CN108924057B - Port flow intelligent control system of on-cloud system - Google Patents

Port flow intelligent control system of on-cloud system Download PDF

Info

Publication number
CN108924057B
CN108924057B CN201810872051.1A CN201810872051A CN108924057B CN 108924057 B CN108924057 B CN 108924057B CN 201810872051 A CN201810872051 A CN 201810872051A CN 108924057 B CN108924057 B CN 108924057B
Authority
CN
China
Prior art keywords
port
flow
traffic
information
cloud system
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.)
Active
Application number
CN201810872051.1A
Other languages
Chinese (zh)
Other versions
CN108924057A (en
Inventor
单健锋
江鹏辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Koubei Network Technology Co Ltd
Original Assignee
Zhejiang Koubei Network Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang Koubei Network Technology Co Ltd filed Critical Zhejiang Koubei Network Technology Co Ltd
Priority to CN201810872051.1A priority Critical patent/CN108924057B/en
Publication of CN108924057A publication Critical patent/CN108924057A/en
Application granted granted Critical
Publication of CN108924057B publication Critical patent/CN108924057B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/70—Admission control; Resource allocation
    • H04L47/76—Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate

Abstract

The application provides a port flow intelligent control system of a cloud system, a flow characteristic prediction method and device, a flow data processing method and device and a port flow control method and device applied to the system. The system port flow intelligent control system comprises: the offline data center stores flow data provided by the cloud system; the port computing center predicts the port flow characteristics of the cloud system according to the flow data and sends port size adjustment or flow indication information to the cloud system, the cloud system adjusts the ports according to the indication information and sends the adjusted port information to the port forwarding center, and the port forwarding center sends the flow data of the client to a proper port of the cloud system according to the adjusted port information. The port flow intelligent control system can predict the flow characteristics of each port in advance, so that each port is controlled to perform automatic adjustment and automatic flow limitation according to different flows, and automatic flow distribution is realized without manual intervention.

