or similar functions. The terms "first", "second", and the like do not necessarily limit the number and execution order, and the terms "first", "second", and the like do not necessarily differ. In the embodiments of the present application, the words "exemplary" or "such as" are used to indicate examples, illustrations or illustrations, and any embodiment or design described as "exemplary" or "e.g.," should not be construed as preferred or advantageous over other embodiments or designs. The use of the terms "exemplary" or "such as" are intended to present relevant concepts in a concrete fashion for ease of understanding.

The network parameter configuration process provided in the embodiment of the present application is introduced below with reference to a scenario in which a network device configures network parameters of its own ECN (an ECN watermark probability, and an ECN watermark waterline).

Referring to fig. 3 and 4, the configuration process for the network parameters of the ECN may include three phases (states): the steady-state probing phase, the throughput probing phase (Probe _ BW) and the delay probing phase (Probe _ RTT) can be switched between different phases (states) by a finite state machine. Steady state detection phase (state 1): detecting the throughput, and entering a throughput detection stage or a time delay detection stage when the throughput of the network equipment is detected to be in a steady state (stable); throughput probing phase (state 2): exciting an ECN upper waterline (Kmax) and an ECN lower waterline (Kmin) to be adjusted up and exciting an ECN marking probability (DropPcnt) to be adjusted down until the throughput reaches an upper limit, and returning to a steady-state detection stage; delay probing phase (state 3): and (3) taking ' the reduction of data cache delay or the increase of the number of data throughput channels ' as positive feedback under the condition that the throughput is not reduced ', exciting the downward regulation of an ECN upper waterline and an ECN lower waterline, and exciting the upward regulation of ECN marking probability until the minimum data cache delay and/or the maximum number of data throughput channels are reached, and returning to the steady-state detection stage. In the embodiment of the present application, the finite state machine is a mathematical model representing a finite number of phases (states) and behaviors such as transitions and/or actions among the phases, and transitions among a steady-state probing phase, a throughput probing phase and a delay probing phase may be implemented by the finite state machine, for example, in the steady-state probing phase, throughput is detected, and when it is detected that throughput of the network device is in a steady state, the network device enters the throughput probing phase or the delay probing phase, and transitions from the steady-state probing phase to the throughput probing phase or the delay probing phase are implemented.

Specifically, when the network device runs, the running time of the network device can be divided into a plurality of time periods, network statistical data corresponding to the time periods are obtained, and then the ECN water-up line, the ECN water-down line and the ECN marking probability can be determined and configured according to the network statistical data of different time periods, so that the ECN water-up line, the ECN water-down line and the ECN marking probability are dynamically adapted to the current network transmission characteristics (network traffic model) of the network device, and the network transmission performance is guaranteed. The time lengths corresponding to the multiple time periods divided by the network device may be the same or different, and this is not limited in this embodiment of the application.

As an example, the network device may divide its running time according to a set duration (or period) to obtain a plurality of time periods. The network statistical data acquired by the network device for any time period includes network transmission characteristics such as the number of forwarding bytes, the average depth of a cache queue, a timestamp for sending a packet, throughput, the number of data throughput channels, and the like of the network device in the time period, where the throughput of the network device in the time period may be determined according to a ratio of the number of forwarding bytes of the network device in the time period to a duration of the time period (for example, a difference between a maximum timestamp and a minimum timestamp for sending a packet in the time period), and the network statistical data may further include values corresponding to network parameters such as an ECN water-up line, an ECN water-down line, and an ECN marking probability of the network device in the time period.

The network parameter configuration of the ECN is described below with reference to a specific configuration phase.

Firstly, the method comprises the following steps: steady state detection phase

In order to avoid interference caused by factors such as flow fluctuation and the like on network statistical data acquired by the network equipment and ensure the accuracy and reliability of the network parameter configuration of the ECN, when the network equipment performs the network parameter configuration of the ECN, firstly, according to the throughputs of network equipment in a plurality of continuous time periods, calculating whether the difference value of the throughputs of two adjacent time periods of the network equipment is continuously less than a throughput fluctuation threshold value for a plurality of times so as to detect whether the flow of the network equipment enters a steady state, if the difference between the throughputs calculated for successive times is greater than or equal to the throughput fluctuation threshold, it indicates that there is a large fluctuation in the traffic through the network device, the method may bring great interference to the determination and configuration of the network parameters of the ECN, the network device keeps the network parameters of the ECN unchanged, and continues to calculate whether the throughput difference value of two adjacent time periods of the network device is continuously smaller than the throughput fluctuation threshold for multiple times.

