JP2014087031A - Communication traffic predicting device, communication traffic predicting method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the accuracy of results of communication traffic prediction in traffic to which a plurality of communication services are multiplexed, and at the same time to reduce its calculation amount.SOLUTION: In a communication traffic predicting device, multiple communication services are divided into predetermined multiple groups. The communication traffic predicting device comprises: divided-group traffic measuring means for calculating the traffic of each communication service from traffic data, to which the multiple communication services are multiplexed, according to a processing method predetermined for each of the divided groups; traffic predicting means for predicting traffic of each group at a future time point according to the processing method predetermined for each of the divided groups on the basis of the calculated traffic of each communication service; and required facility calculation means for integrating the predicted traffic of each group at the future time point and for calculating a facility capacity absorbing short-term variations based on the integrated traffic.

Description

  The present invention predicts future traffic for a communication network in which various communication services such as a telephone service, an Internet connection service, and a video distribution service (multi-channel broadcasting service and video on demand service) are multiplexed. It is related to the technology.

  Each communication service has a quality of service or a quality of user experience that is necessary for the user of the communication service to use it comfortably. The quality of such a communication service is ensured and at the lowest possible cost. What can be provided is the purpose of the operation management business of the communication network equipment, and in order to achieve this goal, a business plan related to the design and operation of the equipment is formulated.

  For the above purpose, when planning the business plan for designing and operating communication network equipment, it is necessary to estimate the future traffic from several months to one year, which is the period required for construction such as equipment expansion. It is very important to predict the amount of the most busy traffic.

  Conventionally, future prediction of communication traffic has been performed using a method such as regression analysis with reference to a database in which past traffic data is accumulated for a long period of time. Specific contents of the traffic prediction technique are disclosed in, for example, the following patent documents or non-patent documents.

JP2004-23114 Japanese Patent No. 3737713 JP 07-235985 A JP2012-182677

Matsuda Jun et al. (1991), Switch traffic management technology based on daily traffic prediction method, NTT R & D Vol.40, No. 12, pp.1581-1588

  The first problem to be solved by the present invention is as follows.

  The traffic of a communication service originally has unique traffic characteristics depending on the system specifications of each communication service and the behavior of a user who uses the communication service. Therefore, it is extremely difficult to increase the accuracy of traffic prediction by predicting while adding up the traffic in which many communication services are multiplexed.

  On the other hand, a requirement for the conventional traffic prediction technique to be effective is the presence of traffic data over a sufficiently long period in the past. The communication network is a large-scale system, and the number of facilities to be configured is enormous. Therefore, the conventional technology has a problem that the traffic data to be maintained becomes huge and the amount of calculation for processing becomes enormous. When the traffic data to be maintained becomes huge and the amount of calculation for processing becomes enormous, there is a problem that, for example, the cost for traffic prediction increases and the processing speed decreases.

  The present invention has been made in view of the above points, and can improve the accuracy of communication traffic prediction results in traffic in which many communication services are multiplexed. At the same time, the required traffic data amount and calculation amount can be reduced. An object of the present invention is to provide a communication traffic prediction technique that can be reduced.

  The second problem to be solved by the present invention after solving the first problem is as follows.

  In a communication network, the daily fluctuation range of the traffic volume with respect to the entire communication service provided is very large compared to the conventional network. This is because there is a very large difference between the predicted maximum traffic volume and the traffic volume that occurs frequently under normal conditions, but on the other hand, the probability that the event that generates the maximum traffic volume will occur. Will gradually become very small. In the design and operation of communication network facilities, simply predicting the maximum amount of traffic, which is an extremely small probability, raises the cost of communication networks and communication services.

  In view of the above-mentioned problems, the present invention has a degree of probability of occurrence from the traffic amount that frequently occurs under normal conditions to the maximum level of traffic that rarely occurs in the daily fluctuation range of the traffic amount predicted in the future. It also aims to output multiple different traffic predictions.

In order to achieve the above object, the present invention is a communication traffic prediction apparatus for a communication network and its facilities provided by multiplexing a plurality of communication services,
The plurality of communication services are classified into a plurality of predetermined groups, and the plurality of communication services stored in the traffic database are multiplexed according to a traffic amount calculation processing method predetermined for each of the classified groups. Traffic metering means for calculating the traffic volume of each communication service from the traffic data;
Based on the traffic amount of each communication service calculated by the traffic distribution means, traffic prediction for predicting the traffic amount at a future time point of each group according to the traffic amount prediction processing method predetermined for each classified group. Means,
And a necessary equipment calculating means for integrating the traffic volume predicted by the traffic prediction means at a future time point and calculating the equipment capacity for absorbing short-term fluctuations based on the integrated traffic volume. Is configured as a communication traffic prediction apparatus.

  The traffic prediction means performs a plurality of types of traffic predictions with different degrees of probability of occurrence in the prediction of the traffic volume at the future time point, and the necessary equipment calculation means each of the traffic volumes by the plurality of types of traffic predictions A plurality of types of equipment capacities with different degrees of occurrence probability may be calculated.

The plurality of classified groups include, for example, a multicast type, a distribution type, an AC type, or a connection type related to other network connection to the communication network,
In the traffic amount calculation processing method in the multicast type, the traffic amount of the multicast transmission source is calculated as the traffic amount of the target equipment of each communication service,
In the traffic type calculation processing method in the distribution type, the traffic amount branched to the target facility from the traffic amount of the distribution source is calculated as the traffic amount of the target facility of each communication service,
In the traffic amount calculation processing method in the AC type, the traffic amount that summarizes the traffic amount of the communication service classified as the AC type is calculated as the traffic amount of the target facility,
In the connection type traffic amount calculation processing method, the traffic amount of the target facility is calculated using the change amount of the traffic amount before and after the other network is connected.

Further, for example, in the traffic amount prediction processing method in the multicast type, the traffic amount is predicted on the assumption that there is no change in the traffic amount of each communication service,
In the traffic amount prediction processing method in the distribution type, a prediction is performed by calculating an amount obtained by branching a predicted amount of traffic of a distribution source of each communication service to a target facility,
In the traffic type prediction processing method in the AC type, the traffic amount is predicted by collectively collecting communication services classified into the AC type,
In the traffic type prediction processing method in the connection type, the traffic amount of the target facility is predicted based on the traffic fluctuation at the connection point to which another network is connected.

  The present invention can also be configured as a communication traffic prediction method and a computer executed by the communication traffic prediction apparatus as a program for causing each computer to function as each means in the communication traffic prediction apparatus.

  According to the present invention, by classifying a large number of communication services into a relatively small number of routine processing groups, and preparing a traffic processing meter and traffic prediction that are well-established for each classification in advance, The accuracy of the traffic prediction result can be improved, and at the same time, the required traffic data amount and calculation amount can be reduced.

  In addition, according to the present invention, the daily fluctuation range of the traffic volume predicted in the future can be performed with traffic predictions having different degrees of probability of occurrence from the traffic volume that frequently occurs under normal conditions to the highest level traffic that rarely occurs. Multiple outputs are possible. Therefore, the decision to develop a facility design and operation plan to provide a cheaper communication network by reducing the facility capacity as much as possible even if the service provider takes the risk of degradation of service quality to an acceptable level. However, it will be possible in the design and operation of communication network equipment.

It is a functional block diagram of the communication traffic prediction apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart explaining operation | movement of the communication traffic prediction apparatus which concerns on the 1st Embodiment of this invention. It is a function block diagram of the communication traffic prediction apparatus which concerns on the 2nd Embodiment of this invention. It is a flowchart explaining operation | movement of the communication traffic prediction apparatus which concerns on the 2nd Embodiment of this invention. It is a function block diagram of the communication traffic prediction apparatus which concerns on the 3rd Embodiment of this invention. It is a flowchart explaining operation | movement of the communication traffic prediction apparatus which concerns on the 3rd Embodiment of this invention. It is a figure which shows the example of the output result of communication traffic prediction.

  Embodiments of the present invention will be described below with reference to the drawings. Each embodiment described below is only an example, and embodiments to which the present invention is applied are not limited to the following embodiments.

[First embodiment]
First, a first embodiment of the present invention will be described below. FIG. 1 is a functional configuration diagram of a communication traffic prediction apparatus 10 according to the first embodiment. FIG. 2 is a flowchart showing the operation of the communication traffic prediction apparatus 10 according to the first embodiment.