Description

Port flow intelligent control system of on-cloud system
Technical Field
The application relates to the field of port flow control of a system, in particular to an intelligent port flow control system of a cloud-based system. The application further provides a flow characteristic prediction method and device, a flow data processing method and device, and a port flow control method and device applied to the port flow control system of the on-cloud system.
Background
With the development of the internet, various cloud service systems based on the internet are in a variety, and the flow data accessing the cloud service systems are more and more, especially for the cloud service systems in the catering industry, the flow data accessing the cloud service systems during a meal spot or an activity is very concentrated. Under the background, the size of a cloud service system port is artificially set according to the change condition of flow data, and the flow data access requirement of the catering industry cannot be met. Therefore, in the cloud service system of the catering industry, one core problem is that: how to ensure that the port of the cloud service system is not punctured by the centrally accessed traffic data.
In order to solve the above problem, the patent document of application No. 201610143420.4 provides a network flow control method, which determines traffic pipes for transmitting received traffic data according to the type of the traffic data, each of which can allocate bandwidth weight according to the traffic type, reasonably schedule the traffic data and send the traffic data to a port. Although the scheme reasonably allocates the bandwidth through the flow data classification, and sends the flow data to the designated port, the allocation of the flow data which are accessed in a centralized manner is realized to a certain extent, so that the port of the service system is not broken down, the catering industry can still not realize the effective protection of the port of the cloud service system, and the running stability of the cloud service system can not be ensured.
Disclosure of Invention
The application provides a port flow intelligent control system of a cloud system, so that the running stability of a cloud service system is ensured. The application further provides a flow characteristic prediction method and device, a flow data processing method and device, and a port flow control method and device applied to the port flow control system of the on-cloud system.
The application provides a port flow intelligence control system of system on cloud, includes: the system comprises an offline data center, a port calculation center, a cloud system and a port forwarding center;
the off-line data center is used for storing the flow data provided by the cloud system;
the port computing center is used for acquiring flow data stored by the off-line data center, predicting port flow characteristics of the on-cloud system according to the flow data, and sending an indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system to the cloud system according to the port flow characteristics;
the cloud system is used for obtaining adjusted port bandwidth information and/or port flow information according to the indication message sent by the port calculation center, and sending the adjusted port bandwidth information and/or the adjusted port flow information to the port forwarding center;
and the port forwarding center is used for sending the flow data obtained from the client to the port of the cloud system according to the adjusted port bandwidth information and/or the adjusted port flow information sent by the cloud system.
Optionally, the offline data center is further configured to perform traffic data cleaning on the traffic data provided by the cloud system, and provide the cleaned traffic data to the port computing center;
the port computing center is specifically configured to obtain the cleaned flow data provided by the offline data center, and predict port flow characteristics of the system on the cloud according to the cleaned flow data.
Optionally, the offline data center is specifically configured to delete negative samples outside a preset positive-negative sample ratio from the flow data provided by the cloud system according to the preset positive-negative sample ratio of the flow data and the number of positive samples in the flow data provided by the cloud system, so as to obtain cleaned flow data.
Optionally, the offline data center is further configured to perform discretization processing on the traffic data provided by the cloud system, obtain discretized traffic data, and provide the discretized traffic data to the port computation center;
the port computing center is specifically configured to obtain discretized flow data provided by the offline data center, and predict port flow characteristics of the system on the cloud according to the discretized flow data.
Optionally, the offline data center is specifically configured to store, by using a sparse matrix, the traffic data provided by the cloud system, obtain the traffic data after sample size expansion, and provide the traffic data after sample size expansion to the port computation center, where the number of the traffic data after sample size expansion is greater than the number of the traffic data provided by the cloud system;
the port computing center is specifically configured to obtain the flow data provided by the offline data center after the sample size is expanded, and predict a port flow characteristic of the system on the cloud according to the flow data after the sample size is expanded.
Optionally, the flexible port computing center is specifically configured to extract traffic data time information and traffic data peak information from traffic data stored in the offline data center, and predict a port traffic characteristic of the cloud system according to the traffic data time information and the traffic data peak information by using a bayesian network model, where the bayesian network model is a network model used to predict the port traffic characteristic of the cloud system according to the traffic data time information and the traffic data peak information.
Optionally, the flexible port calculation center is further configured to perform unified clustering on the flow data stored in the offline data center according to a time dimension and a peak dimension to obtain flow data time information and flow data peak information after the unified clustering, perform normalization processing on the flow data time information and the flow data peak information after the unified clustering to obtain normalized flow data time information and flow data peak information, and train the bayesian network model by using the normalized flow data time information and the flow data peak information.
Optionally, the port traffic characteristics include at least one of the following information:
a peak flow value of a port of a system on the cloud within a period of time;
traffic change information of a port of a system on the cloud within a period of time;
total traffic information for ports of the on-cloud system over a period of time;
average traffic information per unit time over a time range for a port of a system on the cloud.
Optionally, the port computing center is specifically configured to determine, according to the port traffic characteristics, a maximum port bandwidth and/or a maximum port traffic threshold of a port of the cloud system within a period of time;
the method comprises the step of sending an indication message for adjusting port bandwidth and/or port traffic of the on-cloud system to the cloud system, wherein the indication message comprises port maximum bandwidth and/or port maximum traffic threshold of a port of the on-cloud system in a period of time.
Optionally, the cloud system is specifically configured to obtain adjusted port bandwidth information and/or port traffic information according to an indication message sent by the port computing center and used for adjusting the port bandwidth and/or port traffic of the system on the cloud, and send the adjusted port bandwidth information and/or the adjusted port traffic information to the port forwarding center.
Optionally, the port of the cloud system is configured to send alarm information to the cloud system when the traffic data of the port reaches or exceeds the bearable peak value;
the cloud system can also be used for automatically adjusting the port bandwidth and/or the port flow according to alarm information sent by a system port on the cloud, and sending adjusted port bandwidth information and/or adjusted port flow information to the port forwarding center.
Correspondingly, the application also provides a flow characteristic prediction method, which comprises the following steps:
acquiring flow data;
predicting port flow characteristics of the system on the cloud according to the flow data;
and sending an indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system to the cloud system according to the port flow characteristics of the on-cloud system.
Optionally, the obtaining of the traffic data and predicting, according to the traffic data, a port traffic characteristic of the cloud system specifically include:
acquiring cleaned flow data;
and predicting the port flow characteristics of the cloud system according to the cleaned flow data.
Optionally, the obtaining of the traffic data and predicting, according to the traffic data, a port traffic characteristic of the cloud system specifically include:
obtaining discretized flow data;
and predicting port flow characteristics of the system on the cloud according to the discretized flow data.
Optionally, the obtaining of the traffic data and predicting, according to the traffic data, a port traffic characteristic of the cloud system specifically include:
obtaining flow data after sample size expansion;
and predicting the port flow characteristics of the system on the cloud according to the flow data after the sample size is expanded.
Optionally, the obtaining of the traffic data and predicting, according to the traffic data, a port traffic characteristic of the cloud system specifically include:
extracting flow data time information and flow data peak value information from flow data stored in the offline data center;