If the throughput difference calculated for a plurality of times is smaller than the throughput fluctuation threshold value, it indicates that the flow of the network device is relatively stable, the network device enters a steady state stage, the flow fluctuation and other factors do not cause relatively large interference to the determination and configuration of the network parameters of the ECN, and the network device enters a throughput detection stage or a delay detection stage.

In a possible implementation, after determining that the steady-state phase is entered, the network device performs polling selection on the throughput detection phase and the delay detection phase, that is, the throughput detection phase or the delay detection phase selected after two adjacent steady-state phases are different, for example, the throughput detection phase is selected after the last steady-state phase, and the delay detection phase is selected after the next steady-state phase.

Optionally, when there are multiple network parameters of the ECN, such as an ECN water-up line, an ECN water-down line, and an ECN marking probability, the network device may configure the ECN water-up line, the ECN water-down line, and the ECN marking probability in sequence according to a sequence of first configuring (adjusting) the ECN water-up line, then configuring the ECN water-down line, and finally configuring the ECN marking probability. It should be understood that, when there are multiple network parameters of the ECN, the present embodiment does not limit the order of configuring the multiple network parameters of the ECN, for example, an order of first configuring an ECN lower pipeline, then configuring an ECN upper pipeline, and finally configuring an ECN marking probability may also be adopted.

II, secondly: throughput detection phase

Referring to FIG. 5, when configuring the ECN pipeline:

the network device may first adjust up a value of the ECN water-up line of a next time period according to a first step corresponding to the ECN water-up line, wait to enter the next time period, and after the network device enters the next time period, the next time period is used as a current time period of the network device, the network device obtains network statistical data of the current time period, and determines throughput of the network device of the current time period, where the first step corresponding to the ECN water-up line may be 20%, 30%, 50%, and the like of the value of the default ECN water-up line.

After the network equipment determines the throughput of the network equipment in the current time period, comparing the throughput of the current time period with the throughput of the previous time period, and if the throughput of the network equipment in the previous time period is increased compared with the throughput of the network equipment in the previous time period, continuously adjusting the value of an ECN (engineering-centric networking) water-up line in the next time period according to a first step length corresponding to the ECN water-up line; if the throughput of the network device in the previous time period is not increased compared with the current time period, the network device may set the value of the ECN waterline in the next time period to the value of the ECN waterline corresponding to the previous time period, and enter the steady-state detection stage.

In one possible implementation, to avoid falling into a locally optimal solution for the configuration of the ECN waterline, if the throughput of the network device in the current time period is not increased compared to the previous time period, the network device may further continue to adjust the value of the ECN waterline in the next time period up by the first step corresponding to the ECN waterline with the probability of alpha1, set the value of the ECN waterline in the next time period to the value of the ECN waterline in the previous time period with the probability of (1-alpha1), and enter the steady-state detection phase. Wherein alpha1 is not more than 1, optionally alpha1 is not more than 50%, such as 10%, 15%, etc., although the value of alpha1 may also be randomly selected within a certain value range (such as 5% -30%), or dynamically attenuated or increased within a certain value range (such as attenuated within a certain value range along with the increase of the value of the waterline on the ECN).

In one possible implementation, in order to ensure that the ECN water-up line matches the current network traffic model of the network device quickly and efficiently, if the throughput of the network device in the current time period is increased compared to the throughput of the previous time period, the network device may further increase the first step size corresponding to the ECN water-up line, for example, increase the first step size from 40% of the value of the default ECN water-up line to 60% of the value of the default ECN water-up line; if the throughput of the network device in the current time period is not increased compared with the throughput of the previous time period, the network device may further down-regulate a first step size corresponding to the ECN waterline, for example, down-regulate the first step size from 40% of the default ECN waterline to 20% of the default ECN waterline.

The same applies to the configuration of the lower ECN pipeline and the configuration of the upper ECN pipeline, and when the lower ECN pipeline is configured, the above-mentioned process of configuring the upper ECN pipeline may be referred to.