  The communication traffic prediction apparatus 10 according to the first embodiment includes a number of communication networks that are managed and operated, including an IP telephone service, an Internet connection service, an IP multi-channel broadcasting service, and a video on demand service (VOD service). The target is a communication network that provides multiple communication services in the same packet class. It includes the case where a plurality of different communication service providers provide the communication service.

  As shown in FIG. 1, a communication traffic prediction apparatus 10 according to the first embodiment includes a traffic data acquisition unit 20, a traffic database unit 30, a communication service analysis / classification unit 40, a traffic subtotaling unit 50, and a traffic prediction for each communication service. Unit 60, necessary equipment calculation unit 70, and output unit 80.

  The communication service analysis / classification unit 40 includes an analysis unit 42 and a classification unit 43. The traffic meter 50 includes a first classification type meter 51, a second classification type meter 52, and a third classification meter 53. The communication service-specific traffic prediction unit 60 includes a first classification type prediction unit 61, a second classification type prediction unit 62, a third classification type prediction unit 63, a subscription number prediction unit 64, and a traffic prediction unit 65 per subscription. The necessary facility calculation unit 70 includes an integration unit 71 and a short time fluctuation estimation unit 72.

  The communication traffic prediction device 10 is connected to the communication network 1 and the network management system 2.

  In the present embodiment, it is assumed that the traffic data 4 including a standard MIB, flow data, packet capture, and other formats can be periodically acquired from each facility of the communication network 1. Further, it is assumed that the network configuration information 3 can be acquired from the management system 2 of the communication network 1 at any time. Further, it is assumed that the subscriber data 5 of the communication service can be acquired from the management system 2 of the communication network 1 at any time.

  Hereinafter, the operation of each functional unit of the communication traffic prediction apparatus 10 according to the first embodiment will be described along the procedure of the flowchart shown in FIG. The step numbers described below correspond to the step numbers shown in FIG.

  First, the operation of the traffic data acquisition unit 20 will be described.

  Step S100) The traffic data acquisition unit 20 periodically acquires various traffic data 4 from each facility of the communication network 1. In addition, the traffic data acquisition unit 20 acquires the network configuration information 3 from the network management system 2 when a change occurs in the connection status of the equipment configuring the communication network 1. Further, the traffic data acquisition unit 20 acquires subscriber number data for each communication service from the network management system 2 every day. The traffic data acquisition unit 20 may calculate an optimal acquisition schedule that satisfies requirements such as an acquisition source for various data to be acquired, an acquisition time and a cycle, and the like.

  The traffic database unit 30 accumulates the data acquired by the traffic data acquisition unit 20 over a long period of time.

  Next, the communication service analysis / classification unit 40 will be described.

  Step S110) First, various necessary data are read from the traffic database unit 30. Next, the provided communication service information 41 is read. The provided communication service information 41 includes list information of the provided communication services, and information for specifying the traffic and flow by each communication service from various data.

  Step S120) Next, the analysis unit 42 performs analysis prepared in advance for various data regarding each communication service from the information described in the provided communication service information 41, and outputs an analysis result. The analysis method includes statistical analysis processing for various types of traffic data, or analysis of the arrival and departure IP addresses of flow data and packet capture data.

  Step S130) Next, in the classification unit 43, the analysis result of the analysis unit 42 is input, and the communication service is classified into one of the standard processing classification groups prepared in advance. At this time, a determination condition for the analysis result performed by the analysis unit 42 can be set in advance.

  Steps S141 to S143) In the traffic meter 50, the standard processing prepared in advance for the classification group to which the communication service determined by the classification unit 43 belongs is performed on the communication service, and the traffic amount of the communication service is Outputs the result of dividing from the traffic volume.

  In the first embodiment, it is assumed that three classifications from the first classification type to the third classification type are set. Correspondingly, the traffic metering unit 50 includes the first classification type metering unit. 51, the 2nd classification type subtotaling part 52, and the 3rd classification type subtotaling part 53, and each performs the division corresponding to a classification type. In this embodiment, the communication service analysis / classification unit 40 is provided. However, the communication service analysis / classification unit 40 is not provided with the communication service analysis / classification unit 40 by including the function of the communication service analysis / classification unit 40 in the traffic distribution unit 50. It is good.

  Steps S151 to S153) Subsequently, in the traffic prediction unit 60 for each communication service, a standard process prepared in advance is performed on the classification group to which the communication service determined by the classification unit 43 belongs. Based on the traffic amount data of the communication service divided by the subtotaling unit 50, the result of the traffic prediction at the future time t ^ of the communication service is output.

  The traffic prediction unit 60 for each communication service corresponds to the three classifications from the first classification type to the third classification type, and includes a first classification type prediction unit 61, a second classification type prediction unit 62, and a third classification type prediction unit 63. And predicting corresponding to each classification type.

  In the traffic prediction of each communication service, the number of subscribers of the communication service and the traffic amount per subscriber are divided into the number of subscribers in the number-of-subscriptions prediction unit 64 and the number of subscribers per subscription is estimated in the traffic prediction unit 65 per subscription. Estimate the traffic volume. It should be noted that the number of subscribers can be obtained by including the function of the subscription number prediction unit 64 and the function of the traffic prediction unit 65 in each of the first classification type prediction unit 61, the second classification type prediction unit 62, and the third classification type prediction unit 63. The prediction unit 64 and the traffic prediction unit 65 may not be provided.

  Step S158) The subscription number prediction unit 64 predicts the optimum subscription number of the communication service from the tendency of the monthly increase in the subscription number of each communication service belonging to the first classification type to the third classification type. In the case of the communication service that is about to spread as a new innovation, prediction based on the theory of S-shaped growth curve is performed. The prediction based on the theory of the S-shaped growth curve means that when the time axis is logarithmically converted, the increasing process of the number A is divided into three periods, the first period is predicted by a convex function downward, The period is predicted with a straight line, and the third period is predicted with an upward convex function. The detection of the time point that divides the first and second periods is the time when the net increase in a single month starts to decrease, and the prediction of the first and second periods is more accurate than the prediction based on the conventional regression line. Can be predicted.

  Step S159) The traffic prediction unit 65 per subscription predicts the optimum traffic amount per subscription of the communication service from the fluctuation trend of the traffic amount per subscription of the communication service belonging to the first classification type to the third classification type.

  Next, the operation of the necessary equipment calculation unit 70 will be described. The necessary facility calculation unit 70 includes an integration unit 71 and a short time fluctuation estimation unit 72.

  Step S160) First, the integrating unit 71 calculates the traffic amount obtained by integrating the traffic prediction of the future time t ^ of each communication service calculated by the service-specific traffic prediction unit 60 for all the communication services provided. To do.

  Step S170) Next, in the short-time fluctuation estimation unit 72, in order to ensure the service quality required for each communication service, which is calculated by the service-specific traffic prediction unit 60, the short-time fluctuation of the packet that is necessary is calculated. Estimate the capacity to absorb. The short-time variation estimation unit 72 can output a calculation result based on performance data of each device or actual data obtained by a load verification experiment, or an estimation result created by computer simulation or a mathematical model.

  Step S180) Finally, in the output unit 80, the result of the necessary facility calculation unit 70 is input, and the facility amount necessary for providing all the communication services provided at the future time t ^ is output.

  The communication traffic prediction apparatus 10 according to the present embodiment can be realized by causing one or a plurality of computers to execute a program describing the processing contents described in this specification. That is, the function of each unit of the communication traffic prediction apparatus 10 is a program corresponding to processing executed by each unit of the communication traffic prediction apparatus 10 using hardware resources such as a CPU, a memory, and a hard disk built in the computer. It is possible to implement by executing. Further, the program can be recorded on a computer-readable recording medium (portable memory or the like), stored, or distributed. It is also possible to provide the program through a network such as the Internet or electronic mail.

  The traffic data acquisition unit 20 and the traffic database unit 30 are provided in a device different from the communication traffic prediction device 10, and the communication traffic prediction device 10 accesses the device to acquire various data and perform traffic prediction. It is good as well.

  As described above, the point that the present invention can be implemented by the computer and the program is the same in the other embodiments.

[Second Embodiment]
Next, a second embodiment of the present invention will be described below. FIG. 3 is a functional configuration diagram of the communication traffic prediction apparatus 10 according to the second embodiment. FIG. 4 is a flowchart showing the operation of the communication traffic prediction apparatus 10 according to the second embodiment.

  The communication traffic prediction apparatus 10 according to the second embodiment uses the same IP telephone service, Internet connection service, IP multi-channel broadcasting service, and video on demand service (VOD service) in the managed and operated communication network. Targets communication networks that provide multiple packet classes.