predicting port flow characteristics of the cloud system by using a Bayesian network model according to the flow data time information and the flow data peak value information;
the Bayesian network model is a network model used for predicting port traffic characteristics of the on-cloud system according to traffic data time information and traffic data peak value information.
Optionally, the port traffic characteristics of the system on the cloud include at least one of the following information:
a peak flow value of a port of the on-cloud system within a period of time;
traffic change information of a port of a system on the cloud within a period of time;
total traffic information for ports of the on-cloud system over a period of time;
average traffic information per unit time over a time range for a port of a system on the cloud.
Optionally, the sending, to the cloud system, an instruction message for adjusting a port bandwidth and/or a port traffic of the cloud system according to the port traffic characteristic of the cloud system specifically includes:
determining the maximum port bandwidth and/or the maximum port traffic threshold of a port of the on-cloud system within a period of time according to the port traffic characteristics;
and sending an indication message for adjusting the port bandwidth and/or the port traffic of the on-cloud system to the cloud system, wherein the indication message comprises the port maximum bandwidth and/or the port maximum traffic threshold of the port of the on-cloud system in a period of time.
Correspondingly, the application also provides a flow data processing method, which comprises the following steps:
storing flow data provided by the cloud system;
and cleaning the stored flow data, and providing the cleaned flow data for a port calculation center.
Optionally, the cleaning of the stored flow data and the provision of the cleaned flow data to the port calculation center specifically include:
and deleting negative samples outside the proportion from the flow data provided by the cloud system according to the preset proportion of positive and negative samples of the flow data and the number of positive samples in the flow data provided by the cloud system to obtain cleaned flow data, and providing the cleaned flow data to a port calculation center.
Optionally, the storing the flow data provided by the cloud system, performing flow data cleaning on the stored flow data, and providing the cleaned flow data to the port computing center, further includes:
and carrying out discretization processing on the flow data provided by the cloud system to obtain discretized flow data, and providing the discretized flow data to a port calculation center.
Optionally, the storing the flow data provided by the cloud system, performing flow data cleaning on the stored flow data, and providing the cleaned flow data to the port computing center specifically includes:
storing the flow data provided by the cloud system by using a sparse matrix;
obtaining the flow data after sample size expansion, and providing the flow data after sample size expansion for a port calculation center, wherein the quantity of the flow data after sample size expansion is more than that of the flow data provided by the cloud system.
Correspondingly, the application also provides a port control method of the cloud system, which comprises the following steps:
acquiring indication information of port adjustment of a system on the cloud;
according to the indication information of the port adjustment of the on-cloud system, adjusting the bandwidth information and/or the port flow indication information of the port of the on-cloud system to obtain the adjusted port bandwidth information and/or the adjusted port flow indication information of the on-cloud system;
and sending the adjusted port bandwidth information and/or the adjusted port traffic information of the on-cloud system to the port forwarding center according to the port bandwidth information and/or the port traffic information of the on-cloud system.
Optionally, the adjusting, according to the indication information of the port adjustment of the cloud system, the bandwidth information and/or the port traffic information of the port of the cloud system to obtain the adjusted port bandwidth information and/or port traffic information of the cloud system specifically includes:
and obtaining adjusted port bandwidth information and/or port traffic information according to the indication message for adjusting the port bandwidth and/or the port traffic of the on-cloud system sent by the port computing center, and sending the adjusted port bandwidth information and/or the adjusted port traffic information to the port forwarding center.
Optionally, the obtained indication message for adjusting the port bandwidth and/or the port traffic of the system on the cloud specifically includes: a port maximum bandwidth and/or a port maximum traffic threshold for a port of the system on the cloud over a period of time.
Optionally, the method further includes:
receiving alarm information sent by a port of a cloud system;
automatically adjusting the port bandwidth and/or the port flow according to alarm information sent by a system port on the cloud;
and sending the adjusted port bandwidth information and/or the adjusted port flow information to the port forwarding center.
Correspondingly, the present application also provides a flow characteristic prediction apparatus, including:
an acquisition unit configured to acquire traffic data;
the prediction unit is used for predicting port flow characteristics of the system on the cloud according to the flow data;
and the sending unit is used for sending an indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system to the cloud system according to the port flow characteristics of the on-cloud system.
Correspondingly, the present application also provides a traffic data processing apparatus, including:
the storage unit is used for storing the flow data provided by the cloud system;
the cleaning unit is used for cleaning the stored flow data;
and the sending unit is used for providing the cleaned flow data to the port calculation center.
Correspondingly, the present application further provides a port control device of a cloud system, including:
the acquisition unit is used for acquiring indication information of port adjustment of the system on the cloud;
the adjusting unit is used for adjusting the bandwidth information and/or the port flow information of the port of the on-cloud system according to the indication information of the port adjustment of the on-cloud system to obtain the adjusted port bandwidth information and/or the adjusted port flow information of the on-cloud system;
and the sending unit is used for sending the adjusted port bandwidth information and/or the adjusted port traffic information of the on-cloud system to the port forwarding center according to the port bandwidth information and/or the port traffic information of the on-cloud system.
Compared with the prior art, the method has the following advantages:
the application provides a port flow intelligent control system of a cloud system, in particular to a system for intelligently controlling flexible change of ports of the cloud system according to different flows. The port flow intelligent control system comprises: the offline data center stores flow data provided by the cloud system; the port computing center obtains the flow data, predicts the port flow characteristics of the cloud system according to the flow data, and sends an indication message for adjusting the port bandwidth or the port flow of the cloud system to the cloud system; the cloud system obtains adjusted port bandwidth information or port flow information according to the indication message, and sends the adjusted port bandwidth information or the adjusted port flow information to the port forwarding center; and the port forwarding center sends the flow data obtained from the client to the port of the cloud system according to the adjusted port bandwidth information or the adjusted port flow information sent by the cloud system. The port flow intelligent control system can predict the flow characteristics of each port in advance according to the flow data of a current period of time, thereby controlling each port to carry out automatic adjustment and automatic flow limitation according to the difference of flow and realizing automatic flow distribution without manual intervention.
The application also provides a flow characteristic prediction method, and particularly relates to a method for predicting flow characteristics through an intelligent algorithm. The flow characteristic prediction method comprises the following steps: acquiring flow data, predicting port flow characteristics of the on-cloud system by using an intelligent algorithm according to the flow data, and predicting current port flow data according to the predicted port flow characteristics of the on-cloud system; and sending an indication message for adjusting the port bandwidth or the port flow of the system on the cloud to the cloud system according to the predicted current port flow data. The flow characteristic prediction method can predict the current flow data characteristics of the port of the cloud system through an intelligent algorithm according to the flow data information in a current period of time, and sends an adjusting instruction according to the flow data characteristics, so that the bandwidth and the flow of the port can be adjusted in advance.
Drawings
Fig. 1 is a schematic diagram of an embodiment of a port traffic intelligent control system of a cloud system according to the present application;
FIG. 2 is a flow chart of an embodiment of a flow characteristic prediction method of the present application;
FIG. 3 is a flow chart of an embodiment of a traffic data processing method according to the present application;
fig. 4 is a flowchart illustrating an embodiment of a port control method of the cloud system according to the present application;
fig. 5 is a scene diagram of an embodiment of a port traffic intelligent control system of the cloud system according to the present application;
fig. 6 is a complete structural diagram of a port traffic intelligent control system of the on-cloud system according to the present application;
FIG. 7 is a schematic view of an embodiment of a flow characteristic prediction apparatus according to the present application;
