JP6562358B2 - Network management server, network management method, and program - Google Patents

Description

  The present disclosure relates to a network management system, a network management method, and a program for controlling a traffic path and a flow rate according to a network load.

  The traffic flowing through the network varies depending on the situation of users and services. For example, if the connection point of a user or service moves, the route through which traffic flows changes accordingly. The flow rate also changes depending on the number of users and the popularity of services. These changes can cause traffic to concentrate on specific links and switches, causing congestion. In addition, access may be concentrated on a specific server, exceeding the capacity. To avoid these overload conditions, the network operator controls the traffic path and flow rate.

Takeshi Fukumoto, Masami Iio, Kiyoshi Ueda, “Efforts to Realize Future Network Control Nodes”, NTT Technical Journal, pp. 23-27, Mar. 2012. Yosuke Rema, Masanori Sugawara, “Study on an automatic estimation system for the number of people using an environmental sensor”, http: // staff. Miyakyo-u. ac. jp / ˜msugawa / phase_tohoku201503_zanma. pdf (search May 24, 2016) “Developed and put to practical use“ next-generation high-performance weather radar (MP radar) ”capable of accurately grasping local weather phenomena”, http://www8.cao.go.jp/cstp /Tyousakai/suisin/haihu13/sanko1-2.pdf (searched on May 24, 2016)

  The conventional network management system monitors the traffic state and performs control according to the change. However, this approach can only be controlled after the traffic changes. For this reason, there is a risk that some influence is exerted on the communication quality before the control is executed. For example, Non-Patent Document 1 states that “the system load status is monitored, and in high-load systems, the application software is installed on the server of the resource pool and expanded.” ing.

  For example, when a large influence such as live distribution is clearly predicted, a change in traffic is predicted in advance and control may be performed in advance. However, this approach is limited to planned events where traffic changes can be clearly predicted. There is a problem that it is not possible to cope with so-called flash cloud accompanying sudden news.

  Therefore, an object of the present invention is to provide a network management system, a network management method, and a program thereof that can cope with the above-described flash cloud.

  In order to achieve the above object, the network management system according to the present invention performs network management based on a wide range of information without being limited to the traffic state.

Specifically, a network management system according to the present invention is a network management system that controls the route and flow rate of network traffic,
One or more sensors connected to the network and collecting information affecting traffic;
A predicting unit that predicts traffic fluctuations based on the sensor information;
Based on the prediction of the prediction unit, an adjustment execution unit that changes the channel number or position of the access point that the user accesses;
It is characterized by providing.

Further, the network management method according to the present invention is a network management method for controlling the route and flow rate of network traffic,
Collect information that affects traffic from one or more sensors connected to the network, predict traffic fluctuations based on the sensor information, and change the number or location of channels of access points that users access It is characterized by that.

  This network management system includes a network in which one or more “sensors” and one or more management servers are connected. A “sensor” is any device that can acquire information outside the network and transmit it to the management server via the network. The sensor transmits the acquired information to the management server periodically or in accordance with an instruction from the management server.

  The management server predicts future traffic changes based on information collected from the sensors. Then, the management server instructs the optimum route at the optimum time with respect to the predicted traffic change, and controls the flow rate. That is, the management server predicts the flash cloud based on information from the sensor, and performs network management according to the prediction.

  Therefore, the present invention can provide a network management system, a network management method, and a program thereof that are compatible with a flash cloud.

  For example, the sensor acquires information for estimating the number of people, and the prediction unit estimates the number of people around the access point from the sensor information and predicts traffic fluctuations. It is characterized by that.

  For example, the sensor acquires information related to weather, and the prediction unit estimates the number of people around the access point from the information of the sensor and predicts traffic fluctuation.

For example, the network management system further includes a cache server for temporarily storing data,
The sensor acquires information related to a disaster, the prediction unit predicts an increase in traffic in the network or interruption of communication between the network and the outside from the information of the sensor, and the adjustment execution unit includes the prediction The data is moved from the external server outside the network to the cache server based on the prediction of the network.