  The IP multi-channel broadcasting service is assumed to be realized using multicast. A plurality of communication service providers may provide an IP telephone service, an Internet connection service, an IP multi-channel broadcasting service, and a VOD service. However, in the second embodiment, the IP multi-channel broadcasting service includes P1 and P2. The VOD service is provided by the V1 and V2 companies. When referring only to the IP multi-channel broadcasting service provided by the P1 company, the IP multi-channel broadcasting service (P1) is described.

  As shown in FIG. 3, the communication traffic prediction apparatus 10 according to the second embodiment includes a traffic data acquisition unit 20, a traffic database unit 30, a traffic subtotal unit 50, a traffic prediction unit 60 for each communication service, and a necessary equipment calculation unit. 70 and an output unit 80. The traffic meter 50 includes an MC type meter 54, a distribution meter 55, and an AC meter 56. The traffic prediction unit 60 for each communication service includes an MC type prediction unit 66, a distribution type prediction unit 67, and an AC type prediction unit 68. The necessary equipment calculation unit 70 includes an integration unit 71 and a short-time fluctuation estimation unit 72. The communication traffic prediction device 10 is connected to the communication network 1 and the network management system 2.

  Regarding the traffic data that can be acquired, in the present embodiment, it is assumed that the traffic data 4 based on the standard MIB is measured once an hour in each facility of the communication network 1. Here, since it is assumed that multiple packets are provided in the same packet class, the total value of traffic of all communication services is measured in the standard MIB.

  It is assumed that the network configuration information 3 can be obtained from the communication network management system 2 at any time. Further, it is assumed that the communication service subscriber number data 5 can be acquired from the communication network management system 2 at any time.

  Hereinafter, the operation of each functional unit of the communication traffic prediction apparatus 10 according to the second embodiment will be described along the procedure of the flowchart shown in FIG. The step numbers described below correspond to the step numbers shown in FIG.

  First, the operation of the traffic data acquisition unit 20 will be described.

  Step S200) The traffic data acquisition unit 20 acquires only the traffic data of the daily busy hour with respect to the standard MIB acquired every hour in each facility of the communication network 1 in order to reduce the data capacity and the calculation amount. In addition, the traffic data acquisition unit 20 acquires the network configuration information 3 from the network management system 2 when a change occurs in the connection status of the equipment configuring the communication network 1. Further, the traffic data acquisition unit 20 acquires subscriber number data for each communication service from the network management system 2 every day. The traffic data acquisition unit 20 may calculate an optimal acquisition schedule that satisfies requirements such as an acquisition source for various data to be acquired, an acquisition time and a cycle, and the like.

  In the second embodiment, the traffic data of each facility is acquired only in one day of the busiest time of day, so the traffic data unless otherwise specified is the traffic data of the busiest time of the facility. And

  The traffic database unit 30 accumulates the data acquired by the traffic data acquisition unit 20.

  Step S210) In the second embodiment, the traffic meter 50 reads various data such as provided communication service information.

  The provided communication service information in the second embodiment includes (1) a list of routine processing groups, (2) a list of provided communication services, and (3) a routine processing group to which the communication service belongs. Correspondence relationship is described.

Specifically, in the second embodiment, the following information is described.
(1) As a list of routine processing groups, three types of MC type, distribution type, and AC type are described.
(2) As a list of provided communication services, IP telephone service, Internet connection service, IP multi-channel broadcast service (P1), IP multi-channel broadcast service (P2), VOD service (V1), VOD service (V2) 6 are described.
(3) The correspondence relationship with the fixed processing group to which the communication service belongs is that the MC type IP multi-channel broadcast service (P1), IP multi-channel broadcast service (P2), the distribution type VOD service (V1), VOD service ( V1), the IP phone service and the Internet connection service are described in the AC type.

  Step S230) Based on the provided communication service information, the provided communication service can be classified into the standard processing groups prepared in advance. That is, in this embodiment, it can be classified into three types: MC type, distribution type, and AC type.

  As shown in FIG. 3, in the second embodiment, each function unit in the traffic meter unit 50 is based on the provided information service information 41 read from the traffic database unit 30 and the setting information corresponding to the corresponding routine processing group. Then, processing for identifying and acquiring relevant traffic data is performed.

  The setting information corresponding to the standard processing group includes MC type setting information 44 and distribution type setting information 45.

  The MC type setting information 44 specifies, for each communication service classified as MC type, information specifying a multicast transmission source and a multicast distribution tree as information specifying traffic data corresponding to the communication service. Information to be described.

  More specifically, the MC type setting information 44 in the second embodiment specifies the multicast transmission source equipment of the P1 company as information for specifying traffic data corresponding to the IP multi-channel broadcasting service (P1). Information specifying the multicast source tree and information specifying the multicast distribution tree and information specifying the traffic data corresponding to the IP multi-channel broadcasting service (P2) Information to be described.

  The distribution type setting information 45 includes, for each communication service classified as a distribution type, information specifying the VOD service source equipment as information specifying traffic data corresponding to the communication service, and all VOD services. Information for specifying a facility in a communication network through which a VOD service is provided to a service subscriber and information on the number of VOD service subscribers who receive the VOD service through the facility are described.

  More specifically, in the second embodiment, in the distribution type setting information 45, the equipment of the source of the VOD service of the V1 company is specified as information specifying the traffic data corresponding to the VOD service (V1). Information, information identifying the equipment in the communication network through which the VOD service (V1) is provided to all VOD service (V1) subscribers, and the VOD that receives the VOD service (V1) through the equipment Information on the number of service (V1) subscribers, information on identifying traffic data corresponding to the VOD service (V2), information identifying the source equipment of the VOD service of the V1 company, and all VOD services (V2) Information identifying the equipment in the communication network through which the VOD service (V2) is provided to other subscribers And VOD service (V2) subscribers information undergoing VOD service (V2) via, are described.

  Next, the operation of the traffic meter 50 will be described in detail.

  Step S241) First, the operation of the MC type meter 54 will be described. Since the traffic volume of the communication service classified as the MC type has a feature that matches the traffic volume of a specific facility on the communication network, the MC type metering unit 54 performs traffic division using this. The IP multi-channel broadcasting service is classified as MC type.

  In the second embodiment, the traffic volume of the IP multi-channel broadcasting service classified as MC type is the traffic volume measured at the multicast transmission source realizing the IP multi-channel broadcasting service. it can.

  That is, the MC type metering unit 54 reads the MC type setting information, specifies information for specifying the multicast transmission source realizing the IP multi-channel broadcasting service, and leads from the transmission source to all multicast distribution destination facilities. Using the setting information of the multicast distribution tree as an input, the traffic volume of the IP multi-channel broadcasting service on all facilities on the communication network 1 is obtained.

Here, the traffic volume of the IP multi-channel broadcasting service (P1) by the company P1 is defined as follows:
F_x_p1 (t1): the traffic volume of the IP multi-channel broadcasting service (P1) of the traffic prediction target equipment at the time point t1;
S_x_p1 (t1): Traffic volume of the source of the IP multi-channel broadcasting service (P1) at time t1,
The MC type metering unit 54 can calculate F_x_p1 (t1) by the following relational expression.

F_x_p1 (t1) = S_x_p1 (t1)
However, if the equipment does not belong to the multicast distribution tree, F_x_p1 (t1) = 0.

Similarly, the traffic volume of the IP multi-channel broadcasting service (P2) by company P2 is defined as follows:
F_x_p2 (t1): the traffic volume of the IP multi-channel broadcasting service (P2) of the traffic prediction target equipment at the time point t1;
S_x_p2 (t1): Traffic volume of the source of the IP multi-channel broadcasting service (P2) at time t1,
The MC type metering unit 54 can calculate F_x_p2 (t1) by the following relational expression.

F_x_p2 (t1) = S_x_p2 (t1)
However, if the equipment does not belong to the multicast distribution tree, F_x_p2 (t1) = 0.

  In general, even if there are n or more companies that provide IP multi-channel broadcasting service, the same applies, and the total traffic of n companies is defined as the total amount of MC traffic.

  That is, if F_x (t1) is the total MC traffic of the traffic prediction target equipment at time t1, the total MC traffic can be calculated by the following relational expression.

F_x (t1) = F_x_p1 (t1) + F_x_p2 (t1) + ... + F_x_pn (t1)
= S_x_p1 (t1) + S_x_p2 (t1) + ... + S_x_pn (t1)
Step S242) Next, the operation of the distribution type metering unit 55 will be described. Since the traffic volume of the communication service classified as the distribution type has a characteristic that the traffic volume of a specific facility on the communication network is branched, the distribution-type subtractor 55 uses the traffic distribution meter using the traffic volume. I do.