FIG. 8 is a schematic diagram of an embodiment of a traffic data processing apparatus according to the present application;
fig. 9 is a schematic diagram of an embodiment of a port control device of the on-cloud system according to the present application.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.
As described above, in the existing traffic data control method, a service pipe for transmitting traffic data is often determined according to the type of received traffic data, a bandwidth weight of each pipe is allocated according to the service type, and the traffic data is reasonably scheduled and transmitted to a designated port through the pipe. Although the scheme reasonably distributes the bandwidth and sends the flow data to the designated port through flow data classification, the reasonable distribution of the flow data which are accessed in a centralized manner is realized to a certain extent, so that the port of the service system is not broken down, the effective protection of the port of the cloud service system can not be realized particularly in the catering industry, and the running stability of the cloud service system can not be ensured.
The application provides a port flow intelligent control system of system on cloud, its core thought is as the basis of system port flow control on whole cloud with the off-line data center, with the core of system port flow intelligent control on whole cloud with port calculation center. The port flow intelligent control system can predict the flow characteristics of each port in advance according to the flow data of a current period of time, thereby controlling each port to carry out automatic adjustment and automatic flow limitation according to the difference of flow and realizing automatic flow distribution without manual intervention.
The following describes an embodiment of the port traffic intelligent control system based on the on-cloud system of the present application in detail. In addition, in the following description, detailed descriptions will be made for functions of the offline data center, the port computation center, the cloud system, and the port forwarding center in the present system, respectively. Please refer to fig. 1, which is a schematic diagram of an embodiment of a port traffic intelligent control system of a cloud system according to the present application.
In this embodiment, the port traffic intelligent control system of the cloud system includes the following components:
and the offline data center 101 is used for storing the flow data provided by the cloud system.
In this embodiment, when the port computing center directly applies the intelligent algorithm to the traffic data stored in the offline data center, an ideal result cannot be obtained in most cases, because the actually stored traffic data is often unevenly distributed and has a large amount of dirty data, the client traffic data provided by the offline data center cannot be directly used by the flexible computing center. Therefore, before sending data to the flexible computing center, the off-line data center needs to perform processing operations such as data cleaning, positive and negative sample proportioning, continuous data discretization and the like on the stored flow data, so that the flow data stored in the off-line data center can be used.
In this embodiment, if the off-line data center is used to process the stored flow data so that the flow data can be used by the port calculation center, the dirty data needs to be removed by data cleaning, then the negative samples outside the preset ratio value are deleted from the flow data provided by the cloud system according to the preset ratio of the positive and negative samples of the flow data and the number of the positive samples in the flow data provided by the cloud system to obtain the cleaned flow data, the cleaned flow data is further discretized by continuous data discretization operation to obtain discretized flow data, the discretized flow data is subjected to sample amplification by sparse matrix storage, the quality of the flow data is improved, the flow data with the sample size expanded is obtained, and the flow data with the sample size expanded is sent to the port calculation center. At this time, the traffic data stored in the offline data center can be applied to an intelligent algorithm for operation, so that the port traffic characteristics of the system on the cloud can be predicted.
In this embodiment, the offline data center performs traffic data cleaning on the stored traffic data, that is, the traffic data center performs real-time monitoring on the stored traffic data in the process of storing the traffic data forwarded by the cloud system, finds dirty data in off-peak periods in time and marks the dirty data, and performs identification cleaning on the marked dirty data by using a cleaning unit of the offline data center to retain the dirty data in peak periods. It should be noted that the dirty data in the general sense means data in the system that is not in a given range or meaningless to actual services, and particularly in this embodiment, the dirty data means data meaningless to predicting flow characteristics by using an intelligent algorithm.
Because continuous data cannot be directly analyzed and processed through an intelligent algorithm, stored continuous flow data needs to be discretized, namely, a group of continuous flow data values are divided into a plurality of specific discretized sets according to a certain rule. After the continuous data are discretized, the discretized flow data are expanded through sparse matrix storage, and the discretized flow data after sample volume expansion are provided for the port calculation center. Of course, the method for sample amplification is not limited to using sparse matrix storage for sample amplification, and methods such as convex optimization or antagonistic neural network may be used for sample amplification.
The port computing center 102 is configured to obtain traffic data stored in the offline data center, predict a port traffic characteristic of the cloud system according to the traffic data, and send an indication message for adjusting a port bandwidth and/or a port traffic of the cloud system to the cloud system according to the port traffic characteristic.
In this embodiment, the port computing center is a core of a port flow control system of the whole cloud system, and can perform intelligent prediction according to flow data provided by offline data, where the specific prediction process includes: extracting time information, flow data value information and flow data peak value information from flow data provided by the off-line data center, extracting characteristics of the extracted flow data information by adopting an SVM algorithm, carrying out clustering operation on the extracted characteristic data by utilizing a Kmeans algorithm, so that data with similar time characteristics and flow data characteristics are counted together to form a characteristic pool, establishing a Bayesian network model according to the flow characteristic information and the time characteristic information in the characteristic pool, and predicting the current flow data characteristics of a port of the on-cloud system in advance. It should be noted that the bayesian network model is a network model for predicting port traffic characteristics of the cloud system according to traffic data time information and traffic data peak information, and the clustering operation is not limited to clustering operation using a Kmeans algorithm, and may also be clustering operation using a Kmedoids algorithm or a Clarans algorithm.
The port computing center obtains current flow data characteristics of the port of the on-cloud system through the method, determines the maximum port bandwidth or the maximum port flow threshold of the port of the on-cloud system in a current period of time according to the current flow data characteristics of the port of the on-cloud system, and sends an indication message for adjusting the port bandwidth, the port flow limit value and the bearable peak value of the on-cloud system to the cloud system according to the maximum port bandwidth or the maximum port flow threshold in the current period of time, wherein the indication message specifically refers to information for adjusting the port bandwidth, the port flow limit value and the specific adjustment quantity of the bearable peak value of the on-cloud system. The port traffic data characteristics comprise at least one of traffic peak information, traffic change information, total traffic information and average traffic information per unit time of a port of the on-cloud system in a period of time.
The cloud system 103 is configured to obtain adjusted port bandwidth information and/or port traffic information according to the indication message sent by the port computing center, and send the adjusted port bandwidth information and/or the adjusted port traffic information to the port forwarding center.
In this embodiment, after receiving indication information for adjusting a port bandwidth or port traffic sent by a port computing center, the adjusting, by the cloud system, the port bandwidth or the port traffic of the system on the cloud according to the indication information specifically includes: and after the adjusted port information is obtained, the adjusted port number of the specific port, the current bandwidth and the current flow limit value of the port, the bearable peak value and other information are sent to a port forwarding center.
In addition, when the flow data of the client side intensively accesses to or exceeds the bearable peak value of the cloud system port, the cloud system port can also detect abnormal information of overhigh flow data through the IP bound by the background and send alarm information to the cloud system, the cloud system adjusts the port bandwidth or the port flow according to the alarm information sent by the port, and sends the adjusted port bandwidth information or the adjusted port flow information to the port forwarding center. The cloud system specifically means that the bandwidth or the port flow of each port is adjusted according to the alarm information sent by the port, so that the port of the cloud system can flexibly change according to different flows, and the port of the cloud system is prevented from being broken down due to overlarge short-time flow.
