CA3151273A1 - Method, apparatus, electronic device, and computer-readable medium for traffic control - Google Patents
Method, apparatus, electronic device, and computer-readable medium for traffic controlInfo
- Publication number
- CA3151273A1 CA3151273A1 CA3151273A CA3151273A CA3151273A1 CA 3151273 A1 CA3151273 A1 CA 3151273A1 CA 3151273 A CA3151273 A CA 3151273A CA 3151273 A CA3151273 A CA 3151273A CA 3151273 A1 CA3151273 A1 CA 3151273A1
- Authority
- CA
- Canada
- Prior art keywords
- prediction
- traffic
- prediction result
- cycle
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000003062 neural network model Methods 0.000 claims abstract description 41
- 238000011217 control strategy Methods 0.000 claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 210000000225 synapse Anatomy 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 238000004891 communication Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000007796 conventional method Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
A method, an apparatus, an electronic device, and a computer-readable medium for traffic control are disclosed. The method includes: collecting request data of a target application; predicting a traffic according to the request data using a traffic prediction model, so as to obtain a prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model; generating a traffic control strategy according to the prediction result related to the target cycle; and controlling the traffic of the target application in the target cycle according to the traffic control strategy. The neural network model provides precise traffic prediction, so that traffic monitoring strategies specific to prediction results can be generated, thereby facilitating reasonable distribution of the traffic, and enabling warning of potential traffic bursts to be given according to prediction results.
Description
METHOD, APPARATUS, ELECTRONIC DEVICE, AND COMPUTER-READABLE
MEDIUM FOR TRAFFIC CONTROL
BACKGROUND OF THE INVENTION
Technical Field [0001] The present invention relates to the technical field of computers, and more particularly to a method, an apparatus, an electronic device, and a computer-readable medium for traffic control.
Description of Related Art
MEDIUM FOR TRAFFIC CONTROL
BACKGROUND OF THE INVENTION
Technical Field [0001] The present invention relates to the technical field of computers, and more particularly to a method, an apparatus, an electronic device, and a computer-readable medium for traffic control.
Description of Related Art
[0002] With the rapid development of network technologies, applications based on networks keep increasing in both number and complexity. These applications continuously devour network resources, leading to network congestion. For solving network congestion, network traffic control has become a hot topic. Network traffic control is measures to control network data traffic using software or hardware. To implement it, the first step is to build up a traffic control strategy, and then control traffic according to this strategy. In the prior art, common traffic control strategies may depend on manual maintenance by operations and maintenance personnel or may involve using a linear prediction model to make linear prediction of network traffic so as to realize dynamic monitoring of network traffic.
[0003] In the known traffic control strategies as describe previously, manual maintenance is less flexible and has relatively low traffic usage. While the practice of using a linear prediction model to set a traffic control strategy is more flexible than manual maintenance, a linear prediction has inherent errors, which can even aggravate when used in complex scenes, leading to waste of resources.
SUMMARY OF THE INVENTION
Date Recue/Date Received 2022-03-08
SUMMARY OF THE INVENTION
Date Recue/Date Received 2022-03-08
[0004] The present invention provides a method, an apparatus, an electronic device, and a computer-readable medium for traffic control, which can precisely predict traffic situations and implement traffic control schemes specific to the predicted situations.
[0005] The present invention provides the following schemes.
[0006] In a first aspect, the present invention provides a method for traffic control, which comprises:
[0007] collecting request data of a target application;
[0008] predicting a traffic according to the request data using a traffic prediction model, so as to obtain a prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model;
[0009] generating a traffic control strategy according to the prediction result related to the target cycle; and
[0010] controlling the traffic of the target application in the target cycle according to the traffic control strategy.
[0011] Further, the step of predicting a traffic according to the request data using a traffic prediction model, so as to obtain a prediction result related to a target cycle comprises:
[0012] entering the request data to said traffic prediction models of different prediction cycles for traffic prediction, so as to obtain the prediction results related to at least two different cycles; and
[0013] analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0014] Further, the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
[0015] using the prediction result of a short said cycle to correct the prediction result of a long said cycle; and Date Recue/Date Received 2022-03-08
[0016] taking the corrected prediction result of the longer cycle as the prediction result of the target cycle.
[0017] Further, the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
[0018] performing fit analysis on the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0019] Further, the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
[0020] calculating a mean of the prediction results related to the different cycles; and
[0021] taking the mean of the prediction results related to the different cycles as the prediction result of the target cycle.
[0022] Further, the traffic prediction model is constructed through:
[0023] acquiring a sample training set for traffic prediction, which includes input samples and result samples;
[0024] inputting the input sample to the initialized neural network model, so as to obtain a prediction result set;
[0025] comparing the prediction result set to the result samples, so as to obtain prediction errors;
[0026] adjusting the neural network model according to the prediction errors;
[0027] using the adjusted neural network model to repeatedly make prediction based on the input samples; and
[0028] comparing the prediction errors to standard error conditions, and when the prediction errors satisfy the standard error conditions, determining that the neural network model corresponding to the prediction errors is the traffic prediction model.