  In the second embodiment, for a VOD service classified as a distribution type, the traffic volume of the VOD service on all facilities on the communication network can be obtained by the following procedure.

  The distribution-type subtracting unit 55 reads the distribution-type setting information, specifies the traffic data corresponding to the VOD service, specifies the VOD service source equipment, and sends VOD to all VOD service subscribers. Information on the facilities in the communication network through which the service is provided and information on the number of VOD service subscribers who receive the VOD service through the facilities are input, and the information on all facilities on the communication network Obtain the traffic volume of the VOD service.

  Here, the traffic volume of the VOD service (V1) by the company V1 is defined as follows.

F_y_v1 (t1): Traffic volume of the VOD service (V1) of the equipment subject to traffic prediction at time t1;
S_y_v1 (t1): Traffic volume of the source of the VOD service (V1) at time t1;
A_y_v1 (t1): Total number of subscribers of VOD service (V1) at time t1;
B_y_v1 (t1): Number of subscribers who receive VOD service (V1) via the equipment at time t1.
At this time, the distribution-type subtractor 55 calculates F_y_v1 (t1) according to the following relational expression.

F_y_v1 (t1) = S_y_v1 (t1) * B_y_v1 (t1) / A_y_v1 (t1)
However, if the equipment does not go through the VOD service, F_y_v1 (t1) = 0.

  Similarly, the traffic volume of the VOD service (V2) by company V2 is defined as follows.

F_y_v2 (t1): Traffic volume of the VOD service (V2) of the equipment subject to traffic prediction at time t1;
S_y_v2 (t1): Traffic volume of the source of the VOD service (V2) at time t1;
A_y_v2 (t1): Total number of subscribers of the VOD service (V2) at time t1;
B_y_v2 (t1): Number of subscribers who receive VOD service (V2) via the equipment at time t1.
At this time, the distribution-type subtractor 55 calculates F_y_v2 (t1) according to the following relational expression.

F_y_v2 (t1) = S_y_v2 (t1) * B_y_v2 (t1) / A_y_v2 (t1).
However, if the equipment does not go through the VOD service, F_y_v2 (t1) = 0.

  In general, even when there are n or more companies that provide the VOD service, the same applies, and the total traffic of the n companies is defined as the total amount of distribution-type traffic.

In other words, if F_y (t1) is the total amount of distribution-type traffic of the traffic prediction target equipment at time t1, F_y (t1) can be calculated by the following relational expression.
F_y (t1)
= F_y_v1 (t1) + F_y_v2 (t1) + ... + F_y_vn (t1)
= S_y_v1 (t1) * B_y_v1 (t1) / A_y_v1 (t1) + S_y_v2 (t1) * B_y_v2 (t1) / A_y_v2 (t1) + ...
+ S_y_vn (t1) * B_y_vn (t1) / A_y_vn (t1).
Step S243) Subsequently, the operation of the AC type metering unit 56 will be described. The traffic volume of communication services classified as an AC type flows in a complicated manner on the communication network, and it is difficult to find the relationship with the traffic volume of a specific facility. There is a feature that the unevenness of traffic characteristics is reduced by multiplexing various applications by users. In order to further strengthen this feature, it is effective to perform a traffic distribution summarizing communication services classified into the AC type, and the processing amount for the traffic distribution can be reduced.

  Based on such a viewpoint, in the second embodiment, for the Internet connection service and the IP telephone service classified as the AC type, the total is defined as the AC type traffic total, and the total is performed.

  Here, it is defined as follows.

F_z (t1): Total amount of AC-type traffic of the equipment subject to traffic prediction at time t1;
F (t1): Traffic volume of the equipment at time t1.
At this time, the AC-type metering unit 56 calculates F_z (t1) by the following relational expression.

F_z (t1) = F (t1)-F_x (t1) -F_y (t1)
Next, the operation of the communication service-specific traffic prediction unit 60 will be described in detail.

  Step S251) First, the operation of the MC type prediction unit 66 will be described. Since the traffic volume of the communication service classified as MC type has a feature that matches the traffic volume of a specific facility on the communication network, the traffic prediction using this is originally performed. IP multi-channel is classified as MC type.

  In the IP multi-channel broadcasting service classified as MC type in the second embodiment, the number of distributed programs hardly changes and is almost constant, so the corresponding multicast traffic amount is also almost constant.

  Therefore, the MC type prediction unit 66 sets a constant value for traffic prediction of communication services classified into the MC type. That is, the traffic prediction of the IP multi-channel broadcast service (P1) and IP multi-channel broadcast service (P2) classified as MC type is also set to a constant value.

  Here, it is defined as follows.

F ^ _x_p1 (t ^): Traffic volume of the IP multi-channel broadcasting service (P1) of the equipment subject to traffic prediction at the future time t ^;
F_x_p1 (t1): Traffic volume of the IP multi-channel broadcasting service (P1) of the equipment at the latest time point t1.
At this time, the MC type prediction unit 66 determines F ^ _x_p1 (t ^) according to the following relational expression.

F ^ _x_p1 (t ^) = F_x_p1 (t1)
Similarly, if you define
F ^ _x_p2 (t ^): Traffic volume of the IP multi-channel broadcasting service (P2) of the equipment subject to traffic prediction at the future time t ^;
F_x_p2 (t1): Traffic volume of the IP multi-channel broadcasting service (P2) of the equipment at the latest time t1,
The MC type prediction unit 66 determines F ^ _x_p2 (t ^) according to the following relational expression.

F ^ _x_p2 (t ^) = F_x_p2 (t1)
In general, even when there are n or more companies that provide IP multi-channel broadcasting services, the same is true, and the total amount of MC traffic, which is the total traffic of n companies, can be predicted.

  In other words, if F ^ _x (t ^) is the total amount of MC traffic of the traffic prediction target equipment at the future time t ^, F ^ _x (t ^) can be calculated by the following relational expression.

F ^ _x (t ^) = F ^ _x_p1 (t ^) + F ^ _x_p2 (t ^) + ... + F ^ _x_pn (t ^)
Step S252) Next, the operation of the distribution type prediction unit 67 will be described. Since the traffic volume of the communication service classified as the distribution type has a characteristic that the traffic volume of a specific facility on the communication network is branched, the traffic prediction using this is performed.

  In the second embodiment, for the VOD service classified as the distribution type, the traffic prediction of the VOD service on all the facilities on the communication network can be performed by the following procedure.

  That is, the distribution-type prediction unit 67 performs prediction on the traffic of the VOD service source, and according to the ratio of the number of subscribers who use the VOD service via the traffic prediction target equipment to the VOD service subscribers. As a result of branching, the traffic volume of the equipment is predicted.

  Here, the traffic volume of the VOD service (V1) by the company V1 is defined as follows.

F ^ _y_max_v1 (t ^): Maximum value of the traffic volume of the future VOD service (V1) of the equipment subject to traffic prediction at the future time t ^;
A ^ _y_v1 (t ^): Number of VOD service (V1) subscribers at a future time t ^;
B ^ _y_v1 (t ^): Number of subscribers who receive VOD service (V1) via the equipment to be predicted at a future time t ^;
V_y_v1_max: Maximum value of traffic volume per subscriber from the past to the present in the VOD service (V1) distribution source equipment.

  At this time, the delivery type prediction unit 67 determines F ^ _y_max_v1 (t ^) by the following equation.

F ^ _y_v1_max (t ^) = V_v1_max * B ^ _y_v1 (t ^) / A ^ _y_v1 (t ^)
However, here, the VOD service is considered as a new innovation service, and the number of subscribers is predicted based on the theory of the S-shaped growth curve.

  That is, A ^ _y_v1 (t ^) represents the number of subscribers at the end of every month in the last 12 months of the VOD service (V1) as a function based on the theory of the S-shaped growth curve. In addition, B ^ _y_v1 (t ^) is a function based on the theory of S-shaped growth curve for the number of subscribers at the end of every month in the last 12 months receiving the VOD service (V1) via the equipment to be predicted.

  Similarly, the definition is as follows.

F ^ _y_v1_ave (t ^): Frequent value of the traffic volume of the VOD service (V1) of the equipment subject to traffic prediction at the future time t ^;
V_y_v1_ave: Average traffic volume per subscriber for the previous month in the VOD service (V1) distribution source equipment.
At this time, the delivery type prediction unit 67 determines F ^ _y_v1_ave (t ^) as follows.