When marking the ECN with probability configuration:

the network device may first adjust a value of the ECN marking probability of a next time period according to a first step corresponding to the ECN marking probability, wait to enter the next time period, and after the network device enters the next time period, the next time period is used as a current time period of the network device, and the network device obtains network statistical data of the network device of the current time period, and determines a throughput of the network device of the current time period. Wherein the ECN marking probability may correspond to a first step size of 0.1, 0.2, etc.

After the network equipment determines the throughput of the network equipment in the current time period, comparing the throughput of the current time period with the throughput of the previous time period, and if the throughput of the network equipment in the previous time period is increased compared with the throughput of the network equipment in the current time period, continuously reducing the value of the ECN marking probability of the next time period according to the first step corresponding to the ECN marking probability; if the throughput of the network device in the current time period is not increased compared with the throughput of the network device in the previous time period, the network device may set the value of the ECN marking probability in the next time period as the value of the ECN marking probability corresponding to the previous time period, and enter a steady-state detection stage.

In one possible implementation, to avoid trapping in the configuration of ECN marking probabilities into a locally optimal solution, if the throughput of the network device in the current time period is not increased compared with the previous time period, the network device can further continuously adjust the value of the ECN marking probability in the next time period according to the first step corresponding to the ECN marking probability with the probability of alpha2, setting the value of the ECN marking probability of the next time period as the value of the ECN marking probability corresponding to the previous time period according to the probability of (1-alpha2), entering a steady-state detection stage, wherein alpha2 is not greater than 1, optionally alpha2 is also not greater than 50%, such as 10%, 15%, etc., of course, the value of alpha2 may also be randomly selected within a certain range (e.g., 5% -30%), or dynamically attenuated or increased within a certain range (e.g., attenuated within a certain range as the value of ECN marking probability decreases).

Thirdly, the method comprises the following steps: time delay detection phase

Referring to FIG. 6, when configuring the ECN pipeline:

the network device may first adjust a value of the ECN water-up line of a next time period according to a second step corresponding to the ECN water-up line, wait to enter the next time period, and after the network device enters the next time period, the next time period is used as a current time period of the network device, the network device obtains network statistical data of the current time period, and determines throughput, data cache delay, and data throughput channel number of the network device of the current time period, where the second step corresponding to the ECN water-up line may be the same as or different from a first step corresponding to the ECN water-up line, such as 20%, 30%, 40%, and the like of the value of the default ECN water-up line.

After the network equipment determines the throughput, the data cache delay and the data throughput channel number of the network equipment in the current time period, comparing the throughput, the data cache delay and the data throughput channel number of the current time period with those of the previous time period, and if the throughput of the network equipment is not reduced and the data cache delay is reduced or the data throughput channel number is increased compared with the previous time period, continuing to reduce the value of an ECN (engineering-centric networking) water-up line in the next time period according to a second step corresponding to the ECN water-up line; if the throughput of the network device is decreased or the data buffering delay is not decreased and the number of data throughput channels is not increased when the current time period is compared with the previous time period, the network device may set the value of the ECN water-up line of the network device in the next time period to the value of the ECN water-up line corresponding to the previous time period, and enter the steady-state detection stage.

In one possible implementation, in order to avoid falling into a locally optimal solution for the configuration of the ECN pipeline, if the throughput of the network device is decreased in the current time period compared to the previous time period, or the data buffering delay is not decreased and the number of data throughput channels is not increased, the network device may further continue to decrease the value of the ECN pipeline in the next time period according to the second step corresponding to the ECN pipeline with the probability of alpha3, set the value of the ECN pipeline in the next time period to the value of the ECN pipeline corresponding to the previous time period with the probability of (1-alpha3), and enter the steady-state detection phase. Wherein alpha3 is not more than 1, optionally alpha3 is not more than 50%, such as 10%, 20%, etc., although the value of alpha3 may also be randomly selected within a certain value range (such as 5% -25%), or dynamically attenuated or increased within a certain value range (such as attenuated within a certain value range along with the decrease of the value of the waterline on the ECN).

In one possible implementation, to ensure that the ECN pipeline matches the current network traffic model of the network device quickly and efficiently, if the throughput of the network device in the current time period is not decreased compared to the throughput of the previous time period, and the data buffering delay is decreased or the number of data throughput channels is increased, the network device may further increase a second step corresponding to the ECN pipeline, for example, the second step is increased from 40% of the value of the default ECN pipeline to 60% of the value of the default ECN pipeline; if the throughput of the network device in the current time period is decreased compared with the throughput of the previous time period, or the data buffering delay is not decreased and the number of data throughput channels is not increased, the network device may further down-regulate a second step size corresponding to the ECN water-up line, for example, down-regulate the second step size from 40% of the default ECN water-up line to 20% of the default ECN water-up line.

The same adjustment of the ECN lower waterline as the ECN upper waterline can be referred to the above adjustment process of the ECN upper waterline.

When marking the ECN with probability configuration:

the network device may first adjust up a value of ECN marking probability of a next time period according to a second step corresponding to the ECN marking probability, wait to enter the next time period, and after the network device enters the next time period, the next time period is used as a current time period of the network device, the network device obtains network statistical data of the current time period, and determines throughput, data cache delay, and data throughput channel number of the network device in the current time period. The second step length corresponding to the ECN marking probability may be the same as or different from the first step length corresponding to the ECN marking probability, for example, may be 0.1, 0.2, and the like.

After the network equipment determines the throughput, the data cache delay and the data throughput channel number of the network equipment in the current time period, comparing the throughput, the data cache delay and the data throughput channel number of the current time period with those of the previous time period, and if the throughput of the network equipment is not reduced and the data cache delay is reduced or the data throughput channel number is increased compared with the previous time period, continuously adjusting the value of the ECN marking probability of the next time period according to a second step length corresponding to the ECN marking probability; if the throughput of the network device is decreased or the data buffering delay is not decreased and the number of data throughput channels is not increased when the current time period is compared with the previous time period, the network device may set the value of the ECN marking probability of the next time period to the value of the ECN marking probability corresponding to the previous time period, and enter the steady-state detection stage.

In one possible implementation, to avoid trapping in the configuration of ECN marking probabilities into a locally optimal solution, if the throughput of the network device is not decreased in the current time period compared to the previous time period, and the data buffering delay is decreased or the number of data throughput channels is increased, the network device may also continue to adjust the value of the ECN marking probability for the next time period up by a second step corresponding to the ECN marking probability with a probability of alpha4, setting the value of the ECN marking probability of the next time period as the value of the ECN marking probability corresponding to the previous time period according to the probability of (1-alpha4), entering a steady-state detection stage, wherein alpha4 is not greater than 1, optionally alpha4 is also not greater than 50%, such as 10%, 20%, etc., certainly, the value of alpha4 can also be randomly selected within a certain value range (e.g., 5% -30%), or dynamically attenuated or increased within a certain value range (e.g., attenuated within a certain value range with the increase of the value of ECN marking probability).

A network parameter configuration method provided in an embodiment of the present application is described below, and is applicable to a communication system architecture shown in fig. 1, where a network device configures a network parameter of the network device, as shown in fig. 7, a specific flow of the method includes:

s701, network statistical data corresponding to a first time period are obtained by the network equipment, and the network statistical data comprise network transmission characteristics of the network equipment in the first time period and a first value corresponding to a specified network parameter.

In the embodiment of the application, when the network device runs, the running time of the network device can be divided into a plurality of time periods, network data corresponding to the time periods are respectively counted to obtain network statistical data including each time period, and then the network parameters can be determined and configured according to the network transmission characteristics of the different time periods, so that the network parameters are dynamically adapted to the network transmission characteristics of the network device, and the network transmission performance is guaranteed. The time lengths corresponding to the multiple time periods divided by the network device may be the same or different, which is not limited in this application embodiment, and as an example, the network device may divide the running time of the network device according to a set time length (or period) to obtain multiple time periods.

In addition, in order to avoid interference caused by factors such as flow fluctuation on network statistical data acquired by the network equipment and ensure the accuracy and reliability of network parameter configuration, in one possible implementation, before the network equipment determines and configures the specified network parameters, the network equipment determines that the throughput difference value calculated for multiple times continuously is smaller than the throughput fluctuation threshold value.

That is, before the network device determines and configures the network parameters, it first calculates, according to the throughputs of the network device in a plurality of consecutive time periods, whether the throughput difference between two adjacent time periods of the network device is continuously smaller than a throughput fluctuation threshold for a plurality of times to detect whether the traffic of the network device enters a steady state, if the throughput difference calculated for a plurality of consecutive times is smaller than the throughput fluctuation threshold, it indicates that the traffic of the network device is relatively steady, the network device enters the steady state, the traffic fluctuation and other factors do not cause relatively large interference to the determination and configuration of the specified network parameters, and the network device determines and configures the specified network parameters.

After the network device determines that the throughput difference calculated for a plurality of consecutive times is smaller than the throughput fluctuation threshold, the network device may modify the value of the specified network parameter according to the first adjustment direction and the first adjustment step size, and configure the value of the specified network parameter in the next time period as the modified value according to the first adjustment direction and the first adjustment step size. And after determining that the throughput difference calculated for a plurality of times is smaller than the throughput fluctuation threshold value, taking each time period in which the network equipment is currently located as a first time period of the network equipment, and acquiring network statistical data corresponding to the first time period.

In the embodiment of the present application, the specified network parameters may include, but are not limited to, one or more of an ECN watermark-up line, an ECN watermark-down line, an ECN marking probability, a PFC pipeline, an RED packet loss pipeline, and a RED packet loss threshold; network transmission characteristics may include, but are not limited to, one or more of throughput, data buffering latency, number of data throughput channels, etc.; the network statistical data of the first time period acquired by the network equipment comprises the network transmission characteristics of the network equipment in the first time period and a first value corresponding to the designated network parameter.

In addition, it should be understood that, in the embodiment of the present application, when the specified network parameters are different, the first adjustment direction, the first step size, the second adjustment direction, the second step size, and the like for modifying the specified network parameters may be the same or different. For example: when the designated network parameter is an ECN waterline, the first adjusting direction is up-regulation and the second adjusting direction is down-regulation, and when the designated network parameter is an ECN waterline, the first adjusting direction is up-regulation and the second adjusting direction is down-regulation and is the same as the ECN waterline; when the designated network parameter is ECN marking probability, the first adjusting direction is down-regulation, and the second adjusting direction is up-regulation, which is different from an ECN up-line.

In one possible implementation, specifying a first adjustment direction of the network parameter is beneficial to improving throughput of the network device, and specifying a second adjustment direction of the network parameter is beneficial to reducing data buffering delay of the network device and/or increasing the number of data throughput channels of the network device, where the first adjustment direction and the second adjustment direction are generally opposite.

S702, the network equipment determines a second value corresponding to the specified network parameter according to the network statistical data.

Taking the network transmission characteristics including the throughput as an example, in one possible implementation, when the throughput of the network device increases in a first time period relative to an initial time period, where the initial time period is a time period before the first time period, it is described that the specified network parameter is adjusted according to the first adjustment direction, which is beneficial to improving the throughput of the network device, and the network device modifies the first value corresponding to the specified network parameter according to the first adjustment direction and the first step length to obtain the second value of the network parameter. When the throughput of the network device in the first time period does not rise with respect to the initial time period, it is described that the designated network parameter is adjusted according to the first adjustment direction, which is not beneficial to improving the throughput of the network device, the network device may back a value corresponding to the designated network parameter to an initial value corresponding to the initial time period, and the initial value corresponding to the initial time period is used as a second value of the designated network parameter, where the initial time period is a time period before the first time period.

In addition, in order to avoid trapping the configuration of the network parameter into a locally optimal solution, when determining that the throughput in the first time period does not rise relative to the throughput in the initial time period, the network device may further back the value corresponding to the specified network parameter to the initial value corresponding to the initial time period according to a first probability (e.g., 1-alpha), and use the initial value corresponding to the initial time period as a second value of the specified network parameter; and modifying the first value corresponding to the specified network parameter according to the second probability (such as alpha) and the first adjustment direction and the first step length to obtain a second value corresponding to the specified network parameter, wherein the value of the alpha is not more than 1, such as 10%, 15% and the like.

As an example, when determining that the throughput in the first time period does not increase relative to the throughput in the initial time period, the network device may back, according to the first probability, the value corresponding to the ECN waterline to the initial value corresponding to the initial time period, use the initial value corresponding to the initial time period as the second value of the ECN waterline, and modify, according to the second probability, the first value corresponding to the ECN waterline according to the first adjustment direction and the first adjustment step size, to obtain the second value corresponding to the ECN waterline. The sum of the first probability and the second probability is 1, and the configuration may be performed according to the requirement, for example, the first probability is configured to be 90%, and the second probability is configured to be 10%.