[0029] Further, the step of adjusting the neural network model according to the prediction errors comprises:
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
[0030] correcting a weight of every synapse in the neural network model according to the prediction errors using a backpropagation algorithm.
[0031] In a second aspect, the present invention provides an apparatus for traffic control, which comprises:
[0032] a data-collecting unit, for collecting request data of a target application;
[0033] a traffic-predicting unit, for predicting a traffic according to the request data using a traffic prediction model, so as to obtain a traffic prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model;
[0034] a strategy-generating unit, for generating traffic control strategy according to the traffic prediction result related to the target cycle; and
[0035] a traffic-controlling unit, for controlling the traffic of the target application in the target cycle according to the traffic control strategy.
[0036] Further, the traffic-predicting unit comprises:
[0037] a prediction module, entering the request data to said traffic prediction models of different prediction cycles for traffic prediction, so as to obtain the prediction results related to at least two different cycles; and
[0038] an analysis module, for analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0039] Further, the analysis module is specifically for:
[0040] using the prediction result of a short said cycle to correct the prediction result of a long said cycle; and
[0041] taking the corrected prediction result of the longer cycle as the prediction result of the target cycle.
[0042] Further, the analysis module is specifically for:
[0043] performing fit analysis on the prediction results related to the different cycles, so as to Date Recue/Date Received 2022-03-08 obtain the prediction result related to the target cycle.
[0044] Further, the analysis module is specifically for:
[0045] calculating a mean of the prediction results related to the different cycles; and
[0046] taking the mean of the prediction results related to the different cycles as the prediction result of the target cycle.
[0047] Further, the apparatus further comprises a traffic prediction model training module, for:
[0048] acquiring a sample training set for traffic prediction, which includes input samples and result samples;
[0049] inputting the input sample to the initialized neural network model, so as to obtain a prediction result set;
[0050] comparing the prediction result set to the result samples, so as to obtain prediction errors;
[0051] adjusting the neural network model according to the prediction errors;
[0052] using the adjusted neural network model to repeatedly make prediction based on the input samples; and
[0053] comparing the prediction errors to standard error conditions, and when the prediction errors satisfy the standard error conditions, determining that the neural network model corresponding to the prediction errors is the traffic prediction model.
[0054] Further, the step of adjusting the neural network model according to the prediction errors comprises:
[0055] correcting a weight of every synapse in the neural network model according to the prediction errors using a backpropagation algorithm.
[0056] In a third aspect, the present invention provides an electronic device, which comprises:
[0057] one or more processors; and
[0058] a memory associated with the one or more processors, the memory storing program instructions, the program instructions, when read and executed by the one or more Date Recue/Date Received 2022-03-08 processors, executing any of the methods of the first aspect.
[0059] In a fourth aspect, the present invention provides a computer-readable medium, storing therein a computer program, wherein the program, when executed by a processor, implements any of the methods of the first aspect.
[0060] The embodiments of the present invention provide the following technical effects:
[0061] 1. In the embodiment of the present invention, the neural network model provides precise traffic prediction, so that traffic monitoring strategies specific to prediction results can be generated, thereby facilitating reasonable distribution of the traffic, and enabling warning of potential traffic bursts to be given according to prediction results;
[0062] 2. The technical scheme of the embodiment of the present invention performs analysis according to the traffic prediction results related to different prediction cycles so as to obtain a prediction result related to a target cycle, which is more precise than the prediction result obtained using the known linear prediction method; and
[0063] 3. The technical scheme of the embodiment of the present invention continuously trains and adjusts the traffic prediction model according to prediction errors, thereby eliminating the defects of the conventional methods about inability to predict sudden incidents.
[0064] Of course, it is not necessary to achieve all of the foregoing advantages for implementation of any product of the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
BRIEF DESCRIPTION OF THE DRAWINGS
[0065] To better illustrate the technical schemes as disclosed in the embodiments of the present Date Recue/Date Received 2022-03-08 invention, accompanying drawings referred in the description of the embodiments below are introduced briefly. It is apparent that the accompanying drawings as recited in the following description merely provide a part of possible embodiments of the present invention, and people of ordinary skill in the art would be able to obtain more drawings according to those provided herein without paying creative efforts, wherein:
[0066] FIG. 1 is a flowchart of a method according to one embodiment of the present invention;
[0067] FIG. 2 is a structural diagram of an apparatus according to one embodiment of the present invention; and
[0068] FIG. 3 is a structural diagram of a computer system according to one embodiment of the present invention.
DETAILED DESCRIPTION OF THE INVENTION
DETAILED DESCRIPTION OF THE INVENTION
[0069] The following description will be made with reference to the accompanying drawings of embodiments of the present invention and detail the technical schemes of the embodiments with clarity and completeness. It is obvious that the described embodiments are merely a part of all possible embodiments of the present invention, but not all of them.