F ^ _y_v1_ave (t ^) = V_y_v1_ave * B ^ _y_v1 (t ^) / A ^ _y_v1 (t ^)
By performing each calculation process as described above, the traffic volume of the VOD service (V1) for the forecasted equipment at the future time t ^ is the traffic volume that frequently appears under normal conditions F ^ _y_v1_ave (t ^) And the largest class of traffic that occurs rarely can be output as F ^ _y_v1_max (t ^).

  In general, even when there are n or more companies that provide the VOD service, the same is true, and the total amount of distribution-type traffic that is the total traffic of the n companies can be predicted.

In other words, if you define
F ^ _y_max (t ^): Maximum value of the total amount of distributed traffic of the equipment to be predicted for traffic at the future time t ^;
F ^ _y_ave (t ^): Frequent value of the total amount of distribution-type traffic for the equipment to be predicted for traffic at the future time t ^,
F ^ _y_max (t ^) and F ^ _y_ave (t ^) can be calculated as follows.

F ^ _y_max （t ^） ＝ V_v1_max ＊ B ^ _y_v1 (t ^) ／ A ^ _y_v1 (t ^) ＋ V_v2_max ＊ B ^ _y_v2 (t ^) ／ A ^ _y_v2 (t ^) ＋ ・ ・ ・ ＋ V_vn_max ＊ B ^ _y_vn (t ^) / A ^ _y_vn (t ^),
F ^ _y_ave (t ^) = V_y_v1_ave * B ^ _y_v1 (t ^) / A ^ _y_v1 (t ^) + V_y_v2_ave * B ^ _y_v2 (t ^) / A ^ _y_v2 (t ^) + ... + V_y_vn_ave * B ^ _y_vn (t ^) / A ^ _y_vn (t ^).

  As described above, the person in charge of traffic prediction and equipment design has the merit of being able to examine the work of equipment design / operation planning from both the risk of service quality and the equipment cost.

  Step S253) Further, the operation of the AC type prediction unit 68 will be described. The traffic volume of communication services classified as AC type flows in a complicated manner on the communication network, and it is difficult to find the relationship with the traffic volume of a specific facility, but very much like the Internet connection service. There is a feature that the unevenness of traffic characteristics is reduced by multiplexing various applications by users. In order to further strengthen this feature, it is effective to predict the traffic that summarizes the communication services classified into the AC type, and the processing amount for the traffic meter can be reduced.

  Therefore, in the second embodiment, for the Internet connection service and the IP telephone service classified as the AC type, a prediction is made with respect to the total AC type traffic that is the total.

  Here, F ^ _z (t ^) is defined as the total amount of AC traffic of the equipment to be predicted for traffic at the future time t ^.

  The number of subscribers in the total amount of AC traffic is determined as the larger value of the number of subscribers of the Internet connection service and the IP telephone service.

  Here, it is defined as follows.

U_z (t1): Number of subscribers of total AC traffic at time t1;
A_z_q1 (t1): Number of Internet connection service subscribers at time t1;
A_z_q2 (t1): Number of IP telephone service subscribers at time t1.

  Then, U_z (t1) can be determined as follows.

U_z (t1) = MAX {A_z_q1 (t1), A_z_q1 (t1)}
The total amount of AC traffic can be predicted as follows.

  First, the definition is as follows.

F ^ _z_max (t ^): Maximum value of the total amount of AC traffic for the equipment to be predicted for traffic at the future time t ^;
U ^ _z (t ^): Number of subscribers of total traffic at the future time t ^;
V ^ _z_max (t ^): Maximum predicted monthly amount of traffic per subscriber of total AC traffic at a future time t ^.

  At this time, F ^ _z_max (t ^) is calculated as follows.

F ^ _z_max (t ^) = U ^ _z (t ^) * V ^ _z_max (t ^)
However, U ^ _z (t ^) is a function obtained by extrapolating the end-of-month data on the number of subscribers of the total amount of AC traffic over the last 12 months using the least square method. Further, V ^ _z_max (t ^) is a function obtained by extrapolating the maximum monthly value in the data obtained by calculating the traffic per subscriber of the AC traffic total to the last 12 points by the least square method.

  The definition is as follows.

F ^ _z_ave (t ^): Frequent value of the total amount of AC-type traffic for the equipment to be predicted for traffic at the future time t ^;
V ^ _z_ave (t ^): The predicted monthly average value of the traffic volume per subscriber of the total traffic volume at the future time t ^.

  At this time, F ^ _z_ave (t ^) can be calculated as follows.

F ^ _z_ave (t ^) = U ^ _z (t ^) * V ^ _z_ave (t ^)
However, V ^ _z_ave (t ^) is a function obtained by extrapolating the monthly average value in the data obtained by calculating the traffic per subscriber of the AC traffic total amount by the least square method for the latest 12 points.

  As described above, at the future time t ^, the traffic forecast target equipment F ^ _z_ave (t ^), which is frequently predicted under normal conditions, and the largest traffic volume that rarely occurs By outputting two values of a certain F ^ _z_max (t ^), the person in charge of traffic prediction and equipment design has the merit of being able to examine the business from both aspects of service quality risk and equipment cost.

  Next, the operation of the necessary equipment calculation unit 70 will be described in detail.

  Step S260) First, in the integrating unit 71, the traffic amount calculated by the communication service-specific traffic predicting unit 60 at the future time t ^ of each communication service is integrated for all provided communication services. Is calculated.

  Step S270) Next, in the short-term fluctuation estimation unit 72, the traffic data measurement is performed in order to ensure the service quality required for each communication service by using the result calculated by the communication service-specific traffic prediction unit 60 as an input. Estimate the capacity of the equipment that absorbs short-term fluctuations in packets required for the period.

  In the second embodiment, F ^ _x (t ^) + F ^ _y_ave (t ^) + F ^ is used to predict the traffic volume that frequently appears under normal conditions for the equipment to be traffic predicted at a future time t ^. _z_ave (t ^) and F ^ _x (t ^) + F ^ _y_max (t ^) + F ^ _z_max (t ^) as inputs to predict the maximum amount of traffic that rarely occurs. In order to ensure the required service quality, the capacity of the future time t ^ is calculated in order to absorb the short-term fluctuations of the packets required for the traffic data measurement period.

  The short-time fluctuation estimation unit 72 can output a calculation result based on performance data of each device or performance data obtained by a load verification experiment, or an estimation result created by computer simulation or a mathematical model. However, in the second embodiment, the equipment necessary for ensuring the quality of service required for all communication services with respect to the traffic volume based on the traffic data measurement period, based on the load verification experiment of each device. Assume that the quantity is known to be d times (constant).

At this time, the short-term fluctuation estimation unit 72 sets d * (F ^ _x (t ^) as a facility quantity necessary for the traffic volume that frequently appears under normal conditions for the facility to be predicted for traffic at a future time t ^. + F ^ _y_ave (t ^) + F ^ _z_ave (t ^)), d * (F ^ _x (t ^) + F ^ _y_max (t ^) + F as the amount of equipment necessary for the maximum amount of traffic that occurs rarely ^ _z_max (t ^))
Are output respectively.

  Step S280) Finally, in the output unit 80, the processing result of the necessary facility calculation unit 70 is input, and the facility amount necessary for providing all the communication services provided at the future time t ^ is output. The

  D * (F ^ _x (t ^) + F ^ _y_ave (t ^) + F ^ _z_ave (t ^)), which is the prediction of the traffic volume that frequently occurs under normal conditions at the future time t ^ And traffic prediction by outputting two values of d * (F ^ _x (t ^) + F ^ _y_max (t ^) + F ^ _z_max (t ^)), which is the largest amount of traffic that occurs rarely And the person in charge of equipment design has the merit of being able to consider the business from both aspects of service quality risk and equipment cost.

[Third embodiment]
A third embodiment of the present invention will be described below. FIG. 5 is a functional configuration diagram of the communication traffic prediction apparatus 10 according to the third embodiment. FIG. 6 is a flowchart showing the operation of the communication traffic prediction apparatus 10 according to the third embodiment.

  The communication traffic prediction apparatus 10 according to the third embodiment includes other network connection services as communication services in addition to an IP telephone service, an Internet connection service, an IP multi-channel broadcasting service, and a video on demand service (VOD service). The target is a communication network provided by multiplexing in the same packet class.

  The IP multi-channel broadcasting service is assumed to be realized using multicast. A plurality of communication service providers may provide an IP telephone service, an Internet connection service, an IP multi-channel broadcast service, and a VOD service. In the third embodiment, the IP multi-channel broadcast service includes P1 and P2. The VOD service is provided by the V1 and V2 companies. When referring only to the IP multi-channel broadcasting service provided by the P1 company, the IP multi-channel broadcasting service (P1) is described.

  The difference between the third embodiment and the second embodiment is that there is a communication network (N1) operated by a third party in the third embodiment, and mutual communication from the connection point. Traffic flowing through the network is flowing in and out. Communication services flowing in and out from the connection point are collectively defined as another network connection service (N1).

  Here, in the present embodiment, the time t0 when the communication network (N1) is connected is known, and the traffic volume of each facility before and after that time is stored in the traffic database. .

  As shown in FIG. 5, the communication traffic prediction apparatus 10 according to the third embodiment includes a traffic data acquisition unit 20, a traffic database unit 30, a traffic subtotal unit 50, a traffic prediction unit 60 for each communication service, and a necessary equipment calculation unit. 70 and an output unit 80. The traffic meter 50 includes an MC type meter 54, a distribution meter 55, an AC meter 56, and a connected meter 57. The traffic prediction unit 60 for each communication service includes an MC type prediction unit 66, a distribution type prediction unit 67, an AC type prediction unit 68, and a connection type prediction unit. The necessary equipment calculation unit 70 includes an integration unit 71 and a short-time fluctuation estimation unit 72. The communication traffic prediction device 10 is connected to the communication network 1 and the network management system 2. Further, as described above, the communication network (N1) 8 is connected to the communication network 1 via the connection point 9.

  With regard to the traffic data that can be acquired, in this embodiment, it is assumed that the traffic data is measured once per hour according to the standard MIB of the communication network equipment. Here, it is assumed that the total value of traffic data of multiplexed communication services is measured for all communication network facilities. Further, it is assumed that the information on the number of subscribers of the communication service can be periodically acquired from the management system 2 of the communication network.

  Hereinafter, the operation of each functional unit of the communication traffic prediction apparatus 10 according to the third embodiment will be described along the procedure of the flowchart shown in FIG. The step numbers described below correspond to the step numbers shown in FIG.

  Step S300) The traffic data acquisition unit 20 acquires only the traffic data of the daily busy hour from the standard MIB acquired every hour in each facility of the communication network 1 in order to reduce the data capacity and the calculation amount. The subscriber count data for each communication service is acquired from the network management system 2 only at the end of every month. The network configuration information 3 is acquired when a change occurs in the network configuration. The traffic data acquisition unit 20 may calculate an optimal acquisition schedule that satisfies conditions such as an acquisition source, an acquisition time, and a period for various types of data to be acquired.

  The difference from the second embodiment is that traffic data of the other network connection service (N1) is acquired at the connection point 9 with the communication network (N1).

  In the third embodiment, the traffic data of each facility is acquired only in one day of the busiest day, so the traffic data unless otherwise specified is the busiest traffic data of the facility. And

  The traffic database unit 30 accumulates the data acquired by the traffic data acquisition unit 20.

  Step S310) In the third embodiment, the traffic meter 50 reads various data such as provided communication service information.

  The provided communication service information in the third embodiment includes (1) a list of routine processing groups, (2) a list of provided communication services, and (3) a routine processing group to which the communication service belongs. Correspondence relationship is described.

Specifically, in the third embodiment, the following information is described.
(1) As a list of routine processing groups, four types of MC type, distribution type, connection type, and AC type are described.
(2) As a list of provided communication services, IP telephone service, Internet connection service, IP multi-channel broadcast service (P1), IP multi-channel broadcast service (P2), VOD service (V1), VOD service (V2) 6 are described.
(3) The correspondence relationship with the fixed processing group to which the communication service belongs is that the MC type IP multi-channel broadcast service (P1), IP multi-channel broadcast service (P2), the distribution type VOD service (V1), VOD service ( V1), the other network connection service (N1) as the connection type, and the IP telephone service and the Internet connection service as the AC type.

  Step S330) Based on the provided communication service information, the provided communication service can be classified into the standard processing groups prepared in advance. That is, in this embodiment, it can be classified into four types: MC type, distribution type, connection type, and AC type.

  As shown in FIG. 5, in the third embodiment, each functional unit in the traffic meter 50 is based on the provided information service information 41 read from the traffic database 30 and the setting information corresponding to the corresponding routine processing group. Then, processing for identifying and acquiring the corresponding traffic data is performed.

  The setting information corresponding to the standard processing group includes MC type setting information 44, distribution type setting information 45, and connection point information 46.

  The MC type setting information 44 specifies, for each communication service classified as MC type, information specifying a multicast transmission source and a multicast distribution tree as information specifying traffic data corresponding to the communication service. Information to be described.

  Since the MC type setting information 44 in the third embodiment is the same as that in the second embodiment, a detailed description thereof will be omitted.

  The distribution type setting information 45 includes, for each communication service classified as a distribution type, information specifying the VOD service source equipment as information specifying traffic data corresponding to the communication service, and all VOD services. Information for specifying a facility in a communication network through which a VOD service is provided to a service subscriber and information on the number of VOD service subscribers who receive the VOD service through the facility are described.

  Since the delivery type setting information 45 of the third embodiment is the same as that of the second embodiment, a detailed description thereof will be omitted.

  Next, the operation of the traffic meter 50 will be described in detail.

  Since the operation of the MC type metering unit 54 (step S341) and the operation of the distribution type metering unit 55 (step S342) are the same as those in the second embodiment, the description thereof is omitted.

  Step S343) The operation of the connection type metering unit 57 will be described. The traffic volume of the other network connection service (N1) classified as the connection type flows in the communication network in the same way as the AC type, and it is considered difficult to find the relationship with the traffic volume of a specific facility. The change in the traffic volume of each facility before and after the communication network (N1) is connected can be measured by the traffic data. Therefore, it is considered that the amount of change in the traffic of each facility before and after connection is equal to the total traffic amount of the communication service classified as the other network connection type service passing through each facility at the time of connection.

  The total traffic volume of the communication service classified as this other network connection type service is defined as the total amount of connection type traffic.

  Here, it is defined as follows.

F_w (t0): Total amount of connected traffic in the equipment subject to traffic prediction at connection time t0;
F (t0 +): Traffic volume at the traffic prediction target equipment at time t0 + immediately after connection time t0;
F (t0-): Traffic amount in the traffic prediction target equipment at time t0- immediately before connection time t0.

  At this time, F_w (t0) is calculated as follows.

F_w (t0) = F (t0 +)-F (t0-)
Next, at the connection point, traffic data of the connection point from the connection time point t0 to the latest time point t1 is accumulated. Therefore, at the connection point, it is possible to calculate the increase rate of the connection type traffic from the connection time t0 to the latest time t1.

  It is defined as follows.

S_w (t0): Traffic volume of connection type traffic at connection point t0 at the connection point;
S_w (t1): the traffic volume of the connection type traffic at the most recent time t1 (t0 <t1) at the connection point;
R_w (t0, t1): The increase rate of the connection type traffic between the time point t1 and the connection time point t0 at the connection point.

  At this time, the following relationship holds.

R_w (t0, t1) = S_w (t1) / S_w (t0)
The increase ratio R_w (t0, t1) obtained above is originally a ratio limited to the connection point, but by assuming that it is equal to the increase ratio of the total amount of connected traffic applied to all devices on the communication network, The total amount of connection-type traffic at a certain past time t1 after the connection time t0 in the traffic prediction target equipment can be divided.

  That is, assuming that F_w (t1) is the total amount of connected traffic of the traffic prediction target equipment at time t1, the connected-type subtractor 57 calculates F_w (t1) by the following relational expression.

F_w (t1) = R_w (t0, t1) * F_w (t0)
Step S344) Subsequently, the operation of the AC type metering unit 56 will be described. The traffic volume of communication services classified as AC type flows in a complicated manner on the communication network, and it is difficult to find the relationship with the traffic volume of a specific facility, but very much like the Internet connection service. There is a feature that the unevenness of traffic characteristics is reduced by multiplexing various applications by users. In order to further strengthen this feature, it is effective to perform a traffic distribution summarizing communication services classified into the AC type, and the processing amount for the traffic distribution can be reduced.

  In the third embodiment, for the Internet connection service and the IP telephone service classified as the AC type, the total is defined as the AC type traffic total and is divided.

  Here, it is defined as follows.

F_z (t1): Total amount of AC-type traffic of the equipment subject to traffic prediction at time t1;
F (t1): Traffic volume of the equipment at time t1.