And the port forwarding center 104 is configured to send traffic data obtained from the client to a port of the cloud system according to the adjusted port bandwidth information and/or the adjusted port traffic information sent by the cloud system.
In this embodiment, the port forwarding center may directly receive the accessed client traffic data, and before forwarding the client traffic data to each port of the cloud system, it needs to receive the adjusted port information sent by the cloud system first, and according to the received port information, forward the traffic information obtained from the client to a suitable port of the cloud system. It should be noted that the port information includes information such as a port number of a specific port, a bandwidth adjusted by the port, a size of a current limit, and a bearable peak value.
The system is preferably applied to port flow control of a cloud system in the catering industry, and please refer to fig. 5, which is a scene diagram of an embodiment of an intelligent port flow control system of a cloud system in the embodiment;
generally, related flow data in the catering industry have the characteristics of high real-time performance, concentrated flow data and short peak value. Therefore, the real-time performance of port computation central traffic characteristic prediction is very important.
Firstly, client flow data initiates a request to a cloud system of a merchant through a classical network, an offline data center stores flow data flooded by a client in real time and processes the flow data, according to the characteristic that the flow data in a catering industry is concentrated and the peak value is short, dirty data in off-peak time periods are removed, positive and negative sample proportion is carried out, a port calculation center reads client access flow data of the merchant within a period of time after being processed by the offline data center in real time, real-time operation and analysis are carried out by utilizing a machine learning algorithm, the flow data characteristics to be accessed by a cloud system port of the current merchant are predicted in advance, instruction information for adjusting the cloud system port of the merchant is sent to an industry cloud system, the industry cloud system adjusts the port bandwidth size, the flow limiting value and the bearable peak value according to the adjustment instruction information, and sends the adjusted port number, the port number and the port number of a specific port to a port forwarding center, The port forwarding center automatically addresses according to received port traffic characteristic information, forwards traffic data accessed by a current client to a proper port, the traffic data reaches a cloud system of a merchant through the port, and the cloud system of the merchant returns a request result to the client after receiving a client access request.
In the above embodiment, a port traffic intelligent control system of a cloud system is provided, please refer to fig. 6, which is a complete structure diagram of the port traffic intelligent control system of the cloud system according to the present application. The port flow intelligent control system stores flow data provided by the cloud system through an offline data center; the port computing center obtains the flow data, predicts port flow characteristics of the cloud system according to the flow data, and sends an indication message for adjusting the port bandwidth and/or the port flow of the cloud system to the cloud system; the cloud system obtains the adjusted port bandwidth information and/or the adjusted port flow information according to the indication message, and sends the adjusted port bandwidth information and/or the adjusted port flow information to the port forwarding center; and the port forwarding center sends the flow data obtained from the client to the port of the cloud system according to the adjusted port bandwidth information and/or the adjusted port flow information sent by the cloud system. The port flow intelligent control system can predict the flow characteristics of each port in advance according to the flow data of a current period of time, thereby controlling each port to carry out automatic adjustment and automatic flow limitation according to the difference of flow and realizing automatic flow distribution without manual intervention.
Corresponding to the port flow intelligent control system of the cloud system, the present application also provides a flow characteristic prediction method, which may be applied to the port flow intelligent control system of the cloud system, please refer to fig. 2, which is a flowchart of an embodiment of the flow characteristic prediction method of the present application. Because the embodiment of the method is similar to the embodiment of the system, the description is relatively simple, and the related points are only explained with reference to the embodiment of the system, and the embodiment of the method described below is only schematic.
In this embodiment, a method for predicting flow characteristics includes the following steps:
step S201: and acquiring flow data.
The flow data obtained by the port calculation center in this step may be at least one of flow data after data cleaning, flow data after positive and negative sample matching, and discretized flow data, or discretized flow data obtained after sequential processing such as dirty data cleaning, positive and negative sample matching, and continuous data discretization is performed on the flow data. Among them, a preferred embodiment is provided: after the processed flow data is obtained, the coefficient matrix is used for carrying out storage operation, flow data samples are amplified, data dispersion is strengthened, the quality of flow data obtained by a port calculation center is improved, and therefore the accuracy of a prediction result is improved.
Step S202: and predicting port flow characteristics of the system on the cloud according to the flow data.
In this embodiment, the port computing center predicts port traffic characteristics of the cloud system according to the traffic data; and predicting port flow characteristics of the on-cloud system according to the cleaned flow data or predicting the port flow characteristics of the on-cloud system according to the discretized flow data. The method comprises the following steps of carrying out a series of processing operations such as dirty data cleaning, positive and negative sample proportioning and continuous data discretization on flow data, then carrying out sample volume expansion by using a sparse matrix, strengthening data discretization, and predicting port flow characteristics of a system on the cloud according to the flow data after sample volume expansion.
The flexible port computing center extracts flow data time information and flow data peak value information from flow data stored in the offline data center, and port flow characteristics of the cloud system are predicted by using a Bayesian network model according to the flow data time information and the flow data peak value information.
Step S203: and sending an indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system to the cloud system according to the port flow characteristics of the on-cloud system.
In this embodiment, the port computing center determines, according to the port traffic characteristics, a port maximum bandwidth or a port maximum traffic threshold of a port of the cloud system within a period of time, and sends an indication message for adjusting the port bandwidth or the port traffic of the cloud system to the cloud system. Wherein the indication information comprises a port maximum bandwidth or a port maximum traffic threshold of a port of the on-cloud system over a period of time.
In the foregoing embodiment, a traffic characteristic prediction method is provided, and specifically, a method for predicting traffic characteristics through an intelligent algorithm is provided, where the traffic characteristic prediction method can predict, through the intelligent algorithm, traffic data characteristics of a port of a cloud system in a current period of time in real time according to a difference in traffic, and output an instruction according to the traffic data characteristics, so that port bandwidth adjustment and traffic allocation can be performed in real time.
In addition, the present application also provides a traffic data processing method, which may be applied to a port traffic intelligent control system of a cloud system, please refer to fig. 3, which is a flowchart of an embodiment of the traffic data processing method of the present application.
In this embodiment, a method for processing traffic data includes the following steps:
step S301: and storing the flow data provided by the cloud system.
Step S302: and cleaning the stored flow data, and providing the cleaned flow data for a port calculation center.
In this embodiment, the offline data center performs traffic data cleaning on the stored traffic data, removes dirty data in the off-peak period, and then deletes negative samples outside the preset proportion from the traffic data provided by the cloud system according to the preset proportion of positive and negative samples of the traffic data and the number of positive samples in the traffic data provided by the cloud system, thereby obtaining the cleaned traffic data. Furthermore, the off-line data center can also perform discretization processing on the cleaned flow data to obtain discretized flow data, and provide the discretized flow data for the port calculation center.
In addition, the embodiment provides a preferred implementation manner that the offline data center stores the discretized data by using a sparse matrix, so as to expand the number of samples; the method comprises the steps that an offline data center obtains flow data after sample amount expansion, and provides the flow data after the sample amount expansion for a port calculation center, wherein the quantity of the flow data after the sample amount expansion is more than that of the flow data after discretization.