  The present invention is a program for causing a computer to function as the network management system. The apparatus of the present invention can be realized by a computer and a program, and can be recorded on a recording medium or provided through a network.

  The present invention can provide a network management system, a network management method, and a program thereof that are compatible with a flash cloud.

It is a figure explaining the structure of the network management system which concerns on this invention. It is a sequence diagram explaining operation | movement of the network management system which concerns on this invention. It is a functional block diagram explaining the management system of the network management system which concerns on this invention. It is a flowchart explaining operation | movement of the management system of the network management system which concerns on this invention. It is a figure explaining the structure of the network management system which concerns on this invention. It is a sequence diagram explaining operation | movement of the network management system which concerns on this invention. It is a functional block diagram explaining the management system of the network management system which concerns on this invention. It is a flowchart explaining operation | movement of the management system of the network management system which concerns on this invention. It is a figure explaining the structure of the network management system which concerns on this invention. It is a sequence diagram explaining operation | movement of the network management system which concerns on this invention. It is a functional block diagram explaining the management system of the network management system which concerns on this invention. It is a flowchart explaining operation | movement of the management system of the network management system which concerns on this invention.

  Embodiments of the present invention will be described with reference to the accompanying drawings. The embodiments described below are examples of the present invention, and the present invention is not limited to the following embodiments. In the present specification and drawings, the same reference numerals denote the same components.

(Embodiment 1)
FIG. 1 is a diagram illustrating a network management system 301 according to this embodiment. The network management system 301 is a network management system that controls the traffic path and flow rate of the network 10.
One or more sensors 20 connected to the network 10 and collecting information affecting traffic;
A prediction unit 12 that predicts traffic fluctuations based on information from the sensor 20;
Based on the prediction of the prediction unit 12, an adjustment execution unit 13 that changes the number of channels or the position of the access point 30 accessed by the user;
It is characterized by providing.
Note that the management server 11 includes a prediction unit 12 and an adjustment execution unit 13.

  In particular, the sensor 20 of the network management system 301 obtains information for estimating the number of people, and the prediction unit 12 estimates the number of people around the access point 30 from the information of the sensor 20, and traffic It is characterized by predicting the fluctuation of

As shown in FIG. 1, the network management system 301 is connected to a sensor 20 that can observe information about the number of people. The sensor 20-1 is, for example, a camera. Images taken with a camera can be used to estimate the number of people using image analysis techniques such as face recognition. The sensor 20-2 and the sensor 20-3 are, for example, a CO2 gas sensor or a hygrometer. A technique for estimating the number of people from a CO ₂ gas sensor and a hygrometer has been developed (see, for example, Non-Patent Document 2). If the number of users accessing the access point 30 from these sensors 20 can be known, it can be predicted that the traffic will change accordingly.

  FIG. 2 is a sequence diagram of the network management system 301. For example, the management server 11 increases the number of users who access the access point after a predetermined time when the number of people around the access point 30 (the range in which the access point can be accessed) exceeds a predetermined number based on information from the sensor 20. Judge that. The predetermined time may be a predetermined time, or may be longer or shorter based on the rate of increase in the number of people. When the management server 11 anticipates an increase in traffic due to an increase in users, the number of accommodated users is improved by performing control such as reducing the radio channel allocated per user or increasing the communication capacity allocated to the access point 30. To do. On the other hand, when a decrease in the number of users is detected, a decrease in traffic is expected, so the operation reverse to the above is performed.

  Moreover, if the management server 11 can acquire the distribution and movement of people from the sensor 20, the position of the access point 30 may be moved according to the distribution and movement of persons using an actuator. Actuators may be used for sensors to improve accuracy.

  FIG. 3 is a functional block diagram illustrating the management server 11. FIG. 4 is a flowchart of traffic control performed by the management server 11 based on the movement of a person. The sensor information is received and analyzed by the image analysis unit 14 and the gas / humidity analysis unit 15 (step S11). The number of persons estimation / traffic prediction unit 12 predicts an increase or decrease in the number of users from the sensor information (step S12). The channel optimization unit 16 increases or decreases the allocated channel per user based on the predicted increase or decrease of the number of users (steps S13 and S14). Finally, the channel assignment execution unit 13 sends a message to the wireless access point 30 to change the assignment (step S15).

(Embodiment 2)
FIG. 5 is a diagram illustrating the network management system 302 according to the present embodiment. The network management system 302 is different from the network management system 301 in FIG.

  In particular, the sensor 20 of the network management system 302 acquires information related to weather, and the prediction unit 12 estimates the number of people around the access point 30 from the information of the sensor 20 and predicts traffic fluctuations. It is characterized by.

  As shown in FIG. 5, the network management system 302 is connected to a weather radar (sensors 20-4 and 20-5), and can predict local weather fluctuations (see, for example, Non-Patent Document 3). ). When rainfall is predicted by these sensors 20, the indoor population increases, so that it is possible to predict the movement of the traffic generation location and the traffic change.

FIG. 6 is a sequence diagram of the network management system 302. For example, when the management server 11 obtains a rainfall forecast after a certain time around the access point 30-2 (a range accessible to the access point) by the information from the sensor 20, the access point in the room after the rain starts. It is determined that the number of users who access 30-1 increases. The predetermined time may be the expected rainfall time of the weather radar. When the management server 11 predicts a traffic change due to rainfall, the management server 11 changes the bandwidth assigned to each access point. Specifically, the management server 11 assigns the bandwidth assigned to the access point 30-2 to the router 35. Control is performed to decrease and increase the bandwidth allocated to the access point 30-1, thereby improving the number of users accommodated by the access point 30-1. On the other hand, when the end of rain is detected, the reverse operation is performed.
The network management system 302 can improve the quality of experience of the user by performing reservation control of the network band in accordance with the movement of the traffic generation source according to the weather.

FIG. 7 is a functional block diagram illustrating the management server 11. FIG. 8 is a flowchart of traffic control performed by the management server 11 based on the weather.
The weather prediction unit 17 receives the sensor information and predicts the weather (step S21). The number estimation / traffic prediction unit 12 predicts an increase or decrease in the number of accesses to the access point 30 from the predicted weather (step S22). The channel optimization unit 16 increases or decreases the bandwidth of each access point 30 based on the increase or decrease of the number of accesses (steps S23 and S24). Finally, a message is sent from the channel assignment execution unit 13 to each access point 30 to change the bandwidth assignment (step S25).

(Embodiment 3)
FIG. 9 is a diagram illustrating the network management system 303 of the present embodiment. The network management system 303 differs from the network management system 301 in FIG.

In particular, the network management system 303 further includes a cache server 36 that temporarily stores data,
The sensor 20 acquires information about the disaster, the prediction unit 12 predicts an increase in traffic in the network 10 or a blockage of communication between the network 10 and the outside from the information of the sensor 20, and the adjustment execution unit 13 Based on the twelve predictions, the data is moved from the external server 39 outside the network 10 to the cache server 36.

  As shown in FIG. 9, the network management system 303 is connected to a sensor 20 relating to a disaster such as a seismometer or a rain gauge. The sensor 20-6 is a seismometer, for example. The sensor 20-7 is, for example, a rain gauge. When a major disaster such as an earthquake or heavy rain occurs, user communication requests increase rapidly and traffic increases. In addition, the path between the network 10 and the outside may be lost. With these sensors 20, it can be determined that a disaster has occurred, and it can be predicted that traffic in the network 10 will increase or communication with outside will be interrupted.

  FIG. 10 is a sequence diagram of the network management system 303. For example, when the occurrence of a disaster is detected around the access point 30 based on information from the sensor 20, the management server 11 predicts that traffic in the network 10 increases or communication with the outside is interrupted. Then, the management server 11 requests the required server function from the external server 39 immediately after the occurrence of the disaster (before traffic increases or the route is lost) and moves it to the cache server 36 in the network 10. Communication with the outside becomes unnecessary due to movement of the server function. Specifically, the management server 11 instructs the cache server 36 to move functions based on the detection of the occurrence of a disaster. The cache server 36 requests and downloads necessary data from the external server 39.

  Moreover, the management server 11 reduces the radio channel allocated per user after a disaster occurrence detection. Thereby, the traffic increase of the network 10 can be suppressed at the time of a disaster. On the other hand, the server function is returned to the external server 39 after a certain period of time has elapsed since the occurrence of the disaster or according to an instruction from the network administrator, and the allocated amount per user of the wireless channel is restored.

FIG. 11 is a functional block diagram illustrating the management server 11. FIG. 12 is a flowchart of traffic control performed by the management server 11 based on the occurrence of a disaster.
The seismic intensity analysis unit 18 and the rainfall analysis unit 19 receive the sensor information and determine the occurrence of a disaster (step S31). The person estimation / traffic prediction unit 12 predicts a communication interruption with the outside of the network and an increase in traffic volume based on the determination of the occurrence of a disaster (step S32). For example, the number of persons estimation / traffic prediction unit 12 predicts that communication interruption or an increase in traffic volume occurs when a seismic intensity or rainfall that exceeds a predetermined value is observed. Alternatively, the person estimation / traffic prediction unit 12 may predict an increase in traffic volume proportional to seismic intensity or rainfall. If the seismic intensity or the rainfall is less than a predetermined value, the channel optimization unit 16 maintains the current radio channel allocated per user (step S33). On the other hand, if the seismic intensity or the rainfall is equal to or greater than the predetermined value, the channel optimization unit 16 instructs the cache server 36 to move the server function (Step S34), and reduces the radio channel allocated per user from the current amount. (Step S35). Finally, a message is sent from the channel assignment execution unit 13 to the access point 30 to change the assignment (step S36).

(Other embodiments)
Although some specific examples have been given above, the present invention is not limited to these examples in terms of sensors and prediction methods.

(Point of invention)
By predicting changes in traffic in advance based on various sensor information other than traffic, the network is controlled before the traffic actually starts to change.

(effect)
By predicting the change before the traffic actually starts to change and controlling the network, the risk of some influence on the communication quality can be avoided. In addition, it is possible to cope with sudden changes that are not planned.

10: Network 11: Management server 12: Prediction unit (number estimation / traffic prediction unit)
13: Adjustment execution unit (channel allocation execution unit)
14: Image analysis unit 15: Gas / humidity analysis unit 16: Channel optimization unit 17: Weather prediction unit 18: Seismic intensity analysis unit 19: Rainfall analysis unit 20, 20-1 to 7: Sensors 30, 30-1, 30- 2: Access point 35: Router 36: Cache server 39: External servers 301-303: Network management system

Claims (3)

A network management server that controls the route and flow rate of network traffic,
Based on the one or more sensor information to obtain information about the disaster, a prediction unit that predicts the interruption of communication with the increase or the network and external traffic within the network,
An adjustment execution unit that moves data from an external server outside the network to a cache server inside the network , based on the prediction of the prediction unit, and reduces the number of channels of access points that the user accesses;
A network management server comprising: A network management method in a network management server that controls the route and flow rate of network traffic,
A prediction step prediction unit, based on one or more sensor information to obtain information about the disaster, to predict the interruption of communication with the increase or the network and external traffic within the network,
The adjustment execution unit moves data from an external server outside the network to a cache server inside the network based on the prediction in the prediction step, and reduces the number of access point channels accessed by the user. An adjustment execution step;
The network management method characterized by performing . A program for causing a computer to function as the network management server according to claim 1 .

Images