  At this time, F_z (t1) can be calculated by the following relational expression.

F_z (t1) = F (t1)-F_x (t1) -F_y (t1) -F_w (t1)
Next, the operation of the communication service-specific traffic prediction unit 60 will be described in detail.

  Since the operation of the MC type prediction unit 66 (step S351) and the operation of the distribution type prediction unit 67 (step S352) are the same as those in the second embodiment, description thereof is omitted.

  Step S353) The operation of the connection type prediction unit 69 will be described. The traffic volume of the other network connection service (N1) classified as the connection type flows in the communication network in the same way as the AC type, and it is considered difficult to find the relationship with the traffic volume of a specific facility. The change in traffic volume can be grasped only at the connection point with the communication network (N1). Therefore, it is considered that the traffic fluctuation in the prediction target facility is equal to the traffic fluctuation of the other network connection service (N1) at the connection point.

  In the third embodiment, the total traffic volume of the communication service classified as the other network connection service (N1) is predicted with respect to the total connection traffic volume.

  Here, F ^ _w (t ^) is defined as the total amount of connected traffic of the equipment to be predicted for traffic at the future time t ^, and S ^ w (t ^) is the connection type of the connection point at the future time t ^. Defined as total traffic.

  At this time, when it is assumed that the increasing ratio from the connection time t0 to the latest time t1 (t0 <t1) continues to be equal to the future time t ^ (t1 <t ^), the following relationship holds.

S ^ _w (t ^) = {R_w (t0, t1)} ^ {(t ^ -t0) / (t1-t0)} * S (t0)
Similarly, it is assumed that the following relationship holds.

F ^ _w (t ^) = {R_w (t0, t1)} ^ {(t ^ -t0) / (t1-t0)} * F (t0)
Step S354) Next, the operation of the AC type prediction unit 68 will be described. The traffic volume of communication services classified as AC type flows in a complicated manner on the communication network, and it is difficult to find the relationship with the traffic volume of a specific facility, but very much like the Internet connection service. There is a feature that the unevenness of traffic characteristics is reduced by multiplexing various applications by users. In order to further strengthen this feature, it is effective to predict the traffic that summarizes the communication services classified into the AC type, and the processing amount for the traffic meter can be reduced.

  In the third embodiment, as in the second embodiment, prediction is performed on the total amount of AC traffic, which is the total traffic of communication services classified as an AC type. F ^ _z (t ^) is defined as the total amount of AC traffic for the equipment to be predicted for traffic at the future time t ^. Since the detailed operation of the AC type prediction unit 68 is the same as that of the second embodiment, the description thereof is omitted.

  Next, the operation of the necessary equipment calculation unit 70 will be described in detail.

  Step S360) First, in the integrating unit 71, the traffic amount calculated by the communication service-specific traffic predicting unit 60 at the future time t ^ of each communication service is integrated for all provided communication services. Is calculated.

  Step S370) Next, in the short-term fluctuation estimation unit 72, the traffic data measurement is performed in order to secure the service quality required for each communication service by using the result calculated by the communication service-specific traffic prediction unit 60 as an input. Estimate the capacity of the equipment that absorbs short-term fluctuations in packets required for the period.

  In the third embodiment, F ^ _x (t ^) + F ^ _y_ave (t ^) + F ^ is used to predict the traffic volume that frequently appears under normal conditions for the equipment to be traffic predicted at a future time t ^. _z_ave (t ^) + F ^ _w (t ^), F ^ _x (t ^) + F ^ _y_max (t ^) + F ^ _z_max (t ^) + F ^ _w (t ^), as an input, a future point in time to absorb short-term packet fluctuations required for the traffic data measurement period to ensure the quality of service required for each communication service. Calculate the capacity of t ^.

  The short-time fluctuation estimation unit 72 can output a calculation result based on performance data of each device or performance data obtained by a load verification experiment, or an estimation result created by computer simulation or a mathematical model. However, in the third embodiment, the equipment necessary for ensuring the service quality required for all communication services with respect to the traffic volume based on the traffic data measurement period, through the load verification experiment of each device. Assume that the quantity is known to be d times (constant).

  At this time, the short-term fluctuation estimation unit 72 sets d * (F ^ _x (t ^) as a facility quantity necessary for the traffic volume that frequently appears under normal conditions for the facility to be predicted for traffic at a future time t ^. + F ^ _y_ave (t ^) + F ^ _z_ave (t ^) + F ^ _w (t ^)), d * (F ^ _x (t ^) + F as the amount of equipment necessary for the maximum amount of traffic that occurs rarely ^ _y_max (t ^) + F ^ _z_max (t ^) + F ^ _w (t ^)) is output respectively.

  Step S380) Finally, in the output unit 80, the processing result of the necessary facility calculation unit 70 is input, and the facility amount necessary for providing all the communication services provided at the future time t ^ is output. The

  D * (F ^ _x (t ^) + F ^ _y_ave (t ^) + F ^ _z_ave (t ^) + F, which is the prediction of the traffic volume that frequently occurs under normal conditions at the future time t ^ ^ _w (t ^)) and d * (F ^ _x (t ^) + F ^ _y_max (t ^) + F ^ _z_max (t ^) + F ^ _w (t ^) By outputting the two values of)), the person in charge of traffic prediction and equipment design has the merit of being able to examine the business from both aspects of service quality risk and equipment cost.

  FIG. 7 shows an output result example of communication traffic prediction by the communication traffic prediction apparatus 10 according to the embodiment of the present invention. In FIG. 7, the solid line indicates the predicted value of the traffic amount that frequently appears under normal conditions, and the dotted line indicates the predicted value of the maximum amount of traffic that rarely occurs. The actual traffic value is also shown.

[Summary of Embodiment, Effects]
In the embodiment of the present invention, in the communication traffic prediction apparatus 10 for a communication network and its facilities provided by multiplexing a plurality of communication services, the plurality of communication services are classified into a plurality of predetermined groups. The traffic amount for calculating the traffic volume of each communication service from the traffic data multiplexed with the plurality of communication services stored in the traffic database according to the traffic volume calculation processing method predetermined for each classified group. Based on the traffic amount of each communication service calculated by the traffic counting unit 50 and the traffic counting unit 50, the traffic at each time point of each group in accordance with the traffic volume prediction processing method predetermined for each classified group. Traffic prediction unit 60 for each communication service that predicts the amount Necessary equipment calculation section 70 for integrating the traffic volume at each future time of each group predicted by the traffic prediction section 60 for each communication service and calculating the equipment capacity for absorbing short-term fluctuations based on the integrated traffic volume. And to prepare.

  The traffic prediction unit for each communication service 60 performs a plurality of types of traffic predictions with different degrees of occurrence probability in the prediction of the traffic volume at the future time point, and the necessary equipment calculation means uses the plurality of types of traffic predictions. Integration is performed for each of the traffic amounts, and a plurality of types of equipment capacities having different degrees of occurrence probability are calculated.

  Further, in the present embodiment, the plurality of classified groups include a multicast type, a distribution type, an AC type, or a connection type related to other network connection to the communication network, and the traffic amount calculation processing in the multicast type In the method, the traffic volume of the multicast transmission source is calculated as the traffic volume of the target equipment of each communication service. In the distribution type traffic volume calculation processing method, the traffic volume of the target equipment of each communication service is calculated as the traffic volume of the target equipment of each communication service. The traffic amount branched to the target facility is calculated from the traffic amount, and the traffic amount calculation processing method in the AC type calculates the traffic amount that summarizes the traffic amount of the communication service classified as the AC type as the traffic amount of the target facility. In the traffic type calculation processing method in the connection type, The traffic volume of elephant equipment is calculated using the amount of change in the amount of traffic before and after the another network is connected.

  Further, in the traffic type prediction processing method in the multicast type, the traffic amount is predicted on the assumption that there is no change in the traffic amount of each communication service. In the traffic type prediction processing method in the distribution type, the distribution source of each communication service Predicting by calculating the amount of traffic predicted to be branched to the target equipment, and in the traffic type prediction processing method in the AC type, the communication services classified into the AC type are collectively estimated to predict the traffic amount, In the traffic type prediction processing method in the connection type, the traffic amount of the target facility is predicted based on the traffic fluctuation at the connection point to which another network is connected.

  That is, according to the embodiment of the present invention, a large number of communication services are classified into a relatively small number of groups, and a traffic meter and a process for traffic prediction are preliminarily processed and set for each classification. As a result, the accuracy of the result of communication traffic prediction can be improved, and at the same time, it is possible to reduce the required traffic data amount and calculation amount.

  Furthermore, with regard to the daily fluctuation range of the traffic volume predicted in the future, it is possible to output multiple traffic predictions with different degrees of probability of occurrence, from the traffic volume that occurs frequently under normal conditions to the highest level traffic that occurs rarely It is.

  As a result, even if the service provider accepts the risk of deterioration in service quality to the extent acceptable, the facility design and operation plan is established to provide a cheaper communication network by reducing the equipment capacity as much as possible. Judgment is possible in the design and operation of communication network equipment.

  In any of the above-mentioned patent documents and non-patent documents, in order to improve the accuracy of future prediction of communication traffic, the traffic in which a plurality of communication services are multiplexed is divided into the traffic of individual communication services. This technology consists of a technology for measuring traffic, a technology for predicting traffic that is individually adapted to the traffic of each individual communication service, and a technology for appropriately accumulating the amount of traffic applied to the target equipment for traffic prediction. There is nothing.

  As mentioned above, the invention made by the present inventor has been specifically described based on the above embodiment, but the present invention is not limited to the above embodiment, and various modifications and changes can be made without departing from the scope of the invention. Application is possible.

  For example, in the embodiment of the present invention, communication services are classified into a multicast type, a distribution type, an AC type, a connection type, and the like, but these are examples, for example, a multicast type, a distribution type, an AC type, and a connection type. It is also possible to classify into a group having any one or more of these types and other types. Alternatively, it is possible to classify into a group different from any of the multicast type, distribution type, AC type, and connection type. In addition, multiple traffic predictions with different degrees of occurrence probability are not limited to the prediction of traffic volume that occurs frequently under normal conditions and the prediction of the maximum amount of traffic that occurs infrequently, but includes predictions of other occurrence probabilities. It's okay.

DESCRIPTION OF SYMBOLS 1 ... Communication network 2 ... Network management system 3 ... Network configuration information 4 ... Traffic data 5 ... Communication service subscriber number data 6 ... Internet 7 ... Internet connection point 8 ..Communication network (N1)
DESCRIPTION OF SYMBOLS 9 ... Connection point 10 ... Communication traffic prediction apparatus 20 ... Traffic data acquisition part 30 ... Traffic database part 40 ... Communication service analysis classification | category part 41 ... Provided communication service information 42 ... Analysis unit 43... Classification unit 44... MC type setting information 45... Distribution type setting information 46... Connection type information 50. 52... Second classification type subtractor 53... Third classification type subtractor 54... MC type subtractor 55... Distribution type subtractor 56. ... Connection type subtotaling unit 60 ... Traffic service-specific traffic prediction unit 61 ... First classification type prediction unit 62 ... Second classification type prediction unit 63 ... Third classification type prediction unit 64 .. Subscription number prediction unit 65 ... Traffic prediction unit per subscription 66 ... MC type prediction unit 6 ... Distribution type prediction unit 68 ... AC type prediction unit 69 ... Connection type prediction unit 70 ... Necessary equipment calculation unit 71 ... Integration unit 72 ... Short time fluctuation estimation unit 80 ... Output section

Claims (9)

A communication traffic prediction device for a communication network and its facilities provided by multiplexing a plurality of communication services,
The plurality of communication services are classified into a plurality of predetermined groups, and the plurality of communication services stored in the traffic database are multiplexed according to a traffic amount calculation processing method predetermined for each of the classified groups. Traffic metering means for calculating the traffic volume of each communication service from the traffic data;
Based on the traffic amount of each communication service calculated by the traffic distribution means, traffic prediction for predicting the traffic amount at a future time point of each group according to the traffic amount prediction processing method predetermined for each classified group. Means,
And a necessary equipment calculation means for calculating the equipment capacity for absorbing short-term fluctuations based on the accumulated traffic volume estimated by the traffic prediction means at a future time point. A communication traffic prediction device. The traffic prediction means performs a plurality of types of traffic predictions with different degrees of probability of occurrence in the prediction of the traffic amount at the future time point,
2. The communication according to claim 1, wherein the necessary facility calculation unit performs integration for each of the traffic amounts based on the plurality of types of traffic prediction, and calculates a plurality of types of facility capacities having different degrees of occurrence probability. Traffic prediction device. The plurality of classified groups include a multicast type, a distribution type, an AC type, or a connection type related to another network connection to the communication network,
In the traffic amount calculation processing method in the multicast type, the traffic amount of the multicast transmission source is calculated as the traffic amount of the target equipment of each communication service,
In the traffic type calculation processing method in the distribution type, the traffic amount branched to the target facility from the traffic amount of the distribution source is calculated as the traffic amount of the target facility of each communication service,
In the traffic amount calculation processing method in the AC type, the traffic amount that summarizes the traffic amount of the communication service classified as the AC type is calculated as the traffic amount of the target facility,
3. The communication according to claim 1, wherein, in the connection type traffic amount calculation processing method, the traffic amount of the target equipment is calculated using a change amount of the traffic amount before and after the connection to another network. Traffic prediction device. The plurality of classified groups include a multicast type, a distribution type, an AC type, or a connection type related to another network connection to the communication network,
In the traffic amount prediction processing method in the multicast type, the traffic amount is predicted as if there is no change in the traffic amount of each communication service,
In the traffic amount prediction processing method in the distribution type, a prediction is performed by calculating an amount obtained by branching a predicted amount of traffic of a distribution source of each communication service to a target facility,
In the traffic type prediction processing method in the AC type, the traffic amount is predicted by collectively collecting communication services classified into the AC type,
The communication traffic according to claim 1 or 2, wherein the traffic type prediction processing method in the connection type predicts the traffic amount of the target equipment based on a traffic fluctuation at a connection point to which another network is connected. Prediction device. A communication traffic prediction method executed by a communication traffic prediction apparatus targeting a communication network and its facilities provided by multiplexing a plurality of communication services,
The plurality of communication services are classified into a plurality of predetermined groups, and the plurality of communication services stored in the traffic database are multiplexed according to a traffic amount calculation processing method predetermined for each of the classified groups. A traffic metering step for calculating the traffic volume of each communication service from the traffic data;
Based on the traffic amount of each communication service calculated in the traffic distribution step, the traffic prediction predicts the traffic amount at a future time point of each group according to the traffic amount prediction processing method predetermined for each classified group. Steps,
And a necessary facility calculation step of calculating a facility capacity that absorbs short-term fluctuations based on the accumulated traffic amount based on the accumulated traffic amount at a future time of each group predicted by the traffic prediction step. A communication traffic prediction method. In the prediction of the traffic amount at the future time in the traffic prediction step, a plurality of types of traffic predictions having different degrees of probability of occurrence are performed,
6. The communication according to claim 5, wherein, in the necessary facility calculating step, a plurality of types of facility capacities having different degrees of occurrence probability are calculated by performing integration for each of the traffic amounts based on the plurality of types of traffic prediction. Traffic prediction method. The plurality of classified groups include a multicast type, a distribution type, an AC type, or a connection type related to another network connection to the communication network,
In the traffic amount calculation processing method in the multicast type, the traffic amount of the multicast transmission source is calculated as the traffic amount of the target equipment of each communication service,
In the traffic type calculation processing method in the distribution type, the traffic amount branched to the target facility from the traffic amount of the distribution source is calculated as the traffic amount of the target facility of each communication service,
In the traffic amount calculation processing method in the AC type, the traffic amount that summarizes the traffic amount of the communication service classified as the AC type is calculated as the traffic amount of the target facility,
The communication according to claim 5 or 6, wherein in the connection type traffic amount calculation processing method, the traffic amount of the target facility is calculated using a change amount of the traffic amount before and after the other network is connected. Traffic prediction method. The plurality of classified groups include a multicast type, a distribution type, an AC type, or a connection type related to another network connection to the communication network,
In the traffic amount prediction processing method in the multicast type, the traffic amount is predicted as if there is no change in the traffic amount of each communication service,
In the traffic amount prediction processing method in the distribution type, a prediction is performed by calculating an amount obtained by branching a predicted amount of traffic of a distribution source of each communication service to a target facility,
In the traffic type prediction processing method in the AC type, the traffic amount is predicted by collectively collecting communication services classified into the AC type,
The communication traffic according to claim 5 or 6, wherein in the connection type traffic amount prediction processing method, the traffic amount of a target facility is predicted based on traffic fluctuation at a connection point to which another network is connected. Prediction method.   The program for functioning a computer as each means in the communication traffic prediction apparatus of any one of Claims 1 thru | or 4.

Images