In the above embodiment, a flow data processing method is provided, where flow data provided by a cloud system is first stored, then the stored flow data is cleaned, the cleaned flow data is provided to a port computing center, dirty data in an off-peak period is removed, then negative samples outside a preset proportion are deleted from the flow data provided by the cloud system according to the proportion of positive and negative samples in the preset flow data and the number of positive samples in the flow data provided by the cloud system, so as to obtain cleaned flow data, and further, the cleaned flow data is discretized, so as to obtain discretized flow data. The flow data processing method can obtain high-quality flow data, so that the accuracy of the prediction result of the calculation center is ensured.
Fig. 4 is a flowchart of an embodiment of a method for controlling a port of a cloud system according to the present application. Because the embodiment of the method is similar to the embodiment of the system, the description is relatively simple, and the related points can be referred to the description of the embodiment of the system. The following description of the apparatus embodiments is merely exemplary.
The application provides a port control method of a cloud system, which comprises the following steps:
step S401: and acquiring indication information of port adjustment of the system on the cloud.
Step S402: and adjusting the bandwidth information and/or the port traffic information of the port of the system on the cloud according to the indication information of the port adjustment of the system on the cloud to obtain the adjusted port bandwidth information and/or the adjusted port traffic information of the system on the cloud.
In this embodiment, the cloud system obtains adjusted port bandwidth information or port traffic information according to an indication message sent by the port computing center to adjust the port bandwidth or port traffic of the cloud system, and sends the adjusted port bandwidth information or the adjusted port traffic information to the port forwarding center.
Further, the obtained indication message for adjusting the port bandwidth and/or the port traffic of the cloud system specifically includes: a port maximum bandwidth and/or a port maximum traffic threshold for a port of the system on the cloud over a period of time.
Step S403: and sending the adjusted port bandwidth information and/or the adjusted port traffic information of the on-cloud system to the port forwarding center according to the port bandwidth information and/or the port traffic information of the on-cloud system.
In this embodiment, the cloud system may further receive alarm indication information sent by a port of the cloud system, that is, when the port traffic data of the cloud system approaches or exceeds a current limit value, the port of the cloud system sends alarm indication information indicating that the traffic data is too large to the cloud system, and the cloud system automatically adjusts the port bandwidth or the port traffic according to the alarm information sent by the port of the cloud system, and sends the adjusted port bandwidth information or the adjusted port traffic information to the port forwarding center.
In addition, the port forwarding center may also receive traffic data of the client, and before forwarding the traffic data of the client to each port of the cloud system, it needs to receive adjusted port bandwidth information or adjusted port traffic information sent by the cloud system, and forward the traffic information obtained by the client to a suitable port of the cloud system according to the received bandwidth information or the adjusted port traffic information. It should be noted that the port adjustment information includes port number information of a specific port, bandwidth information after the port is adjusted, current limit information, bearable peak value information, and the like.
Corresponding to the above flow characteristic prediction method, the present application also provides a flow characteristic prediction apparatus, please refer to fig. 7, which is a schematic diagram of an embodiment of the flow characteristic prediction apparatus of the present application. Since the embodiment of the present apparatus is similar to the embodiment of the flow characteristic prediction method described above, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the embodiment of the flow characteristic prediction method, and the following description of the embodiment of the apparatus is only illustrative.
In this embodiment, a flow characteristic prediction apparatus includes:
an obtaining unit 701 is configured to obtain traffic data.
In this embodiment, the flow data acquired by the acquiring unit 701 may be at least one of flow data after data cleaning, flow data after positive and negative sample matching, and discretized flow data, or discretized flow data obtained by sequentially processing the flow data such as dirty data cleaning, positive and negative sample matching, and continuous data discretization.
A predicting unit 702, configured to predict, according to the traffic data, a port traffic characteristic of the system on the cloud.
In this embodiment, the predicting unit 702 predicting the port traffic characteristics of the system on the cloud according to the traffic data includes: and predicting port flow characteristics of the on-cloud system according to the cleaned flow data or predicting the port flow characteristics of the on-cloud system according to the discretized flow data. The method comprises the following steps of carrying out a series of processing operations such as dirty data cleaning, positive and negative sample proportioning and continuous data discretization on flow data, then carrying out sample volume expansion by using a sparse matrix, strengthening data discretization, and predicting port flow characteristics of a system on the cloud according to the flow data after sample volume expansion.
The sending unit 703 is configured to send, to the cloud system, an indication message for adjusting the port bandwidth and/or the port traffic of the cloud system according to the port traffic characteristics of the cloud system.
In this embodiment, the sending unit 703 determines, according to the port traffic characteristics, the maximum port bandwidth or the maximum port traffic threshold of the port of the cloud system within a period of time, and sends an indication message for adjusting the port bandwidth or the port traffic of the cloud system to the cloud system.
In addition, the present application also provides a traffic data processing apparatus, which is similar to the above described traffic data processing method, so that the description is relatively simple, and for relevant points, reference is made to the section of the embodiment of the traffic data processing method, and refer to fig. 8, which is a schematic diagram of the embodiment of the traffic data processing apparatus of the present application.
In this embodiment, a traffic data processing apparatus includes:
the storage unit 801 is configured to store traffic data provided by the cloud system.
A cleaning unit 802, configured to perform traffic data cleaning on the stored traffic data.
In this embodiment, the cleaning unit 802 performs traffic data cleaning on the stored traffic data, removes the dirty data in the off-peak period, and then deletes negative samples outside the preset proportion from the traffic data provided by the storage unit 801 according to the preset proportion of positive and negative samples of the traffic data and the number of positive samples in the traffic data provided by the cloud system, so as to obtain the cleaned traffic data
A sending unit 803, configured to provide the cleaned flow data to the port computing center.
The present application further provides a port control device of a cloud system, which is similar to the above-mentioned port control method of a cloud system, so that the description is relatively simple, and for relevant points, please refer to the port control method of a cloud system in the embodiment section, and refer to fig. 9, which is a schematic diagram of an embodiment of a port control device of a cloud system according to the present application.
In this embodiment, a port control apparatus of a system on the cloud includes the following components:
an obtaining unit 901 is configured to obtain indication information of port adjustment of a system on the cloud.
An adjusting unit 902, configured to adjust bandwidth information and/or port traffic information of a port of the cloud system according to the indication information of port adjustment of the cloud system, to obtain adjusted port bandwidth information and/or port traffic information of the cloud system.
In this embodiment, the cloud system obtains port bandwidth information or port traffic information adjusted by the adjusting unit 902 according to an indication message sent by the port computing center to adjust the port bandwidth or the port traffic of the cloud system, and the adjusting unit 902 sends the adjusted port bandwidth information or the adjusted port traffic information to the port forwarding center.
A sending unit 903, configured to send the adjusted port bandwidth information and/or the adjusted port traffic information of the cloud system to the port forwarding center according to the port bandwidth information and/or the port traffic information of the cloud system.
In this embodiment, the cloud system may further receive alarm indication information sent by a port of the cloud system, that is, when the port traffic data of the cloud system approaches or exceeds the current limit value, the port of the cloud system sends alarm indication information with excessive traffic data to the cloud system, the cloud system automatically adjusts the port bandwidth or the port traffic according to the alarm information sent by the port of the cloud system, and the sending unit 903 sends the adjusted port bandwidth information or the adjusted port traffic information to the port forwarding center.
Although the present application has been described with reference to the preferred embodiments, it is not intended to limit the present application, and those skilled in the art can make variations and modifications without departing from the spirit and scope of the present application, therefore, the scope of the present application should be determined by the claims that follow.