Any other embodiment devised by a person of ordinary skill in the art based on the embodiments of the present invention shall be encompassed in the scope of the present invention.
Any other embodiment devised by a person of ordinary skill in the art based on the embodiments of the present invention shall be encompassed in the scope of the present invention.
[0070] In embodiments of the present invention, technical schemes based on neural network model prediction traffic and preforming traffic control according to prediction results are provided, which solve the problems of the prior-art traffic control schemes about inferior flexibility and low traffic usage caused by manual maintenance and about error-incurred waste of resources as seen in use of a linear prediction model for traffic prediction.
[0071] Some embodiments of the present invention will be described below to explain these schemes.
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
[0072] Embodiment 1
[0073] Referring to FIG. 1, a traffic control method specifically may comprise the following steps.
[0074] Si is about collecting request data of a target application.
[0075] The request data from the interface of the target application are collected in a real-time manner and used to determine the traffic situation of the target application.
The request data from the interface of the target application may include but are not limited to the following eight forms:
The request data from the interface of the target application may include but are not limited to the following eight forms:
[0076] Get, make a request to a specific resource (request specific page information, and return the entity body);
[0077] Post, submit data to the specified resource for processing request (Submit form, Upload files), it may also lead to the establishment of new resources or the modification of existing resources;
[0078] Put, upload the latest content to the specified resource location (the content of the specified document is replaced by the data transmitted from the client to the server);
[0079] Head, with the server get request a consistent response, the response body will not return, get the original information contained in the small message header (and get the request is similar to, there is no specific content in the response returned, for getting headers);
[0080] Delete, Request server delete request-URL Resources marked *(ask the server to delete the page);
[0081] Trace, Display requests received by the server, for testing and diagnosis;
[0082] Opions, Returns the server's support for a specific resource HTML
Request method or web Server send * Test server function (Allow clients to view server performance); and
Request method or web Server send * Test server function (Allow clients to view server performance); and
[0083] Connec, proxy server in HTTP/1.1 protocols changes connection to a pipe.
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
[0084] S2 is about using the traffic prediction model to perform traffic prediction according to the request data, so as to obtain a prediction result related to a target cycle. The traffic prediction model is constructed on the basis of a neural network model.
[0085] Different from the prior art solutions, the embodiment of the present invention uses a traffic prediction model constructed on the basis of a network model to perform traffic prediction. Traffic prediction is the guide to generation or selection of traffic control strategies. Preciseness of traffic prediction results has direct relationship with the correctness of generation or selection of a traffic control strategies, thereby influencing performance of traffic control. The traffic prediction result may reflect a specific traffic value range, or may alternatively be a variation trend of traffic in the target cycle.
[0086] In one embodiment, prediction is performed through:
[0087] S21: entering the request data to said traffic prediction models of different prediction cycles for traffic prediction, so as to obtain the prediction results related to at least two different cycles; and
[0088] S22: analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0089] The traffic prediction models for different prediction cycles are plural traffic prediction models trained respectively using request data collected in different cycles as training samples, such as short cycle traffic prediction models, middle cycle traffic prediction models, and long cycle traffic prediction models. Correspondingly, the obtained prediction results include short cycle prediction results, middle cycle prediction results, and long cycle prediction results.
[0090] The prediction results related to different cycles obtained as described previously are then analyzed so as to obtain a prediction result related to a target cycle. This is mainly about analyzing the prediction results related to cycle of different lengths comprehensively, so Date Recue/Date Received 2022-03-08 as to ascertain the traffic situation that is related to a cycle as long as possible, and conforms, as much as possible, to the actual traffic situation of the target application in a cycle as long as possible. Specifically, the foregoing analysis may be as described in any one or more of the following embodiments.
[0091] In one embodiment, the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
[0092] using the prediction result of a short said cycle to correct the prediction result of a long said cycle; and
[0093] taking the corrected prediction result of the longer cycle as the prediction result of the target cycle.
[0094] The short cycles and the long cycles are identified by sorting the cycles by length. In the present embodiment, the step of using the prediction result of a short said cycle to correct the prediction result of a long said cycle comprises:
[0095] (1) correcting the prediction result related to the longest cycle according to the prediction results related to the other, short cycles; and
[0096] (2) grouping the prediction results of the cycles, and correcting the prediction results related to the relatively long cycles according to the prediction results related to the relatively short cycles in each group, and then obtaining the prediction result related to the target cycle according to the prediction results related to the relatively long cycles in each group (particularly, mean calculation, for example).
[0097] Specifically, the foregoing correction may be realized using the ratio between the prediction results of the short cycles and the long cycles to correct the prediction results of the long cycles, or by setting weights for different cycles. The embodiment of the present invention puts no limitation on how the correction is realized.
