Method, device and computer readable storage medium for adjusting data forwarding queue
By inputting the status parameters of the data forwarding queue into the pre-trained machine learning model, dynamically adjusting the queue control parameters, the problem of low operation and maintenance efficiency of intermediate network equipment is solved, and the automated management of the data forwarding queue is realized, which alleviates network congestion and guarantees data Forwarding performance.
Patent Information
- Application Number
- CN202110264211.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-03-11
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2041-03-11
AI Technical Summary
The prior art is difficult to effectively improve the operation and maintenance efficiency of data forwarding queues in intermediate network equipment. The traditional method requires a lot of manual operation and cannot dynamically adapt to real-time traffic changes, resulting in deterioration of network data forwarding quality.
By entering the state parameters of the data forwarding queue into the pre-trained machine learning model, obtain queue control parameters, dynamically adjust the queue status, such as slowing down the data transmission rate when the queue depth exceeds the active congestion notification waterline, utilizing deep learning networks and heuristic search The algorithm identifies the business type and sets appropriate queue control parameters.
It realizes automatic dynamic adjustment of data forwarding queues, improves the operation and maintenance efficiency of intermediate network equipment, alleviates network congestion and ensures data forwarding delay and throughput.
Smart Images

Figure CN115086236B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of computer technology, and in particular to a method and device for adjusting a data forwarding queue and a non-transitory computer-readable storage medium. Background Art
[0002] In a data center network, ECN (Explicit Congestion Notification) technology is usually used in combination with PFC (Power Factor Correction) technology to operate and maintain various devices in the data center network.
[0003] ECN is a flow-based end-to-end flow control technology. Through the cooperation of the TCP (Transmission Control Protocol) sender, receiver and intermediate network devices, it can sense the congestion of the forwarding queues corresponding to each port in the intermediate network device, and actively slow down the rate at which the TCP sender sends data to the intermediate network device, thereby avoiding network congestion and causing data packet loss, and ensuring the lossless transmission requirements of low-latency services. Summary of the invention
[0004] A technical problem solved by the present disclosure is how to improve the operation and maintenance efficiency of the intermediate network device where the forwarding queue is located.
[0005] According to one aspect of the present disclosure, a method for adjusting a data forwarding queue is provided, comprising: inputting state parameters of the data forwarding queue into a pre-trained machine learning model to obtain queue control parameters of the data forwarding queue; wherein the state parameters include forwarding delay and queue depth of the data forwarding queue, and the queue control parameters include an active congestion notification watermark and a data forwarding queue cache; setting queue control parameters for the data forwarding queue to adjust the state parameters of the data forwarding queue; wherein, when the queue depth does not exceed the active congestion notification watermark, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a first rate; when the queue depth exceeds the active congestion notification watermark, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a second rate; the second rate is greater than the first rate.
[0006] In some embodiments, inputting the state parameters of the data forwarding queue into a pre-trained machine learning model includes: when the forwarding delay is greater than a first threshold or less than a second threshold, or when the queue depth is greater than a third threshold or less than a fourth threshold, inputting the state parameters into the pre-trained machine learning model, wherein the first threshold is greater than the second threshold, and the third threshold is greater than the fourth threshold.
[0007] In some embodiments, the state parameters of the data forwarding queue are input into a pre-trained machine learning model, and obtaining the queue control parameters of the data forwarding queue includes: inputting the state parameters of the data forwarding queue into a pre-trained first deep learning network to identify the business type of the state parameters; if the business type of the state parameters is an existing business, inputting the state parameters into a pre-trained second deep learning network to obtain the first queue control parameters of the data forwarding queue, the first queue control parameters include a first active congestion notification waterline and a first data forwarding queue cache; if the business type of the state parameters is a new business, processing the state parameters using a heuristic search algorithm to obtain the second queue control parameters of the data forwarding queue, the second queue control parameters include a second active congestion notification waterline and a second data forwarding queue cache.
[0008] In some embodiments, the method further includes: using historical state parameters of the data forwarding queue and annotated business type labels to train the first deep learning network so that the first deep learning network can identify the business type of the state parameters, wherein the business type labels include existing business type labels and new business type labels.
[0009] In some embodiments, the first deep learning network is a convolutional neural network.
[0010] In some embodiments, the method further includes: using historical state parameters of the data forwarding queue and empirical values of queue control parameters to train the second deep learning network, so that the first deep learning network can output the first queue control parameters according to the input state parameters.
[0011] In some embodiments, the second deep learning network is a deep Q-network.
[0012] According to another aspect of the present disclosure, there is provided a device for adjusting a data forwarding queue, comprising: a state parameter input module, configured to: input the state parameters of the data forwarding queue into a pre-trained machine learning model to obtain queue control parameters of the data forwarding queue; wherein the state parameters include the forwarding delay and queue depth of the data forwarding queue, and the queue control parameters include an active congestion notification waterline and a data forwarding queue cache; a control parameter setting module, configured to: set queue control parameters for the data forwarding queue to adjust the state parameters of the data forwarding queue; wherein, when the queue depth does not exceed the active congestion notification waterline, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a first rate; when the queue depth exceeds the active congestion notification waterline, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a second rate; the second rate is greater than the first rate.
[0013] In some embodiments, the state parameter input module is configured to input the state parameter into a pre-trained machine learning model when the forwarding delay is greater than a first threshold or less than a second threshold, or when the queue depth is greater than a third threshold or less than a fourth threshold, wherein the first threshold is greater than the second threshold and the third threshold is greater than the fourth threshold.
[0014] In some embodiments, the state parameter input module is configured to: input the state parameters of the data forwarding queue into a pre-trained first deep learning network to identify the business type of the state parameters; if the business type of the state parameters is an existing business, input the state parameters into a pre-trained second deep learning network to obtain the first queue control parameters of the data forwarding queue, the first queue control parameters include a first active congestion notification waterline and a first data forwarding queue cache; if the business type of the state parameters is a new business, process the state parameters using a heuristic search algorithm to obtain the second queue control parameters of the data forwarding queue, the second queue control parameters include a second active congestion notification waterline and a second data forwarding queue cache.
[0015] In some embodiments, the device also includes a first network training module, which is configured to: use the historical state parameters of the data forwarding queue and the annotated business type labels to train the first deep learning network, so that the first deep learning network can identify the business type of the state parameters, wherein the business type labels include existing business type labels and new business type labels.
[0016] In some embodiments, the first deep learning network is a convolutional neural network.
[0017] In some embodiments, the device also includes a second network training module, which is configured to: use the historical state parameters of the data forwarding queue and the empirical values of the queue control parameters to train the second deep learning network, so that the first deep learning network can output the first queue control parameters based on the input state parameters.
[0018] In some embodiments, the second deep learning network is a deep Q-network.
[0019] According to another aspect of the present disclosure, another device for adjusting a data forwarding queue is provided, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute the aforementioned method for adjusting a data forwarding queue based on instructions stored in the memory.
[0020] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the aforementioned method of adjusting a data forwarding queue is implemented.
[0021] The present disclosure can dynamically and automatically set the queue control parameters of the data forwarding queue, thereby realizing dynamic and automatic adjustment of the queue state parameters of the data forwarding queue, thereby improving the operation and maintenance efficiency of the intermediate network device where the forwarding queue is located.
[0022] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or related technologies, the drawings required for use in the embodiments or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0024] Figure 1 A flowchart of a method for adjusting a data forwarding queue according to some embodiments of the present disclosure is shown.
[0025] Figure 2 A flowchart of some embodiments of obtaining queue control parameters using a machine learning model is shown.
[0026] Figure 3 A schematic diagram showing an application example of the method for adjusting a data forwarding queue disclosed in the present invention.
[0027] Figure 4 A schematic diagram of the structure of an apparatus for adjusting a data forwarding queue according to some embodiments of the present disclosure is shown.
[0028] Figure 5 A schematic structural diagram of an apparatus for adjusting a data forwarding queue according to some other embodiments of the present disclosure is shown. DETAILED DESCRIPTION
[0029] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present disclosure and its application or use. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.
[0030] The inventors have found that by setting the queue control parameters of the data forwarding queue, the queue state parameters of the data forwarding queue can be adjusted. For example, when the queue depth of the data forwarding queue exceeds the active congestion notification waterline, the data forwarding queue notifies the data sender to slow down the rate at which the data sender sends data to the intermediate network device, thereby reducing the queue depth of the data forwarding queue. However, if the queue control parameters of the data forwarding queue are set to fixed values, the data forwarding queue cannot dynamically adapt to changes in the flow of real-time data, which may cause the data forwarding quality of the network to deteriorate and fail to guarantee business performance; if the queue control parameters of the data forwarding queue are adjusted in real time, experienced engineers are required to perform a large amount of manual operations to try the set values of the queue control parameters. Therefore, the traditional method of adjusting the data forwarding queue results in low operation and maintenance efficiency of the intermediate network device where the forwarding queue is located.
[0031] In view of this, the present disclosure provides a method for adjusting a data forwarding queue, which can improve the operation and maintenance efficiency of an intermediate network device where the forwarding queue is located.
[0032] First combine Figure 1 Some embodiments of the method of adjusting a data forwarding queue of the present disclosure are described.
[0033] Figure 1 FIG. 1 is a flow chart showing a method for adjusting a data forwarding queue according to some embodiments of the present disclosure. Figure 1 As shown, these embodiments include steps S101 to S102.
[0034] In step S101, the state parameters of the data forwarding queue are input into a pre-trained machine learning model to obtain queue control parameters of the data forwarding queue.
[0035] Among them, the state parameters include the forwarding delay and queue depth of the data forwarding queue, and the queue control parameters include the active congestion notification watermark and the data forwarding queue buffer. After pre-training, the machine learning model can output the queue control parameters of the data forwarding queue based on the input state parameters of the data forwarding queue.
[0036] In some embodiments, when the forwarding delay is greater than a first threshold or less than a second threshold, or when the queue depth is greater than a third threshold or less than a fourth threshold, the state parameters are input into a pre-trained machine learning model, wherein the first threshold is greater than the second threshold and the third threshold is greater than the fourth threshold.
[0037] When the state parameters of the data forwarding queue meet certain threshold conditions, the state parameters are input into a pre-trained machine learning model for processing, which can effectively control the trigger conditions of the predicted queue control parameters and reduce the workload of the machine learning model.
[0038] In step S102, queue control parameters are set for the data forwarding queue to adjust state parameters of the data forwarding queue.
[0039] For example, when the queue depth does not exceed the active congestion notification watermark, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a first rate. When the queue depth exceeds the active congestion notification watermark, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a second rate, where the second rate is greater than the first rate.
[0040] The above embodiment can dynamically and automatically set the queue control parameters of the data forwarding queue through artificial intelligence technology, thereby realizing dynamic and automatic adjustment of the queue state parameters of the data forwarding queue, thereby improving the operation and maintenance efficiency of the intermediate network device where the forwarding queue is located. At the same time, the above embodiment can more accurately set the data forwarding queue cache and active congestion notification waterline, which can alleviate network congestion while ensuring the network performance of the intermediate network device in terms of data forwarding delay and throughput.
[0041] Combine the following Figure 2 Some embodiments of using machine learning models to obtain queue control parameters are described.
[0042] Figure 2 The flowchart of some embodiments of using machine learning models to obtain queue control parameters is shown. Figure 2 As shown, these embodiments include steps S2012 to S2014.
[0043] In step S2012, the state parameters of the data forwarding queue are input into a pre-trained first deep learning network to identify the business type of the state parameters.
[0044] The first deep learning network may be, for example, a convolutional neural network with classification function.
[0045] If the service type of the state parameter is an existing service, step S2013 is executed. In step S2013, the state parameter is input into a pre-trained second deep learning network to obtain a first queue control parameter of the data forwarding queue. The first queue control parameter includes a first active congestion notification waterline and a first data forwarding queue buffer.
[0046] If the service type of the state parameter is a new service, step S2014 is executed. In step S2014, the state parameter is processed using a heuristic search algorithm to obtain a second queue control parameter of the data forwarding queue. The second queue control parameter includes a second active congestion notification watermark and a second data forwarding queue buffer.
[0047] When the state parameters are processed by using a heuristic search algorithm, an evaluation function can be constructed according to the forwarding delay, congestion packet loss rate and queue occupancy rate of the data forwarding queue.
[0048] If the second active congestion notification watermark is set too small, the data sender will send data to the data forwarding queue of the intermediate network device at a lower rate in most cases, which will result in a higher forwarding delay of the data forwarding queue. If the second active congestion notification watermark is set too large, the data sender will send data to the data forwarding queue of the intermediate network device at a higher rate in most cases, which will result in a higher congestion packet loss rate of the data forwarding queue.
[0049] If the second data forwarding queue buffer is set too small, the congestion packet loss rate of the data forwarding queue will be high; if the second data forwarding queue buffer is set too large, the queue occupancy rate of the data forwarding queue will be low, resulting in a waste of resources.
[0050] It can be seen that the setting of the second active congestion notification watermark and the second data forwarding queue cache will affect the forwarding delay, congestion packet loss rate and queue occupancy rate of the data forwarding queue. In view of this, the function value of the evaluation function is proportional to the forwarding delay, congestion packet loss rate and queue occupancy rate of the data forwarding queue, and when the heuristic search algorithm is used for processing, the second active congestion notification watermark and the second data forwarding queue cache that minimize the function value of the evaluation function are sought.
[0051] In some embodiments, step S2010 is further included. In step S2010, the first deep learning network is trained using the historical state parameters of the data forwarding queue and the annotated service type labels, so that the first deep learning network can identify the service type of the state parameters. The service type labels include existing service type labels and new service type labels.
[0052] In some embodiments, step S2011 is further included. In step S2011, the second deep learning network is trained using the historical state parameters of the data forwarding queue and the empirical values of the queue control parameters, so that the first deep learning network can output the first queue control parameters according to the input state parameters.
[0053] Step S2010 lists the process of training the first deep learning network by supervised learning, and step S2011 lists the process of training the second deep learning network by supervised learning. In addition, the second deep learning network can be trained by unsupervised learning, or a deep Q network can be used as the second deep learning network. When a deep Q network is used as the second deep learning network, the idea is similar to the aforementioned heuristic search algorithm. The state parameters of the data forwarding queue are used as the current state s0, the first queue control parameters are used as action a0, and the state parameters of the data forwarding queue after the first queue control parameters are set are used as the next state s1. At the same time, a reward function r is set, and the function value of the reward function r is inversely proportional to the forwarding delay, congestion packet loss rate and queue occupancy rate of the data forwarding queue.
[0054] Those skilled in the art will appreciate that the heuristic search algorithm and the deep Q network in these embodiments may be implemented using existing technologies, and the heuristic search algorithm and the deep Q network themselves are not the inventive points of the present disclosure.
[0055] The above-mentioned embodiment can relatively accurately set the data forwarding queue cache and the active congestion notification waterline, and can alleviate network congestion while ensuring the network performance of the intermediate network equipment in terms of data forwarding delay and throughput.
[0056] Combine the following Figure 3 An application example of the method for adjusting a data forwarding queue disclosed in the present invention is described.
[0057] Figure 3 A schematic diagram showing an application example of the method for adjusting a data forwarding queue disclosed in the present invention is shown. Figure 3 As shown, the application example includes a data forwarding queue, an artificial intelligence chip, and a control parameter configuration device located in an intermediate network device. The workflow of the application example is as follows:
[0058] (1) After receiving the data packet, the data forwarding queue will determine whether to enable the INT (Inband Network Telemetry) function based on the status parameters of the data forwarding queue. If the INT function is enabled, the forwarding delay and queue depth of the data forwarding queue will be encapsulated into the INT metadata, and in the data push Postcard mode, the INT metadata will be directly pushed to the artificial intelligence chip on the local jump device (intermediate network device).
[0059] Specific examples of INT metadata are as follows, including:
[0060] Version number Ver, occupies 2 bits;
[0061] Flags, occupies 9 bits, including the copy permission flag and other reserved bits;
[0062] The number of collected data items is Instruction Count, which occupies 5 bits;
[0063] Max Hop Count, occupies 8 bits and indicates the maximum number of metadata that can be encapsulated in a message;
[0064] Total Hop Count, occupies 8 bits and indicates the number of metadata encapsulated in the current message.
[0065] The specific content of the metadata is Instruction Bitmap, which occupies 8 bits.
[0066] The 8 bits of the Instruction Bitmap field correspond to 8 data options, and each bit represents that the data needs to be collected into the metadata. The field settings of the Instruction Bitmap are shown in Table 1.
[0067] Table 1
[0068] Fields Data Options bit0 Switch ID bit1 Inbound port number bit2 Data forwarding timestamp bit3 Queue Depth bit4 Data inflow timestamp bit5 Queue occupancy bit6 Queue congestion status bit7 Outbound port link utilization
[0069] The data forwarding queue supports configuration based on thresholds to trigger INT metadata push, which can effectively control the triggering conditions of INT metadata push, reduce the processing load of the AI chip, and avoid the impact of a large amount of in-band measurement data on the AI chip.
[0070] (2) The AI chip identifies the state parameters in the INT metadata based on CNN (Convolutional Neural Networks). The state parameters of existing business types are processed using the DQN (Deep QNetwork) algorithm; the state parameters of new business types are processed using a heuristic search algorithm.
[0071] In artificial intelligence chips, classifying and processing status parameters through different algorithms can greatly improve the efficiency of artificial intelligence processing. According to the queue status parameters in the triggered INT push data, the appropriate queue control parameters can be analyzed.
[0072] (3) The control parameter configuration device forwards the queue control parameters obtained by analysis to the data forwarding queue, and configures the queue control parameters for the data forwarding queue. The queue control parameters may specifically include ECN waterline, data forwarding queue buffer, etc.
[0073] When congestion occurs, the data forwarding queue cache is automatically increased to cache burst data traffic to avoid negative impact on service quality caused by packet loss and retransmission. The ECN waterline is automatically lowered to notify the data sender to reduce the sending rate to resolve network congestion.
[0074] When the congestion disappears according to the data forwarding queue cache, on the one hand, the queue cache is automatically lowered to ensure the utilization of cache resources; on the other hand, the ECN waterline is raised to notify the data sender to increase the sending rate, reduce data forwarding delay, and improve business response rate and data throughput.
[0075] Those skilled in the art will understand that edge computing networks usually carry services that are sensitive to data forwarding delays. In the case of unbalanced service traffic, network packet loss and retransmission have a greater impact on the user experience of these services. At the same time, the computer room of the edge computing network is relatively remote and the distribution locations are relatively scattered. When these services are stuck or even interrupted, it is difficult for operation and maintenance personnel to locate network congestion and data packet loss problems in real time and accurately. In the above application example, integrating artificial intelligence chips into the intermediate network equipment of the edge computing network can realize the automated operation and maintenance of the edge computing network, effectively improving the operation and maintenance efficiency of the edge computing network.
[0076] Combine the following Figure 4 Some embodiments of the apparatus for adjusting a data forwarding queue according to the present disclosure are described.
[0077] Figure 4 FIG. 2 is a schematic diagram showing a structure of an apparatus for adjusting a data forwarding queue in some embodiments of the present disclosure. Figure 4 As shown, the device 40 for adjusting the data forwarding queue includes: a state parameter input module 402, configured to: input the state parameters of the data forwarding queue into a pre-trained machine learning model to obtain the queue control parameters of the data forwarding queue; wherein the state parameters include the forwarding delay and queue depth of the data forwarding queue, and the queue control parameters include the active congestion notification waterline and the data forwarding queue cache; a control parameter setting module 403, configured to: set the queue control parameters for the data forwarding queue to adjust the state parameters of the data forwarding queue; wherein, when the queue depth does not exceed the active congestion notification waterline, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a first rate; when the queue depth exceeds the active congestion notification waterline, the data forwarding queue notifies the data sender to send data to the data forwarding queue at a second rate; the second rate is greater than the first rate.
[0078] The above embodiment can dynamically and automatically set the queue control parameters of the data forwarding queue through artificial intelligence technology, thereby realizing dynamic and automatic adjustment of the queue state parameters of the data forwarding queue, thereby improving the operation and maintenance efficiency of the intermediate network device where the forwarding queue is located. At the same time, the above embodiment can more accurately set the data forwarding queue cache and active congestion notification waterline, which can alleviate network congestion while ensuring the network performance of the intermediate network device in terms of data forwarding delay and throughput.
[0079] In some embodiments, the state parameter input module 402 is configured to: input the state parameter into a pre-trained machine learning model when the forwarding delay is greater than a first threshold or less than a second threshold, or when the queue depth is greater than a third threshold or less than a fourth threshold, wherein the first threshold is greater than the second threshold, and the third threshold is greater than the fourth threshold.
[0080] When the state parameters of the data forwarding queue meet certain threshold conditions, the state parameters are input into a pre-trained machine learning model for processing, which can effectively control the trigger conditions of the predicted queue control parameters and reduce the workload of the machine learning model.
[0081] In some embodiments, the state parameter input module 402 is configured to: input the state parameters of the data forwarding queue into a pre-trained first deep learning network to identify the business type of the state parameters; if the business type of the state parameters is an existing business, input the state parameters into a pre-trained second deep learning network to obtain the first queue control parameters of the data forwarding queue, the first queue control parameters include a first active congestion notification waterline and a first data forwarding queue cache; if the business type of the state parameters is a new business, process the state parameters using a heuristic search algorithm to obtain the second queue control parameters of the data forwarding queue, the second queue control parameters include a second active congestion notification waterline and a second data forwarding queue cache.
[0082] In some embodiments, the device 40 also includes a first network training module 400, which is configured to: use the historical state parameters of the data forwarding queue and the annotated business type labels to train the first deep learning network, so that the first deep learning network can identify the business type of the state parameters, wherein the business type labels include existing business type labels and new business type labels.
[0083] In some embodiments, the first deep learning network is a convolutional neural network.
[0084] In some embodiments, the device 40 also includes a second network training module 401, which is configured to: use the historical state parameters of the data forwarding queue and the empirical values of the queue control parameters to train the second deep learning network, so that the first deep learning network can output the first queue control parameters based on the input state parameters.
[0085] In some embodiments, the second deep learning network is a deep Q-network.
[0086] The above-mentioned embodiment can relatively accurately set the data forwarding queue cache and the active congestion notification waterline, and can alleviate network congestion while ensuring the network performance of the intermediate network equipment in terms of data forwarding delay and throughput.
[0087] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0088] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0089] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0090] Combine the following Figure 5 Other embodiments of the apparatus for adjusting a data forwarding queue according to the present disclosure are described.
[0091] Figure 5 FIG. 2 is a schematic diagram showing a structure of an apparatus for adjusting a data forwarding queue in some other embodiments of the present disclosure. Figure 5As shown, the device 50 for adjusting the data forwarding queue includes: a memory 510 and a processor 520 coupled to the memory 510, and the processor 520 is configured to execute the method for adjusting the data forwarding queue in any of the aforementioned embodiments based on instructions stored in the memory 510.
[0092] The memory 510 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs.
[0093] The device 50 for adjusting the data forwarding queue may also include an input / output interface 530, a network interface 540, a storage interface 550, etc. These interfaces 530, 540, 550, the memory 510, and the processor 520 may be connected, for example, via a bus 560. The input / output interface 530 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 540 provides a connection interface for various networked devices. The storage interface 550 provides a connection interface for external storage devices such as SD cards and USB flash drives.
[0094] The present disclosure also includes a non-transitory computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the method for adjusting the data forwarding queue in any of the aforementioned embodiments.
[0095] If the aforementioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
[0096] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0099] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0100] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0101] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.
Claims
1. A method for adjusting a data forwarding queue, comprising: The state parameters of the data forwarding queue are input into the pre-trained machine learning model to obtain the queue control parameters of the data forwarding queue, including: Inputting the state parameters of the data forwarding queue into a pre-trained first deep learning network to identify the service type of the state parameters; If the service type of the state parameter is an existing service, input the state parameter into a pre-trained second deep learning network to obtain a first queue control parameter of a data forwarding queue, wherein the first queue control parameter includes a first active congestion notification waterline and a first data forwarding queue buffer; If the service type of the state parameter is a new service, the state parameter is processed by using a heuristic search algorithm to obtain a second queue control parameter of the data forwarding queue, where the second queue control parameter includes a second active congestion notification waterline and a second data forwarding queue cache; Wherein, the state parameters include the forwarding delay and queue depth of the data forwarding queue, and the queue control parameters include the active congestion notification waterline and the data forwarding queue buffer; The queue control parameters are set for the data forwarding queue to adjust the state parameters of the data forwarding queue; wherein, when the queue depth does not exceed the active congestion notification waterline, the data forwarding queue notifies the data sending end to send data to the data forwarding queue at a first rate; when the queue depth exceeds the active congestion notification waterline, the data forwarding queue notifies the data sending end to send data to the data forwarding queue at a second rate; the second rate is greater than the first rate.
2. The method according to claim 1, wherein: The inputting of the state parameters of the data forwarding queue into the pre-trained machine learning model includes: when the forwarding delay is greater than a first threshold or less than a second threshold, or when the queue depth is greater than a third threshold or less than a fourth threshold, inputting the state parameters into the pre-trained machine learning model, wherein the first threshold is greater than the second threshold, and the third threshold is greater than the fourth threshold.
3. The method according to claim 1, further comprising: The first deep learning network is trained using historical state parameters of the data forwarding queue and the annotated business type labels, so that the first deep learning network can identify the business type of the state parameters, wherein the business type labels include existing business type labels and new business type labels.
4. The method according to claim 1, wherein: The first deep learning network is a convolutional neural network.
5. The method according to claim 1, further comprising: The second deep learning network is trained using historical state parameters of the data forwarding queue and empirical values of queue control parameters, so that the second deep learning network can output the first queue control parameters according to the input state parameters.
6. The method according to claim 1, wherein: The second deep learning network is a deep Q network.
7. A device for adjusting a data forwarding queue, comprising: A state parameter input module is configured to: input the state parameter of the data forwarding queue into a pre-trained machine learning model to obtain the queue control parameter of the data forwarding queue, including: inputting the state parameter of the data forwarding queue into a pre-trained first deep learning network to identify the service type of the state parameter; if the service type of the state parameter is an existing service, inputting the state parameter into a pre-trained second deep learning network to obtain the first queue control parameter of the data forwarding queue, the first queue control parameter including the first active congestion notification watermark and the first data forwarding queue cache; if the service type of the state parameter is a new service, processing the state parameter using a heuristic search algorithm to obtain the second queue control parameter of the data forwarding queue, the second queue control parameter including the second active congestion notification watermark and the second data forwarding queue cache; wherein the state parameter includes the forwarding delay and queue depth of the data forwarding queue, and the queue control parameter includes the active congestion notification watermark and the data forwarding queue cache; The control parameter setting module is configured to: set the queue control parameters for the data forwarding queue to adjust the state parameters of the data forwarding queue; wherein, when the queue depth does not exceed the active congestion notification waterline, the data forwarding queue notifies the data sending end to send data to the data forwarding queue at a first rate; when the queue depth exceeds the active congestion notification waterline, the data forwarding queue notifies the data sending end to send data to the data forwarding queue at a second rate; the second rate is greater than the first rate.
8. The device according to claim 7, wherein: The state parameter input module is configured to: input the state parameter into a pre-trained machine learning model when the forwarding delay is greater than a first threshold or less than a second threshold, or when the queue depth is greater than a third threshold or less than a fourth threshold, wherein the first threshold is greater than the second threshold, and the third threshold is greater than the fourth threshold.
9. The apparatus according to claim 7, further comprising a first network training module, configured to: train the first deep learning network using the historical state parameters of the data forwarding queue and the annotated service type labels, so that the first deep learning network can identify the service type of the state parameters, wherein: The service type labels include existing service type labels and new service type labels.
10. The device according to claim 7, wherein: The first deep learning network is a convolutional neural network.
11. The device according to claim 7 further includes a second network training module, which is configured to: use the historical state parameters of the data forwarding queue and the empirical values of the queue control parameters to train the second deep learning network, so that the second deep learning network can output the first queue control parameters based on the input state parameters.
12. The device according to claim 7, wherein: The second deep learning network is a deep Q network.
13. A device for adjusting a data forwarding queue, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the method for adjusting the data forwarding queue according to any one of claims 1 to 6 based on instructions stored in the memory.
14. A non-transitory computer-readable storage medium, wherein: The non-transitory computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the method for adjusting a data forwarding queue according to any one of claims 1 to 6 is implemented.