Claims (29)

1. The utility model provides a port flow intelligence control system of system on cloud which characterized in that includes: the system comprises an offline data center, a port calculation center, a cloud system and a port forwarding center;
the off-line data center is used for storing the flow data provided by the cloud system;
the port computing center is configured to obtain traffic data stored in the offline data center, predict a port traffic characteristic of a cloud system according to the traffic data, and send an indication message for adjusting a port bandwidth and/or a port traffic of the cloud system to the cloud system according to the port traffic characteristic, where the predicting the port traffic characteristic of the cloud system according to the traffic data includes: extracting flow data time information and flow data peak value information from flow data provided by the offline data center, and predicting port flow characteristics to be accessed by the cloud system according to the flow data time information and the flow data peak value information;
the cloud system is used for obtaining adjusted port bandwidth information and/or port flow information according to the indication message sent by the port calculation center, and sending the adjusted port bandwidth information and/or the adjusted port flow information to the port forwarding center;
and the port forwarding center is used for sending the flow data obtained from the client to the port of the cloud system according to the adjusted port bandwidth information and/or the adjusted port flow information sent by the cloud system.
2. The intelligent port traffic control system according to claim 1,
the off-line data center is also used for carrying out flow data cleaning on the flow data provided by the cloud system and providing the cleaned flow data for the port computing center;
the port computing center is specifically configured to obtain the cleaned flow data provided by the offline data center, and predict port flow characteristics of the system on the cloud according to the cleaned flow data.
3. The intelligent port traffic control system according to claim 2,
the offline data center is specifically configured to delete negative samples outside a preset positive-negative sample ratio from the flow data provided by the cloud system according to the preset positive-negative sample ratio of the flow data and the number of positive samples in the flow data provided by the cloud system, so as to obtain cleaned flow data.
4. The intelligent port traffic control system according to claim 1,
the off-line data center is further used for carrying out discretization processing on the flow data provided by the cloud system, obtaining discretized flow data and providing the discretized flow data to the port computing center;
the port computing center is specifically configured to obtain discretized flow data provided by the offline data center, and predict port flow characteristics of the system on the cloud according to the discretized flow data.
5. The intelligent port traffic control system according to claim 1,
the offline data center is specifically configured to store the flow data provided by the cloud system by using a sparse matrix, obtain the flow data after sample size expansion, and provide the flow data after sample size expansion for the port computation center, where the number of the flow data after sample size expansion is greater than the number of the flow data provided by the cloud system;
the port computing center is specifically configured to obtain the flow data provided by the offline data center after the sample size is expanded, and predict a port flow characteristic of the system on the cloud according to the flow data after the sample size is expanded.
6. The intelligent port traffic control system according to claim 1, wherein the predicting port traffic data characteristics of the system on the cloud in advance according to the traffic data time information and the traffic data peak information includes:
and predicting port traffic characteristics of the cloud system by using a Bayesian network model according to the traffic data time information and the traffic data peak value information, wherein the Bayesian network model is a network model for predicting the port traffic characteristics of the cloud system according to the traffic data time information and the traffic data peak value information.
7. The intelligent port traffic control system according to claim 6, wherein the Bayesian network model is trained by:
and clustering the flow data stored in the off-line data center according to the time dimension and the peak value dimension to obtain the time information and the peak value information of the flow data after clustering, normalizing the time information and the peak value information of the flow data after clustering to obtain the time information and the peak value information of the flow data after normalization, and training the Bayesian network model by using the time information and the peak value information of the flow data after normalization.
8. The intelligent port traffic control system according to claim 1, wherein the port traffic characteristics include at least one of the following information:
a peak flow value of a port of a system on the cloud within a period of time;
traffic change information of a port of a system on the cloud within a period of time;
total flow information of ports of the on-cloud system over a period of time;
average traffic information per unit time over a time range for a port of a system on the cloud.
9. The intelligent port traffic control system according to claim 1,
the port computing center is specifically configured to determine, according to the port traffic characteristics, a port maximum bandwidth and/or a port maximum traffic threshold of a port of the cloud system within a period of time;
the method comprises the step of sending an indication message for adjusting port bandwidth and/or port traffic of the on-cloud system to the cloud system, wherein the indication message comprises port maximum bandwidth and/or port maximum traffic threshold of a port of the on-cloud system in a period of time.
10. The system according to claim 1, wherein the cloud system is specifically configured to obtain adjusted port bandwidth information and/or port traffic information according to the indication message for adjusting the port bandwidth and/or the port traffic of the cloud system, which is sent by the port computing center, and send the adjusted port bandwidth information and/or the adjusted port traffic information to the port forwarding center.
11. The intelligent port traffic control system according to claim 1,
the cloud system port is used for sending alarm information to the cloud system when the flow data of the port reaches or exceeds a bearable peak value;
the cloud system can also be used for automatically adjusting the port bandwidth and/or the port flow according to alarm information sent by a system port on the cloud, and sending adjusted port bandwidth information and/or adjusted port flow information to the port forwarding center.
12. A method for predicting flow characteristics, comprising:
acquiring flow data;
predicting port traffic characteristics of the system on the cloud according to the traffic data, comprising: extracting flow data time information and flow data peak value information from the flow data, and predicting port flow characteristics to be accessed by the cloud system according to the flow data time information and the flow data peak value information;
and sending an indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system to the cloud system according to the port flow characteristics of the on-cloud system.
13. The method according to claim 12, wherein the obtaining of the traffic data and the predicting of the port traffic characteristics of the cloud system according to the traffic data specifically include:
acquiring cleaned flow data;
and predicting the port flow characteristics of the cloud system according to the cleaned flow data.
14. The method for predicting traffic characteristics according to claim 12, wherein the obtaining of the traffic data and the predicting of the port traffic characteristics of the cloud system according to the traffic data specifically include:
obtaining discretized flow data;
and predicting port flow characteristics of the on-cloud system according to the discretized flow data.
15. The method for predicting traffic characteristics according to claim 12, wherein the obtaining of the traffic data and the predicting of the port traffic characteristics of the cloud system according to the traffic data specifically include:
obtaining flow data after sample size expansion;
and predicting the port flow characteristics of the system on the cloud according to the flow data after the sample size is expanded.
16. The method for predicting traffic characteristics according to claim 12, wherein the obtaining of the traffic data and the predicting of the port traffic characteristics of the cloud system according to the traffic data specifically include:
extracting flow data time information and flow data peak value information from flow data stored in an offline data center;
predicting port flow characteristics of the cloud system by using a Bayesian network model according to the flow data time information and the flow data peak value information;
the Bayesian network model is a network model used for predicting port traffic characteristics of the system on the cloud according to traffic data time information and traffic data peak value information.
17. The traffic characteristic prediction method according to claim 12, wherein the on-cloud system port traffic characteristics include at least one of:
a peak flow value of a port of a system on the cloud within a period of time;
traffic change information of a port of a system on the cloud within a period of time;
total traffic information for ports of the on-cloud system over a period of time;
average traffic information per unit time over a time range for a port of a system on the cloud.
18. The traffic characteristic prediction method according to claim 12, wherein the sending, to the cloud system, an instruction message for adjusting a port bandwidth and/or a port traffic of the cloud system according to the port traffic characteristic of the cloud system specifically includes:
determining the maximum port bandwidth and/or the maximum port traffic threshold of a port of the system on the cloud within a period of time according to the port traffic characteristics;
sending an indication message for adjusting the port bandwidth and/or the port traffic of the on-cloud system to the cloud system, wherein the indication message comprises the port maximum bandwidth and/or the port maximum traffic threshold of the port of the on-cloud system in a period of time.
19. A method for processing traffic data, comprising:
storing flow data provided by the cloud system;
the stored flow data is cleaned, the cleaned flow data is provided for a port computing center, and the port computing center is used for predicting port flow characteristics of a cloud system according to the cleaned flow data and sending an indication message for adjusting the port bandwidth and/or the port flow of the cloud system to the cloud system according to the port flow characteristics;
predicting port flow characteristics of a cloud system according to the cleaned flow data, wherein the predicting of the port flow characteristics of the cloud system comprises the following steps: and extracting flow data time information and flow data peak value information from the cleaned flow data, and predicting port flow characteristics to be accessed by the cloud system according to the flow data time information and the flow data peak value information.
20. The flow data processing method according to claim 19, wherein the cleaning of the stored flow data and the provision of the cleaned flow data to the port calculation center specifically include:
and deleting negative samples beyond the proportion from the flow data provided by the cloud system according to the proportion of positive samples and negative samples of the preset flow data and the number of positive samples in the flow data provided by the cloud system to obtain cleaned flow data, and providing the cleaned flow data to a port calculation center.
21. The traffic data processing method according to claim 19, wherein the storing the traffic data provided by the cloud system, performing traffic data cleaning on the stored traffic data, and providing the cleaned traffic data to the port computation center, further comprises:
and carrying out discretization processing on the flow data provided by the cloud system to obtain discretized flow data, and providing the discretized flow data to a port calculation center.
22. The flow data processing method according to claim 19, wherein the storing the flow data provided by the cloud system, performing flow data cleaning on the stored flow data, and providing the cleaned flow data to the port computation center specifically includes:
storing the flow data provided by the cloud system by using a sparse matrix;
obtaining the flow data after sample size expansion, and providing the flow data after sample size expansion for a port calculation center, wherein the quantity of the flow data after sample size expansion is more than that of the flow data provided by the cloud system.
23. A port control method of a system on the cloud is characterized by comprising the following steps:
acquiring indication information of port adjustment of a system on the cloud;
according to the indication information of the port adjustment of the on-cloud system, adjusting the bandwidth information and/or the port flow information of the port of the on-cloud system to obtain the adjusted port bandwidth information and/or the adjusted port flow information of the on-cloud system;
sending the adjusted port bandwidth information and/or the adjusted port traffic information of the on-cloud system to the port forwarding center according to the port bandwidth information and/or the port traffic information of the on-cloud system;
the port adjustment instruction of the on-cloud system is obtained according to predicted port flow characteristics of the on-cloud system, the port flow characteristics are obtained according to flow data time information and flow data peak value information, and the flow data time information and the flow data peak value information are obtained from flow data provided by the cloud system and stored in an off-line data center.
24. The method according to claim 23, wherein the adjusting bandwidth information and/or port traffic information of the port of the cloud system according to the indication information of the port adjustment of the cloud system to obtain the adjusted port bandwidth information and/or port traffic information of the cloud system specifically includes:
and obtaining adjusted port bandwidth information and/or port flow information according to the indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system sent by the port computing center, and sending the adjusted port bandwidth information and/or the adjusted port flow information to the port forwarding center.
25. The method according to claim 23, wherein the obtaining of the indication message for adjusting the port bandwidth and/or the port traffic of the cloud system specifically includes: a port maximum bandwidth and/or a port maximum traffic threshold for a port of the system on the cloud over a period of time.
26. The port control method of the system on the cloud according to claim 23, further comprising:
receiving alarm information sent by a port of a cloud system;
automatically adjusting the port bandwidth and/or the port flow according to alarm information sent by a system port on the cloud;
and sending the adjusted port bandwidth information and/or the adjusted port flow information to the port forwarding center.
27. A flow rate characteristic prediction apparatus, comprising:
an acquisition unit configured to acquire flow data;
the prediction unit is used for predicting port flow characteristics of the system on the cloud according to the flow data;
the sending unit is used for sending an indication message for adjusting the port bandwidth and/or the port flow of the on-cloud system to the cloud system according to the port flow characteristics of the on-cloud system;
the predicting port traffic characteristics of the system on the cloud according to the traffic data includes: extracting flow data time information and flow data peak value information from flow data provided by an offline data center, and predicting port flow characteristics to be accessed by the system on the cloud according to the flow data time information and the flow data peak value information.
28. A traffic data processing apparatus, comprising:
the storage unit is used for storing the flow data provided by the cloud system;
the cleaning unit is used for cleaning the stored flow data;
the sending unit is used for providing the cleaned flow data to the port calculation center;
the port computing center is used for predicting port flow characteristics of a cloud system according to the cleaned flow data and sending an indication message for adjusting the port bandwidth and/or the port flow of the cloud system to the cloud system according to the port flow characteristics;
the predicting port flow characteristics of the cloud system according to the cleaned flow data comprises the following steps: and extracting flow data time information and flow data peak value information from the cleaned flow data, and predicting port flow characteristics to be accessed by the cloud system according to the flow data time information and the flow data peak value information.
29. A port control apparatus of a system on the cloud, comprising:
the acquisition unit is used for acquiring indication information of port adjustment of the system on the cloud;
the adjusting unit is used for adjusting the bandwidth information and/or the port flow information of the port of the on-cloud system according to the indication information of the port adjustment of the on-cloud system to obtain the adjusted port bandwidth information and/or the adjusted port flow information of the on-cloud system;
a sending unit, configured to send, to the port forwarding center, the adjusted port bandwidth information and/or the adjusted port traffic information of the cloud system according to the port bandwidth information and/or the port traffic information of the cloud system;
the port adjustment instruction of the on-cloud system is obtained according to predicted port flow characteristics of the on-cloud system, the port flow characteristics are obtained according to flow data time information and flow data peak value information, and the flow data time information and the flow data peak value information are obtained from flow data provided by the cloud system and stored in an off-line data center.
CN201810872051.1A 2018-08-02 2018-08-02 Port flow intelligent control system of on-cloud system Active CN108924057B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810872051.1A CN108924057B (en) 2018-08-02 2018-08-02 Port flow intelligent control system of on-cloud system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810872051.1A CN108924057B (en) 2018-08-02 2018-08-02 Port flow intelligent control system of on-cloud system