In another possible implementation, when it is determined that the throughput in the first time period does not increase relative to the throughput in the initial time period, the network device may further modify the initial value according to the second adjustment direction and the second step after returning the value corresponding to the specified network parameter to the initial value corresponding to the initial time period, to obtain a second value of the specified network parameter, so as to reduce the data buffering delay of the network device and/or increase the number of data throughput channels of the network device.

S703, the network device configures the specified network parameter of the network device in a second time period as the second value, where the second time period is a time period after the first time period.

In a possible implementation manner, in order to ensure that the specified network parameters quickly and efficiently match the current network transmission characteristics of the network device, after S703, when the throughput of the network device increases in the second time period relative to the first time period, the first step size is adjusted according to the first step size increment; when the throughput of the network device does not increase in the second time period relative to the first time period, the first step size is adjusted by the first step size decrement.

In addition, if the network device modifies the value of the designated network parameter according to the second adjustment direction and the second step length to obtain a second value of the designated network parameter, and configures the value of the designated network parameter in the next time period (the second time period) as the second value, when the throughput of the network device is not decreased and the data buffering delay is decreased or the number of data throughput channels is increased in the next time period (the second time period) in which the value of the designated network parameter is modified according to the second adjustment direction and the second step length, compared with the time period (the first time period) before the value of the designated network parameter is modified according to the second adjustment direction and the second step length, it is indicated that the designated network parameter is adjusted according to the second adjustment direction, which is beneficial to decreasing the data buffering delay or increasing the number of data throughput channels, and the network device continues to modify the value of the designated network parameter (the value of the designated network parameter corresponding to the second time period) according to the second adjustment direction and the second step length, and configures the modified value of the specified network parameter for a time period next to the second time period (third time period).

When the network device modifies the value of the specified network parameter in the next time period (the second time period) according to the second adjustment direction and the second step length, and the throughput of the network device does not increase or the data buffering delay is not reduced and the number of data throughput channels is not increased compared to the time period (the first time period) before the value of the specified network parameter is modified according to the second adjustment direction and the second step length, it is described that the specified network parameter is adjusted according to the second adjustment direction, which is not favorable for reducing the data buffering delay and increasing the number of data throughput channels, and the network device may rollback the value of the specified network parameter in the next time period (the third time period) in the second time period to the value (the first value) of the specified network parameter in the previous time period (the first time period).

In a possible implementation, to avoid falling into a locally optimal solution to the configuration of the network parameter, when the network device is in a next time period (second time period) in which the value of the specified network parameter is modified according to the second adjustment direction and the second step length, and when the throughput of the network device does not increase or the data buffering delay is not reduced and the number of data throughput channels is not increased relative to a time period (first time period) before the value of the specified network parameter is modified according to the second adjustment direction and the second step length, the network device may further back the value of the specified network parameter in the next time period (third time period) in the second time period to the value of the specified network parameter in the previous time period (first time period) (first value) according to a third probability; and modifying the value of the specified network parameter (the value of the specified network parameter corresponding to the second time period) according to the fourth probability and the second adjustment direction and the second step length, and configuring the modified value of the specified network parameter for the next time period (the third time period) of the second time period, wherein the sum of the third probability and the fourth probability is 1, and the configuration can be performed according to requirements, for example, the third probability is configured to be 85%, and the fourth probability is configured to be 15%.

In a possible implementation manner, in order to ensure that the specified network parameters are quickly and efficiently matched with the current network transmission characteristics of the network device, when the throughput of the network device is not reduced in a second time period relative to the first time period, and the data buffering delay is reduced or the number of data throughput channels is increased, adjusting a second step size according to a second step size increment; and when the throughput of the network equipment is reduced relative to the first time period in the second time period, or the data buffering delay is not reduced and the number of data throughput channels is not increased, adjusting the second step size according to the second step size reduction.

Referring to fig. 8, in this embodiment of the application, when the network device runs, the running time of the network device may be divided into a plurality of time periods, and network data corresponding to the plurality of time periods are counted respectively to obtain network statistical data including each time period, so as to determine a value of a specified network parameter according to network transmission characteristics of different time periods, and configure the determined value of the specified network parameter into the network device, so as to ensure transmission performance of the network device. If the network parameter is designated as the ECN waterline, the throughput of the network equipment rises relative to the time period before the ECN waterline value is adjusted by the network equipment, the network equipment continues to adjust the ECN waterline value, and the value of the ECN waterline in the next time period is configured as the ECN waterline value obtained after continuous adjustment, so that the network equipment obtains higher throughput.

In other embodiments, the network device may also send the network statistical data of each time period to the configuration server, and the configuration server dynamically modifies the value of the specified network parameter of the network device according to the network statistical data of the network device in each time period based on the parameter configuration method, and sends the modified value of the network parameter to the network device to adapt to the current network transmission characteristic (network traffic model) of the network device, thereby ensuring the performance of the network.

The above description mainly introduces the scheme provided in the present application from the perspective of network devices. It is understood that in order to implement the above functions, the network device includes a corresponding hardware structure and/or software module (or unit) that performs each function. Those of skill in the art would readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

Based on the above embodiments, the present application provides a communication apparatus, and referring to fig. 9, the communication apparatus 900 includes: a communication interface 901, a processor 902, and a memory 903, wherein:

the communication interface 901, the processor 902 and the memory 903 are connected to each other. Optionally, the communication interface 901, the processor 902 and the memory 903 are connected to each other through a bus 904; the bus 904 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 9, but this does not indicate only one bus or one type of bus.

When the communication apparatus 900 implements the network parameter configuration method shown in fig. 7:

the communication interface 901 is used for receiving and sending data;

the processor 902 is configured to call program instructions stored in the memory 903 to perform the following method: acquiring network statistical data corresponding to a first time period through a communication interface 901, where the network statistical data includes network transmission characteristics of network equipment in the first time period and a first value corresponding to a specified network parameter; determining a second value corresponding to the specified network parameter according to the network statistical data; configuring the specified network parameter of the network device to the second value for a second time period, the second time period being a time period after the first time period.

In one possible design, the network transmission characteristics of the network device during the first time period include a throughput of the network device during the first time period, when the throughput of the network device during the first time period increases relative to an initial time period, the initial time period being a time period before the first time period, and the determining, according to the network statistics, the second value corresponding to the specified network parameter includes: and modifying the first value according to the first adjusting direction and the first step length to obtain the second value.

In one possible design, the network transmission characteristics of the network device during the first time period include a throughput of the network device during the first time period, when the throughput of the network device during the first time period does not rise with respect to an initial time period, the initial time period being a time period before the first time period, and the determining, by the network statistics, the second value corresponding to the specified network parameter includes: determining an initial value corresponding to the specified network parameter in an initial time period; modifying the initial value according to a second adjusting direction and a second step length to obtain a second value; or modifying the first value according to the first adjusting direction and the first step length to obtain the second value.

In one possible design, the method further includes, when the throughput of the network device does not increase relative to the first time period in the second time period, or the data buffering latency is not reduced and the number of data throughput channels is not increased, configuring the specified network parameter of the network device to the first value in a third time period, the third time period being a time period after the second time period.

In one possible design, the method further includes adjusting the second step size by a second step size increment when the throughput of the network device is not decreased in the second time period relative to the first time period and the data buffering delay is decreased or the number of data throughput channels is increased; and when the throughput of the network equipment is reduced relative to the first time period in the second time period, or the data buffering delay is not reduced and the number of data throughput channels is not increased, adjusting the second step size according to a second step size reduction.

In one possible design, before obtaining the network statistics corresponding to the first time period, the method further includes: it is determined through communication interface 901 that the difference in throughput over successive computations is less than the throughput fluctuation threshold.

As another form of the present embodiment, a computer-readable storage medium is provided, on which instructions are stored, and when executed, the instructions can perform the network parameter configuration method in the above method embodiment.

As another form of the present embodiment, there is provided a computer program product containing instructions that, when executed, can perform the network parameter configuration method in the above method embodiment.

As another form of this embodiment, a chip is provided, and when the chip runs, the network parameter configuration method in the foregoing method embodiment may be executed.

In the embodiments of the present application, the processor may be a general-purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic blocks disclosed in the embodiments of the present application. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in a processor.

In the embodiment of the present application, the memory may be a nonvolatile memory, such as a Hard Disk Drive (HDD) or a solid-state drive (SSD), and may also be a volatile memory, for example, a random-access memory (RAM). The memory is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to such. The memory in the embodiments of the present application may also be circuitry or any other device capable of performing a storage function for storing program instructions and/or data.

The technical solutions provided in the embodiments of the present application may be wholly or partially implemented by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, a network appliance, a terminal device or other programmable apparatus. The computer instructions may be stored on a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website, computer, server, or data center to another website, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., Digital Video Disk (DVD)), or a semiconductor medium, among others.

In the embodiments of the present application, the embodiments may refer to each other, for example, methods and/or terms between the embodiments of the method may refer to each other, for example, functions and/or terms between the embodiments of the apparatus and the embodiments of the method may refer to each other, without logical contradiction.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. A method for configuring network parameters, comprising:

acquiring network statistical data corresponding to a first time period, wherein the network statistical data comprise network transmission characteristics of network equipment in the first time period and a first value corresponding to a specified network parameter;

determining a second value corresponding to the specified network parameter according to the network statistical data;

configuring the specified network parameter of the network device to the second value for a second time period, the second time period being a time period after the first time period.

2. The method of claim 1, wherein the network transmission characteristics of the network device during the first time period comprise a throughput of the network device during the first time period, wherein when the network device throughput during the first time period increases relative to an initial time period, the initial time period being a time period prior to the first time period, and wherein determining the second value corresponding to the specified network parameter based on the network statistics comprises:

and modifying the first value according to the first adjusting direction and the first step length to obtain the second value.

3. The method of claim 2, wherein the method further comprises:

adjusting the first step size by a first step size increment when the throughput of the network device increases relative to the first time period in the second time period;

when the throughput of the network device does not increase in the second time period relative to the first time period, adjusting the first step size by a first step size decrement.

4. The method of claim 1, wherein the network transmission characteristics of the network device during the first time period comprise a throughput of the network device during the first time period, wherein the initial time period is a time period prior to the first time period when the network device throughput during the first time period has not risen relative to an initial time period, and wherein determining the second value corresponding to the specified network parameter based on the network statistics comprises:

determining an initial value corresponding to the specified network parameter in an initial time period; modifying the initial value according to a second adjusting direction and a second step length to obtain a second value; or

5. The method of claim 4, wherein the method further comprises:

when the throughput of the network device does not increase in the second time period relative to the first time period, or the data buffering delay is not reduced and the number of data throughput channels is not increased, configuring the specified network parameter of the network device in a third time period to be the first value, wherein the third time period is a time period after the second time period.

6. The method of claim 4 or 5, wherein the method further comprises:

when the throughput of the network equipment is not reduced in the second time period relative to the first time period, and the data caching delay is reduced or the number of data throughput channels is increased, adjusting the second step size according to a second step size increment;

and when the throughput of the network equipment is reduced relative to the first time period in the second time period, or the data buffering delay is not reduced and the number of data throughput channels is not increased, adjusting the second step size according to a second step size reduction.

7. The method of any one of claims 1-6, wherein prior to obtaining network statistics corresponding to the first time period, the method further comprises:

and determining that the difference value of the throughputs calculated for a plurality of times is smaller than the throughput fluctuation threshold value.

8. A communications apparatus, the apparatus comprising:

a memory for storing program instructions;

a communication interface for receiving and transmitting data;

a processor for calling program instructions stored in said memory to perform the following method:

acquiring network statistical data corresponding to a first time period through a communication interface, wherein the network statistical data comprise network transmission characteristics of network equipment in the first time period and a first value corresponding to a specified network parameter;

9. The communications apparatus of claim 8, wherein the network transmission characteristics of the network device during the first time period comprise a throughput of the network device during the first time period, wherein when the network device throughput during the first time period increases relative to an initial time period, the initial time period being a time period prior to the first time period, and wherein determining the second value for the specified network parameter based on the network statistics comprises:

10. The communications apparatus of claim 9, the method further comprising:

11. The communications apparatus of claim 8, wherein the network transmission characteristics of the network device during the first time period comprise a throughput of the network device during the first time period, wherein the initial time period is a time period prior to the first time period when the network device throughput during the first time period has not risen relative to an initial time period, and wherein determining the second value for the specified network parameter based on the network statistics comprises:

12. The communications apparatus of claim 11, the method further comprising:

13. The communication apparatus according to claim 11 or 12, wherein the method further comprises:

14. The communications device of any of claims 8-13, wherein prior to obtaining network statistics corresponding to the first time period, the method further comprises:

determining, by the communication interface, that the difference in throughput over successive times is less than the throughput fluctuation threshold.

15. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a computer program which, when read and executed by one or more processors, implements the method of any one of claims 1-7.

16. A chip, characterized in that it, when running, implements the method according to any one of claims 1-7.