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
[0098] In one embodiment, the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
[0099] performing fit analysis on the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0100] The fit analysis mainly refers to determining a target function according to the discrete prediction results of different cycles by means of adjusting the factor. The target function is the one having the smallest difference with the discrete prediction results of the different cycles.
[0101] In one embodiment, the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
[0102] calculating a mean of the prediction results related to the different cycles; and
[0103] taking the mean of the prediction results related to the different cycles as the prediction result of the target cycle.
[0104] The foregoing mean calculation mainly refers to calculating the arithmetic mean or weighted mean of the prediction results related to different cycles.
[0105] It is to be noted that, in addition to analysis of the prediction results of different cycles as describe above, other approaches may be used to obtain the prediction result of the target cycle, including calculating the median of the prediction results of different cycles, and taking the median as the prediction result of the target cycle. The median-based approach is preferably to be used in cases where the number of cycles is large and the differences among different cycles are small.
[0106] In one embodiment, the traffic prediction model is constructed through:
[0107] S21 ': acquiring a sample training set for traffic prediction, which includes input samples Date Recue/Date Received 2022-03-08 and result samples;
[0108] S22': inputting the input sample to the initialized neural network model, so as to obtain a prediction result set;
[0109] S23': comparing the prediction result set to the result samples, so as to obtain prediction errors;
[0110] S24': based on prediction errors adjusting neural network model;
[0111] S25': using the adjusted neural network model to repeatedly make prediction based on the input samples; and
[0112] S26': comparing the prediction errors to standard error conditions, and when the prediction errors satisfy the standard error conditions, determining that the neural network model corresponding to the prediction errors is the traffic prediction model.
[0113] In S22', the initialized neural network model has an initialized weight. In S24', the synapse weights of the neural network model are corrected according to prediction errors using a backpropagation algorithm. Thereby, the traffic prediction model constructed as described in the embodiment of the present invention has the ability of adaptive learning, and can continuously adjusting synapse weights by comparing the prediction errors and standard errors, thereby eliminating the defects of the conventional methods about inability to predict sudden incidents.
[0114] S3: generating a traffic control strategy according to the prediction result of the target cycle.
[0115] The traffic control strategy may be generated according to the prediction result, or may alternatively be selected from pre-loaded ones that correspond to different prediction results.
[0116] S4 is about controlling the traffic of the target application in the target cycle according to the traffic control strategy.
[0117] As described above, with the relatively precise traffic prediction result, a targeted traffic control strategy may be generated, thereby achieving effective traffic control for the target Date Recue/Date Received 2022-03-08 application.
[0118] Corresponding to the method for traffic control as described above, the embodiment of the present invention further provides a traffic control apparatus, Referring to FIG. 2, it comprises the following components.
[0119] A data collecting unit 201 is for collecting request data of a target application.
[0120] The data collecting unit acquires request data of a target application through the interface of the target application. The request data may include various forms, and the embodiment of the present invention puts no limitation thereon.
[0121] A traffic-predicting unit 202 is for suing the traffic prediction model to perform traffic prediction according to request data, so as to obtain a traffic prediction result related to a target cycle, traffic prediction model based on neural network model.
[0122] In one embodiment, the traffic-predicting unit 202 specifically comprises: traffic prediction models related to different prediction cycles, and specifically comprises:
[0123] a prediction module, for entering the request data to said traffic prediction models of different prediction cycles for traffic prediction, so as to obtain the prediction results related to at least two different cycles; and
[0124] an analysis module, for analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0125] In one embodiment, the analysis module is specifically for:
[0126] using the prediction result of a short said cycle to correct the prediction result of a long said cycle; and
[0127] taking the corrected prediction result of the longer cycle as the prediction result of the target cycle.
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
[0128] The short cycles and the long cycles are identified by sorting the cycles by length. In the present embodiment, the step of using the prediction result of a short said cycle to correct the prediction result of a long said cycle comprises:
[0129] (1) correcting the prediction result related to the longest cycle according to the prediction results related to the other, short cycles; and
[0130] (2) grouping the prediction results of the cycles, and correcting the prediction results related to the relatively long cycles according to the prediction results related to the relatively short cycles in each group, and then obtaining the prediction result related to the target cycle according to the prediction results related to the relatively long cycles in each group (particularly, mean calculation, for example).
[0131] In one embodiment, the analysis module is specifically for:
[0132] performing fit analysis on the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
[0133] In one embodiment, the analysis module is specifically for:
[0134] calculating a mean of the prediction results related to the different cycles; and
[0135] taking the mean of the prediction results related to the different cycles as the prediction result of the target cycle.
[0136] The mean calculation mainly refers to calculating the arithmetic mean or weighted mean of the prediction results related to different cycles.
[0137] In one embodiment, the apparatus of the embodiment of the present invention further comprises a traffic prediction model training module, for:
[0138] acquiring a sample training set for traffic prediction, which includes input samples and result samples;
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
[0139] inputting the input sample to the initialized neural network model, so as to obtain a prediction result set;
[0140] comparing the prediction result set to the result samples, so as to obtain prediction errors;
[0141] adjusting the neural network model according to the prediction errors;
[0142] using the adjusted neural network model to repeatedly make prediction based on the input samples; and
[0143] comparing the prediction errors to standard error conditions, and when the prediction errors satisfy the standard error conditions, determining that the neural network model corresponding to the prediction errors is the traffic prediction model.
[0144] The step of adjusting the neural network model based on prediction errors comprises:
correcting the synapse weights of the neural network model according to prediction errors using a backpropagation algorithm.
correcting the synapse weights of the neural network model according to prediction errors using a backpropagation algorithm.
[0145] A strategy generating unit 203 is for generating traffic control strategy according to the traffic prediction result relate to the target cycle.
[0146] A traffic control unit 204 is for controlling the traffic of the target application in the target cycle according to the traffic control strategy.
[0147] Additionally, an embodiment of the present invention further provides an electronic device, which comprises:
[0148] one or more processors; and
[0149] a memory associated with the one or more processors, the memory storing program instructions, the program instructions, when read and executed by the one or more processors, executing the method for traffic control as provided in the embodiment of the present invention.
[0150] FIG. 3 illustratively depicts a structure of the computer system. It may specifically include Date Recue/Date Received 2022-03-08 a processor 310, a video display adapter 311, a disk driver 312, an I/O port 313, a network port 314, and a memory 320. The processor 310, the video display adapter 311, the disk driver 312, the I/O port 313, the network port 314, and the memory 320 may be communicatively connected through a communication bus 330.
[0151] Therein, the processor 310 may be implemented using a common Central Processing Unit (CPU), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits for executing relevant programs to realize the technical scheme provided by the present invention.
[0152] The memory 320 may be realized using a ROM (Read Only Memory), a RAM
(Random Access Memory), a static storage device, a dynamic storage device or any analog. The memory 320 may store for an operating system 321 that controls operation of the computer system 300, and a basic input/output system (BIOS) for controlling low-level operations of the computer system 300. In addition, it may further store a web browser 323, a data storage management system 324, and an icon and font processing system 325. The icon and font processing system 325 may be an application enabling operations of the foregoing various steps of the embodiments of the present invention. In any case, when the technical scheme provided by present invention is realized using software or firmware, the related program codes are stored in the memory 320 for the processor 310 to call and execute.
(Random Access Memory), a static storage device, a dynamic storage device or any analog. The memory 320 may store for an operating system 321 that controls operation of the computer system 300, and a basic input/output system (BIOS) for controlling low-level operations of the computer system 300. In addition, it may further store a web browser 323, a data storage management system 324, and an icon and font processing system 325. The icon and font processing system 325 may be an application enabling operations of the foregoing various steps of the embodiments of the present invention. In any case, when the technical scheme provided by present invention is realized using software or firmware, the related program codes are stored in the memory 320 for the processor 310 to call and execute.
[0153] The I/O port 313 is for connecting an I/O module for allowing data input and output. The input/output module may be built in the apparatus as a component (not shown), or may be set externally and connected to the apparatus so as to provides corresponding functions.
Therein, the input device may include a keyboard, a mouse, a touch panel, a microphone, various sensors or more. The output device may include a display, an amplifier, a vibrator, an indicator or more.
Date Recue/Date Received 2022-03-08
Therein, the input device may include a keyboard, a mouse, a touch panel, a microphone, various sensors or more. The output device may include a display, an amplifier, a vibrator, an indicator or more.
Date Recue/Date Received 2022-03-08
[0154] The network port 314 is for connecting a communication module (not shown) to allow communication between the disclosed apparatus and external devices. Therein, the communication module may enable communication either in a wired manner (such as through a USB, a network line, etc.) or in a wireless way (such as through a mobile network, WIFI, Bluetooth, etc.).
[0155] The bus 330 comprises a channel allowing information transmission among the components of the device (i.e., the processor 310, the video display adapter 311, the disk driver 312, the I/O port 313, the network port 314, and the memory 320).
[0156] Moreover, the computer system 300 may further obtain information about specific collection conditions from a virtual resources object collection condition information database for its condition determination or the like.
[0157] It is to be noted that while the apparatus in the depicted embodiment works merely by virtue of the processor 310, the video display adapter 311, the disk driver 312, the I/O
port 313, the network port 314, the memory 320, and the bus 330, in practical implementations, the apparatus may further include additional components required for its desired purposes. Moreover, people skilled in the art would appreciate that the disclosed apparatus may only include the minimal number of components for realizing the scheme of the present invention instead of having all these depicted components.
port 313, the network port 314, the memory 320, and the bus 330, in practical implementations, the apparatus may further include additional components required for its desired purposes. Moreover, people skilled in the art would appreciate that the disclosed apparatus may only include the minimal number of components for realizing the scheme of the present invention instead of having all these depicted components.
[0158] Especially, according to the embodiment of the present invention, the process described previously with reference to the flowchart may be realized as a computer software program. For example, an embodiment of the present invention comprises a computer program product, which includes a computer program carried by a computer-readable medium. The computer program includes program codes for executing the method illustrated in the flowchart. In such an embodiment, the computer program may be Date Recue/Date Received 2022-03-08 download and installed by a communication apparatus through a network, or may be installed from a memory, or may be installed from a ROM. When executed by a processor, the computer program executes the functions defined in the method of the embodiment of the present invention as described above.
[0159] It is to be noted that the computer-readable medium of the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. Morre particular examples of the computer-readable storage medium may include but are not limited to electric connection having one or more leads, a portable computer disk, a hard drive, a random-access memory (RAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EPROM or flash), an optical fiber, a compact disc read-only memory (CD-ROM), an optical memory, a magnetic memory, or any suitable combination thereof. In embodiments of the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program.
The program may be used by or with an instruction-executing system, an apparatus or a device.
In embodiments of the present invention, the computer-readable signal medium may comprise data signals that are propagated in the baseband or propagated as a part of carrier waves, and carry computer-readable program codes. The propagatable data signals may be in various forms including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium may alternatively be any computer-readable medium other than a computer-readable storage medium. The computer-readable signal medium may transmit, propagate, or send a program that is to be used by or with an instruction-executing system, an apparatus or a device. The program codes contained in the computer-readable medium may be transmitted through any suitable medium, including but not limited to: a power cord, an optical cable, RF (Radio Frequency), or any suitable combination thereof.
Date Recue/Date Received 2022-03-08
The program may be used by or with an instruction-executing system, an apparatus or a device.
In embodiments of the present invention, the computer-readable signal medium may comprise data signals that are propagated in the baseband or propagated as a part of carrier waves, and carry computer-readable program codes. The propagatable data signals may be in various forms including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium may alternatively be any computer-readable medium other than a computer-readable storage medium. The computer-readable signal medium may transmit, propagate, or send a program that is to be used by or with an instruction-executing system, an apparatus or a device. The program codes contained in the computer-readable medium may be transmitted through any suitable medium, including but not limited to: a power cord, an optical cable, RF (Radio Frequency), or any suitable combination thereof.
Date Recue/Date Received 2022-03-08
[0160] The computer-readable medium may be contained in the aforementioned server, or may alternatively exist separately instead of being installed in the server. The computer-readable medium carries one or more programs. When executed by the server, the one or more programs make the server: in response to its detection result indicating that the accessory mode of a terminal is not activated, acquire the frame rate of the application on the terminal; when the frame rate satisfies screen-off conditions, determine whether a user is acquiring the screen information of the terminal; and in response to its determination result indicating that the user is not acquiring the screen information of the terminal, control the screen to enter the screen-off mode.
[0161] The computer program codes for executing operations of the embodiment of the present invention may be written in one or more program design languages or a combination thereof. The program design languages may include object-orientated program design languages, such as Java, Smalltalk, C++, and may further include conventional procedural program design languages, such as "C" or the like. The program codes may be completely executed on a user computer, partially executed on a user computer, executed as an independent software package, partially executed on a user computer and partially executed on a remote computer, or completely executed on a remote computer or server. Where a remote computer is involved, the remote computer may be connected to the user computer through any kind of networks, such as a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (such as by using an Internet service provider through the Internet).
[0162] The embodiments of the present invention provide the following technical effects:
[0163] 1. In the embodiment of the present invention, the neural network model provides precise traffic prediction, so that traffic monitoring strategies specific to prediction results can be generated, thereby facilitating reasonable distribution of the traffic, and enabling warning Date Recue/Date Received 2022-03-08 of potential traffic bursts to be given according to prediction results;
[0164] 2. The technical scheme of the embodiment of the present invention performs analysis according to the traffic prediction results related to different prediction cycles so as to obtain a prediction result related to a target cycle, which is more precise than the prediction result obtained using the known linear prediction method; and
[0165] 3. The technical scheme of the embodiment of the present invention continuously trains and adjusts the traffic prediction model according to prediction errors, thereby eliminating the defects of the conventional methods about inability to predict sudden incidents.
[0166] The embodiments disclosed herein are described in a progressive sense, and therefore a part in one embodiment may having its details complemented by the description for its counter parts in other embodiments. Every embodiment is such described that it only emphasizes what differentiates it from the other embodiments. Particularly, for a system or an embodiment directed to a system, the subject matter may be described in a simplified way as more details may be learned from the embodiment about its relevant method. The system and system embodiment disclosed herein are merely illustrative, in which a unit described as a separated part may be or may be not physically separated, and a part shown as a unit may be or may be not a physical unit, meaning that it may be located at one site or alternatively be distributed across multiple units in a network. The purpose of an embodiment may be realized using all or some of the described/shown modules according to practical needs. People skilled in the art would understand and implement the present invention without paying creative efforts.
[0167] While some preferred embodiments of the present invention have been described, it is appreciated that, people skilled in the art in light of the basic creative concepts may change and modify theses embodiments in many ways. Thus, the appended claims are Date Recue/Date Received 2022-03-08 intended to be interpreted as being inclusive of theses preferred embodiments and all possible changes and modifications falling within the scope of the present invention.
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
Claims (10)
1. A method for traffic control, comprising:
collecting request data of a target application;
predicting a traffic according to the request data using a traffic prediction model, so as to obtain a prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model;
generating a traffic control strategy according to the prediction result related to the target cycle;
and controlling the traffic of the target application in the target cycle according to the traffic control strategy.
collecting request data of a target application;
predicting a traffic according to the request data using a traffic prediction model, so as to obtain a prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model;
generating a traffic control strategy according to the prediction result related to the target cycle;
and controlling the traffic of the target application in the target cycle according to the traffic control strategy.
2. The method of claim 1, wherein the step of predicting a traffic according to the request data using a traffic prediction model, so as to obtain a prediction result related to a target cycle comprises:
entering the request data to said traffic prediction models of different prediction cycles for traffic prediction, so as to obtain the prediction results related to at least two different cycles; and analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
entering the request data to said traffic prediction models of different prediction cycles for traffic prediction, so as to obtain the prediction results related to at least two different cycles; and analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
3. The method of claim 2, wherein the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
using the prediction result of a short said cycle to correct the prediction result of a long said cycle;
and taking the corrected prediction result of the long cycle as the prediction result of the target cycle.
using the prediction result of a short said cycle to correct the prediction result of a long said cycle;
and taking the corrected prediction result of the long cycle as the prediction result of the target cycle.
4. The method of claim 2, wherein the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
performing fit analysis on the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
Date Recue/Date Received 2022-03-08
performing fit analysis on the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle.
Date Recue/Date Received 2022-03-08
5. The method of claim 2, wherein the step of analyzing the prediction results related to the different cycles, so as to obtain the prediction result related to the target cycle comprises:
calculating a mean of the prediction results related to the different cycles;
and taking the mean of the prediction results related to the different cycles as the prediction result of the target cycle.
calculating a mean of the prediction results related to the different cycles;
and taking the mean of the prediction results related to the different cycles as the prediction result of the target cycle.
6. The method of any of claims 1 through 5, wherein the traffic prediction model is constructed through:
acquiring a sample training set for traffic prediction, which includes input samples and result samples;
inputting the input sample to the initialized neural network model, so as to obtain a prediction result set;
comparing the prediction result set to the result samples, so as to obtain prediction errors;
adjusting the neural network model according to the prediction errors;
using the adjusted neural network model to repeatedly make prediction based on the input samples; and comparing the prediction errors to standard error conditions, and when the prediction errors satisfy the standard error conditions, determining that the neural network model corresponding to the prediction errors is the traffic prediction model.
acquiring a sample training set for traffic prediction, which includes input samples and result samples;
inputting the input sample to the initialized neural network model, so as to obtain a prediction result set;
comparing the prediction result set to the result samples, so as to obtain prediction errors;
adjusting the neural network model according to the prediction errors;
using the adjusted neural network model to repeatedly make prediction based on the input samples; and comparing the prediction errors to standard error conditions, and when the prediction errors satisfy the standard error conditions, determining that the neural network model corresponding to the prediction errors is the traffic prediction model.
7. The method of claim 6, wherein the step of adjusting the neural network model according to the prediction errors comprises:
correcting a weight of every synapse in the neural network model according to the prediction errors using a backpropagation algorithm.
correcting a weight of every synapse in the neural network model according to the prediction errors using a backpropagation algorithm.
8. An apparatus for traffic control, comprising:
a data-collecting unit, for collecting request data of a target application;
a traffic-predicting unit, for predicting a traffic according to the request data using a traffic prediction model, so as to obtain a traffic prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model;
a strategy-generating unit, for generating traffic control strategy according to the traffic prediction result related to the target cycle; and a traffic-controlling unit, for controlling the traffic of the target application in the target cycle Date Recue/Date Received 2022-03-08 according to the traffic control strategy.
a data-collecting unit, for collecting request data of a target application;
a traffic-predicting unit, for predicting a traffic according to the request data using a traffic prediction model, so as to obtain a traffic prediction result related to a target cycle, in which the traffic prediction model is constructed based on a neural network model;
a strategy-generating unit, for generating traffic control strategy according to the traffic prediction result related to the target cycle; and a traffic-controlling unit, for controlling the traffic of the target application in the target cycle Date Recue/Date Received 2022-03-08 according to the traffic control strategy.
9. An electronic device, comprising:
one or more processors; and a memory associated with the one or more processors, the memory storing program instructions, the program instructions, when read and executed by the one or more processors, executing the method of any of claims 1 to 7.
one or more processors; and a memory associated with the one or more processors, the memory storing program instructions, the program instructions, when read and executed by the one or more processors, executing the method of any of claims 1 to 7.
10. A computer-readable medium, storing therein a computer program, wherein the program, when executed by a processor, implements the method of any of claims 1 to 7.
Date Recue/Date Received 2022-03-08
Date Recue/Date Received 2022-03-08
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110249289.0 | 2021-03-08 | ||
CN202110249289.0A CN113079033B (en) | 2021-03-08 | 2021-03-08 | Flow control method and device, electronic equipment and computer readable medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CA3151273A1 true CA3151273A1 (en) | 2022-09-08 |
Family
ID=76612102
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA3151273A Pending CA3151273A1 (en) | 2021-03-08 | 2022-03-08 | Method, apparatus, electronic device, and computer-readable medium for traffic control |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN113079033B (en) |
CA (1) | CA3151273A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113726550A (en) * | 2021-07-21 | 2021-11-30 | 广东电网有限责任公司广州供电局 | Flow prediction method and device, computer equipment and readable storage medium |
CN113835449B (en) * | 2021-11-29 | 2022-03-18 | 常州高凯电子有限公司 | Control method for quickly adjusting valve of flow controller based on pressure fluctuation |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109214584B (en) * | 2018-09-21 | 2022-02-08 | 北京百度网讯科技有限公司 | Method and device for predicting passenger flow |
CN109120463B (en) * | 2018-10-15 | 2022-01-07 | 新华三大数据技术有限公司 | Flow prediction method and device |
CN111222663B (en) * | 2018-11-23 | 2023-12-08 | 北京京东尚科信息技术有限公司 | Data processing method and system, computer system and computer readable medium |
CN111294227A (en) * | 2018-12-10 | 2020-06-16 | 中国移动通信集团四川有限公司 | Method, apparatus, device and medium for neural network-based traffic prediction |
CN112039711B (en) * | 2020-09-08 | 2023-03-24 | 中国联合网络通信集团有限公司 | Flow prediction method and equipment |
-
2021
- 2021-03-08 CN CN202110249289.0A patent/CN113079033B/en active Active
-
2022
- 2022-03-08 CA CA3151273A patent/CA3151273A1/en active Pending
Also Published As
Publication number | Publication date |
---|---|
CN113079033A (en) | 2021-07-06 |
CN113079033B (en) | 2022-09-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA3151273A1 (en) | Method, apparatus, electronic device, and computer-readable medium for traffic control | |
US9584364B2 (en) | Reporting performance capabilities of a computer resource service | |
US8627147B2 (en) | Method and computer program product for system tuning based on performance measurements and historical problem data and system thereof | |
US20070022327A1 (en) | Computer-readable recording medium recording system performance monitoring program, and system performance monitoring method and apparatus | |
CN109726763B (en) | Information asset identification method, device, equipment and medium | |
JP6193393B2 (en) | Power optimization for distributed computing systems | |
CN111177617A (en) | Web direct operation and maintenance method and device based on operation and maintenance management system and electronic equipment | |
CN111404713B (en) | Network resource adjusting method, device and storage medium | |
CN109920192A (en) | Fire alarm method, system and computer readable storage medium | |
US10956229B2 (en) | Managing resource sharing and task bidding on the internet of things (IoT) | |
CN108897673B (en) | System capacity evaluation method and device | |
CN114710499A (en) | Load balancing method, device and medium for edge computing gateway based on computational power routing | |
CN116362359A (en) | User satisfaction prediction method, device, equipment and medium based on AI big data | |
CN104796929B (en) | Network debugging method and device | |
CN116521344B (en) | AI algorithm scheduling method and system based on resource bus | |
WO2020252880A1 (en) | Reverse turing verification method and apparatus, storage medium, and electronic device | |
CN116528335A (en) | Satellite Internet of things access method, device, equipment and medium based on information value | |
CN115952098A (en) | Performance test tuning scheme recommendation method and system | |
CN110633182B (en) | System, method and device for monitoring server stability | |
CN115022206B (en) | Network stability determination method and device, computer equipment and readable storage medium | |
CN114978794B (en) | Network access method, device, storage medium and electronic equipment | |
KR102255252B1 (en) | Method and server for deciding summary value from big raw data | |
CN113419879B (en) | Message processing method, device, equipment and storage medium | |
Truong et al. | Design and Implementation of a Portable Audio Signal Spectrum Analyzer for Noise Management | |
CN117785625A (en) | Method, device, equipment and storage medium for predicting server performance |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request |
Effective date: 20220916 |
|
EEER | Examination request |
Effective date: 20220916 |
|
EEER | Examination request |
Effective date: 20220916 |