Publications (2)

Publication Number Publication Date
CN108924057A CN108924057A (en) 2018-11-30
CN108924057B true CN108924057B (en) 2022-06-03

Family

ID=64394277

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810872051.1A Active CN108924057B (en) 2018-08-02 2018-08-02 Port flow intelligent control system of on-cloud system

Country Status (1)

Country Link
CN (1) CN108924057B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110120975B (en) * 2019-05-06 2020-01-03 南京瑞贻电子科技有限公司 Cloud platform connection port control device and control method
CN112235152B (en) * 2020-09-04 2022-05-10 北京邮电大学 Flow size estimation method and device
CN112202619A (en) * 2020-10-12 2021-01-08 浪潮云信息技术股份公司 Intelligent cloud computing network flow adjusting and optimizing system and method
CN113467525A (en) * 2021-07-28 2021-10-01 中国银行股份有限公司 Interface call flow control method and device
CN116016306A (en) * 2022-12-15 2023-04-25 中国联合网络通信集团有限公司 Data traffic forwarding method, device, equipment and storage medium
CN116915517B (en) * 2023-09-14 2023-11-24 厦门快快网络科技有限公司 Cloud service resource risk security management method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105049262A (en) * 2015-08-24 2015-11-11 中国联合网络通信集团有限公司 Method and device for adjusting flow loads of ports
CN106302221A (en) * 2016-09-12 2017-01-04 中国联合网络通信集团有限公司 Traffic scheduling method based on end office's cloud and system
CN106453111A (en) * 2015-08-11 2017-02-22 中兴通讯股份有限公司 Traffic management method based on aggregation link and traffic management device thereof
CN106712899A (en) * 2016-11-25 2017-05-24 浙江宇视科技有限公司 Port rate adjustment method and device
CN106878116A (en) * 2016-12-06 2017-06-20 新华三技术有限公司 Limit Rate method and apparatus

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11509501B2 (en) * 2016-07-20 2022-11-22 Cisco Technology, Inc. Automatic port verification and policy application for rogue devices

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106453111A (en) * 2015-08-11 2017-02-22 中兴通讯股份有限公司 Traffic management method based on aggregation link and traffic management device thereof
CN105049262A (en) * 2015-08-24 2015-11-11 中国联合网络通信集团有限公司 Method and device for adjusting flow loads of ports
CN106302221A (en) * 2016-09-12 2017-01-04 中国联合网络通信集团有限公司 Traffic scheduling method based on end office's cloud and system
CN106712899A (en) * 2016-11-25 2017-05-24 浙江宇视科技有限公司 Port rate adjustment method and device
CN106878116A (en) * 2016-12-06 2017-06-20 新华三技术有限公司 Limit Rate method and apparatus

Also Published As

Publication number Publication date
CN108924057A (en) 2018-11-30

Similar Documents

Publication Publication Date Title
CN108924057B (en) Port flow intelligent control system of on-cloud system
CN110311812B (en) Network analysis method, device and storage medium
CN104901989B (en) A kind of Site Service offer system and method
CN111741073B (en) Electric power data transmission system based on 5G communication network
CN107708152B (en) Task unloading method of heterogeneous cellular network
CN113242526A (en) Cloud computing server room real-time monitoring system
CN109451534A (en) A kind of dynamic control method and device for QoS flow in the management of 5G system session
CN110830384B (en) Method, device and system for limiting service flow
CN112926791A (en) Computer room temperature distribution prediction method and system
CN106101264A (en) Content distributing network daily record method for pushing, device and system
WO2020093502A1 (en) Nominal bandwidth adjusting method and device
CN115016923A (en) Intelligent processing method for Internet of things data based on edge gateway
CN108684058B (en) LTE cell load evaluation method based on FAHP-ensemble learning
CN107579858A (en) The alarm method and device of cloud main frame, communication system
CN111901134B (en) Method and device for predicting network quality based on recurrent neural network model (RNN)
US20150289162A1 (en) Methods circuits devices systems and associated computer executable code for implementing cell congestion detection in a mobile network
Ahmed et al. Predicting high delays in mobile broadband networks
CN107634881A (en) A kind of network or video traffic detection system and method
CN109617738B (en) Method, system and non-volatile storage medium for user service scaling
CN111476401A (en) Flight schedule optimization method and system
EP3413595B1 (en) Processing method, device and system for service optimization
CN113794646B (en) Monitoring data transmission system and method for energy industry
CN106612241B (en) Service control method and device
CN113301126B (en) Edge computing method suitable for heterogeneous networking gateway
CN115484167A (en) Network slice shutdown method in communication network, computer device, and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant