JP4956381B2 - Traffic analysis apparatus, method, and program - Google Patents

Description

  The present invention relates to a communication technique, and more particularly to a technique for analyzing traffic generated from a communication line.

In general, when providing communication services to users, a line is provided for each user and a communication path assigned to each line is used, or multiple user lines are superimposed on one communication path. Are transferring through the network.
In such a network configuration, it is necessary to predict the traffic generated in the communication path at a future target time, calculate the communication band required for the communication path, and design the communication path and the network based on this communication band. is there.

Conventionally, as a method for predicting the communication band of such a communication path, a technique for predicting a communication band at a future target time point based on time-series traffic information observed for a target communication path in a predetermined observation period (for example, Non-Patent Documents 1 and 2) have been proposed.
In addition, by adding environmental information indicating changes in the population composition, household attributes, and business affiliation of users using the network to the time-series traffic information observed on the target communication path, Techniques for predicting the communication band in (see, for example, non-patent documents 3 and 4) have been proposed.

Hiromichi Kawano et al., “Traffic Prediction Method for Macro Analysis and its Evaluation”, IEICE Transactions, Vol.J82-B, No.6, pp1107-1114, June 1999 Toru Ueda et al., “Application of Time Series Analysis Methods in Telecommunications”, Operations Research, Vol.34, No.10, pp530-534, October 1989 Hoshigochiho et al., “Examination of multiple regression model for call volume”, 1992 IEICE Autumn Meeting, B-503, p3-169, 1992 Toru Ueda et al., "Analysis of communication usage accompanying statistical methods", NTT R & D, Vol. 40, No. 12, p1599-1606, 1991 By David E. McDysan, Darren L. Spohn, translated by Masayuki Murata, "ATM Network", Science and Technology Publishing, pp457-459 Fukaya, `` An Optimal Route Computation Method for GMPLS Survivable Network with a Tabu Search Algorithm for the Weighted Constraint Satisfaction Problem '', Networks2006

  However, in such a conventional technique, the traffic amount of the communication path is predicted based on the traffic information observed on the communication path to be predicted in the observation period, and further on the environmental information about the user who uses the network. However, specific traffic fluctuation factors of individual lines generated on the communication path are not considered, and there is a problem that the traffic amount of the communication path at a future target time cannot be accurately calculated.

  The present invention is for solving such problems, and provides a traffic analysis apparatus, method, and program capable of more accurately calculating the traffic amount of a communication path in consideration of individual traffic variation factors. It is aimed.

In order to achieve such an object, a traffic analysis apparatus according to the present invention includes a data storage unit that stores network status data including observation traffic data indicating a time-series change in the amount of observation traffic in an arbitrary communication path; Based on the network status data of the data storage unit, a model configuration unit that configures a traffic model for estimating a variable factor traffic amount corresponding to the variable factor for each of a plurality of variable factors that vary the traffic amount of an arbitrary communication path And a traffic analysis unit that estimates a variable element traffic amount in the target communication path based on an input analysis target condition for the target communication path, and a target communication by aggregating the estimated variable element traffic amount and a traffic counting unit for calculating a target amount of traffic paths, Mo As a traffic model, a short-term traffic fluctuation model that estimates burst-like short-term traffic volume that occurs in a short period of time that cannot be observed due to constraints such as network device / server functions and performance It is what I did.

In addition, another traffic analysis apparatus according to the present invention includes a data storage unit that stores network status data including observation traffic data indicating a time-series change in an observation traffic amount in an arbitrary communication path, and a network of the data storage unit For each of a plurality of variable elements that change the traffic volume of an arbitrary communication path based on the situation data, a model configuration unit that configures a traffic model that estimates a variable element traffic volume corresponding to the variable element, and an input target Based on the analysis target condition for the communication path, the traffic analysis unit that estimates the variable element traffic amount in the target communication path by the traffic model, and the target traffic amount of the target communication path is obtained by aggregating the estimated variable element traffic amount. and a traffic counting unit that calculates, in the model configuration unit, Torahi' One model is obtained by the target communication path to constitute a utilization start time traffic model for estimating the utilization start time traffic amount generated from the line to be used for new.

In addition, another traffic analysis apparatus according to the present invention includes a data storage unit that stores network status data including observation traffic data indicating a time-series change in an observation traffic amount in an arbitrary communication path, and a network of the data storage unit For each of a plurality of variable elements that change the traffic volume of an arbitrary communication path based on the situation data, a model configuration unit that configures a traffic model that estimates a variable element traffic volume corresponding to the variable element, and an input target Based on the analysis target condition for the communication path, the traffic analysis unit that estimates the variable element traffic amount in the target communication path by the traffic model, and the target traffic amount of the target communication path is obtained by aggregating the estimated variable element traffic amount. and a traffic counting unit that calculates, in the model configuration unit, Torahi' One model is obtained by so configuring the traffic trend changes model to estimate the trend change traffic due to a change in the tendency of increase or decrease the observed traffic volume.

Further, the traffic analysis method according to the present invention is used in a traffic analysis apparatus including a storage unit and an arithmetic processing unit, and includes a network situation including observation traffic data indicating a time-series change in an observation traffic amount in an arbitrary communication path. A traffic analysis method for calculating a target traffic amount that occurs in a target communication path at a future target time based on data, and an arithmetic processing unit varies the traffic amount of an arbitrary communication path based on network status data For each of a plurality of variable elements to be processed, a step of constructing a traffic model for estimating a variable element traffic amount corresponding to the variable element, and an arithmetic processing unit, based on the analysis target condition regarding the target communication path, the target Estimating the amount of variable traffic in a communication path using a traffic model , The arithmetic processing unit, and a step of calculating a target amount of traffic communication paths for by aggregating the estimated variables traffic, the step of configuring the traffic model, as one of the traffic model, network device A short-term traffic fluctuation model for estimating a bursty short-term traffic volume that occurs in a short period of time that cannot be observed due to constraints such as server functions and performances is constructed.

Further, the program according to the present invention includes a storage unit and an arithmetic processing unit, and based on network status data including observation traffic data indicating a time-series change in the observation traffic amount in an arbitrary communication path, a future target time point In the computer of the traffic analysis device that calculates the target traffic amount generated in the target communication path in the calculation processing unit, for each of a plurality of variable elements that change the traffic amount of an arbitrary communication path based on the network status data, The step of constructing a traffic model for estimating the variable element traffic volume corresponding to the variable element, and the arithmetic processing unit, based on the analysis target condition relating to the input target communication path, the variable element traffic volume in the target communication path Estimated by the model and the processing unit. Variables traffic to and a step of calculating a target amount of traffic communication paths for by aggregating the step of configuring the traffic model, as one of the traffic model, network device server function, performance of A short-term traffic fluctuation model that estimates the amount of bursty short-term traffic that occurs in a short period of time that cannot be observed due to constraints is constructed.

  According to the present invention, it is possible to perform modeling in consideration of a life cycle such as a use start period, a normal period, and a maximum bandwidth use period regarding the traffic occurrence state for each traffic for each contracted line accommodated in a communication path. . For this reason, it is possible to calculate the target traffic amount more accurately compared to the traffic prediction based on the environment information surrounding the target communication path as in the past, and for example, it occurs in a future design period preset at the time of design. A target communication path capable of stably satisfying a possible traffic volume can be designed with high accuracy.

Next, an embodiment of the present invention will be described with reference to the drawings.
[Traffic analyzer]
First, a traffic analyzer according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing a configuration of a traffic analysis apparatus according to an embodiment of the present invention.

  The traffic analysis apparatus 10 includes an information processing apparatus such as a server apparatus or a PC terminal as a whole, and is based on a traffic model configured based on network status data 51 acquired from a network status management terminal 20 connected via a communication line. Thus, it has a function of estimating the variable element traffic amount in an arbitrary target communication path, calculating the traffic amount of the target communication path by summing up these variable element traffic amounts, and outputting the calculated traffic amount to the traffic management terminal 30.

  The network status management terminal 20 comprises an information processing device such as a server device or a PC terminal as a whole, and monitors the device configuration status of the network based on various network setting conditions acquired from the target network. A function of acquiring device configuration data 21, a function of acquiring user management data 22 including future plan information such as use start information of each line input from a user management information system (not shown), and the network The traffic analysis device 10 obtains the observation traffic data 23 by observing the traffic amount of the communication path of the communication path, and the network status data 51 including the network device configuration data 21, the user management data 22, and the observation traffic data 23. The function to transmit to.

  The traffic management terminal 30 is composed of an information processing device such as a server device or a PC terminal as a whole. The traffic management terminal 30 has a function of transmitting an analysis target condition 52 related to an analysis target to the traffic analysis device 10 and a traffic analysis result corresponding to the analysis target condition. A function of receiving from the analyzer 10 and displaying the screen. The analysis target condition 52 relates to the analysis target and analysis conditions such as target communication path specifying information indicating which communication path on the network is to be analyzed, and a target time point indicating which traffic position is to be analyzed. Various information.

  In the traffic analysis device 10, the traffic analysis apparatus 10 is configured according to the variation factor for each of a plurality of variation factors that fluctuate the traffic amount of an arbitrary communication path based on the network status data stored in the data storage unit. Construct a traffic model that estimates the variable element traffic volume, and based on the input analysis target conditions for the target communication path, estimate the variable element traffic volume in the target communication path using the traffic model, and estimate the variable element traffic volume To calculate the target traffic amount of the target communication path.

Next, the configuration of the traffic analyzer according to the embodiment of the present invention will be described in detail with reference to FIG.
The traffic analysis apparatus 10 includes a communication interface unit (hereinafter referred to as a communication I / F unit) 11, a data storage unit 12, a storage unit 13, and an arithmetic processing unit 14 as main functional units. It is assumed that the traffic analysis apparatus 10 has a configuration included in a general information processing apparatus such as an operation input unit or a screen display unit in addition to these function units.

  The communication I / F unit 11 includes a dedicated data communication circuit, and performs data communication with the network status management terminal 20 via a communication line, thereby receiving network status data 51 from the network status management terminal 20; By performing data communication with the traffic management terminal 30 via the communication line, it has a function of receiving the analysis target condition 52 from the traffic management terminal 30 and a function of transmitting the analysis result 53 to the traffic management device 30. ing.

  The data storage unit 12 includes a storage device such as a hard disk or a memory, and has a function of storing various processing information used by the arithmetic processing unit 14. Main processing information stored in the data storage unit 12 includes a collected data storage database (hereinafter referred to as a collected data storage DB) 12A, a model configuration database (hereinafter referred to as a model configuration DB) 12B, a model configuration parameter 12C, There is a traffic model 12D and an analysis result storage database (hereinafter referred to as analysis data storage DB) 12E.

The collected data storage DB 12 </ b> A is a database that stores data used for ensuring model reproducibility and the like in the network status data 51.
The model configuration DB 12 </ b> B is a database that stores data necessary for configuring the traffic model 12 </ b> D in the network status data 51.
The model configuration parameter 12C is a parameter indicating various conditions for configuring the traffic model 12D.

The traffic model 12D is a model that is configured for each of a plurality of variable elements that change the traffic volume of an arbitrary communication path and estimates the variable element traffic volume corresponding to the variable element.
The analysis result storage DB 12E is a database that stores analysis results including the traffic amount estimated by the arithmetic processing unit 14 using the traffic model 12D.

  The storage unit 13 includes a storage device such as a hard disk or a memory, and has a function of storing various processing information and programs 13P used by the arithmetic processing unit 14. The program 13P is a program that implements various processing units by being executed by the arithmetic processing unit 14, and is read and stored in advance from an external device or a recording medium via a data input / output function such as the communication I / F unit 11. Stored in the unit 13.

The arithmetic processing unit 14 includes a microprocessor such as a CPU and its peripheral circuits, and reads and executes the program 13P of the storage unit 13 to realize various processing units by cooperating the hardware and the program. It has a function.
Main processing units realized by the arithmetic processing unit 14 include a model configuration preprocessing unit 14A, a model configuration unit 14B, a traffic analysis unit 14C, and a traffic totaling unit 14D.

  Prior to the configuration of the traffic model 12D, the model configuration preprocessing unit 14A is provided to ensure the reproducibility of the model in the network status data 51 received from the network I / F unit 11 from the network status management terminal 20. It has a function of storing data to be used in the collected data storage DB 12A and a function of storing data necessary for configuring the traffic model 12D in the network status data 51 in the model configuration DB 12B.

  Based on the data stored in the model configuration DB 12B and the model configuration parameter 12C, the model configuration unit 14B responds to the variation factors for each of the plurality of variation factors that vary the traffic amount of an arbitrary communication path. It has a function of configuring the traffic model 12D for estimating the variable element traffic amount and a function of storing the configured traffic model 12D in the data storage unit 12.

  The traffic analysis unit 14C has a function of estimating, based on the analysis target condition regarding the target communication path received from the traffic management terminal 30 by the communication I / F unit 11, a traffic variable 12 traffic amount in the target communication path using the traffic model 12D. And a function of storing the analysis result including the estimated variable element traffic amount in the analysis result storage DB 12E.

  The traffic totaling unit 14D calculates the target traffic amount of the target communication path by totaling each variable element traffic amount stored in the analysis result storage DB 12E, and the analysis result 53 including the target traffic amount as the communication I. A function of transmitting to the traffic management terminal 30 via the / F unit 11.

[Model configuration operation]
Next, with reference to FIGS. 2 to 4, the model configuration operation of the traffic analysis apparatus according to the embodiment of the present invention will be described. FIG. 2 is a flowchart showing a model configuration process of the traffic analysis apparatus according to the embodiment of the present invention. FIG. 3 is a flowchart showing model configuration pre-processing. FIG. 4 is a flowchart showing the model configuration process.

The arithmetic processing unit 14 of the traffic analysis apparatus 10 is based on a model configuration instruction from an external device such as the network status management terminal 20 or the traffic management terminal 30 or a model configuration instruction from an operator detected by an operation input unit of the apparatus. Then, the model configuration process of FIG. 2 is executed.
First, the arithmetic processing unit 14 performs model configuration preprocessing as shown in FIG. 3 by the model configuration preprocessing unit 14A (step 100).

  In this model configuration pre-processing, the model configuration pre-processing unit 14A uses the communication I / F unit 11 to manage network status data 51 including network device configuration data 21, user management data 22, and observation traffic data 23. Of the network status data 51 received from the terminal 20, the data used for ensuring the reproducibility of the model is stored in the collected data storage DB 12A, and the data necessary for configuring the traffic model 12D is used for model configuration. Store in DB12B.

Subsequently, the arithmetic processing unit 14 executes model configuration processing as shown in FIG. 4 by the model configuration unit 14B (step 101).
In this model configuration process, the model configuration unit 14B reads the data stored in the model configuration DB 12B and the model configuration parameter 12C from the data storage unit 12, configures the traffic model 12D based on these data, The obtained traffic model 12D is stored in the data storage unit 12.

  Based on various input conditions (input parameters) specified in the analysis target condition 52, the traffic model 12D has, for each of a plurality of variable elements that change the traffic amount of an arbitrary communication path, variable element traffic corresponding to the variable element. It consists of multiple models that estimate quantities (output parameters). Note that the traffic model 12D does not handle only changes in traffic in units of lines, but also includes changes in traffic on communication paths in which a plurality of lines having various attributes are superimposed. It should be possible. For example, the communication traffic volume in a communication path can be observed, but the traffic volume of each line in the communication path accommodated in the communication path cannot be calculated, and the traffic change factor of each line is taken into account. There may be a case where traffic on the communication path is modeled based on the traffic amount on the communication path.

  The model configuration unit 14B includes a long-term traffic usage trend model configuration unit 16A, an observation traffic variation model configuration unit 16B, a short-term traffic variation model configuration unit 16C, a use start traffic model configuration unit 16D, as model configuration processing units. In addition, a traffic tendency change model configuration unit 16E is provided.

  The long-term traffic usage trend change model configuration unit 16A has a function of configuring a long-term traffic usage trend change model 15A that handles a trend of increase / decrease in traffic over time. In this long-term traffic usage trend change model 15A, as a related element, the maximum usage bandwidth (the increase in the trend is saturated once the usage bandwidth is reached) during the period in which the traffic continues to increase or decrease over time. Including. As a specific example of the long-term traffic usage trend change model 15A, there is a model for estimating the long-term traffic volume generated at the target time point based on the trend of increase or decrease in the observed traffic volume over a long period. For example, a linear approximation formula derived from the observed traffic data What is necessary is just to comprise the model for performing the process which estimates long-term traffic amount using approximate functions, such as.

  The observed traffic fluctuation model configuration unit 16B has a function of configuring an observed traffic fluctuation model 15B that handles traffic fluctuations in a sample period that can be monitored as an observed value. As a specific example of the observed traffic fluctuation model 15B, there is a model for estimating the observed traffic fluctuation amount indicating the fluctuation of the observed traffic amount. For example, it is specified by a variation such as a standard deviation of the observed traffic amount or a model configuration parameter 12C. A model for performing the process of estimating the observed traffic fluctuation amount may be configured based on the confidence level for the index.

  The short-term traffic fluctuation model configuration unit 16C has a function of configuring a short-term traffic fluctuation model 15C that handles traffic fluctuations in a short period that cannot be observed due to constraints such as network device / server functions and performance. As a specific example of the short-term traffic fluctuation model 15C, there is a model that estimates a bursty short-term traffic amount that occurs in a short time, and is configured using a known calculation technique (for example, see Non-Patent Document 5). do it.

  The use start traffic model configuration unit 16D configures a use start traffic model 15D that handles traffic at the start of use. The use start traffic model 15D is a model that represents the unknown of use traffic after the start of use, a grace period until the start of full use, and the like. As a specific example of the traffic model 15D at the start of use, the traffic distribution generated for each communication band of the existing line is obtained from the observed traffic data, and the traffic distribution corresponding to the communication band of each line that generates the traffic at the start of use. Is obtained by calculating the combined traffic distribution of the entire traffic at the start of use, and the amount of combined traffic indicating the confidence interval corresponding to the predetermined confidence level of the combined traffic distribution is newly used for the line that uses the target communication path. A model to be estimated as the traffic amount at the start of use generated from the above may be configured.

  The traffic trend change model configuration unit 16E configures a traffic trend change model 15E that handles an event in which traffic tends to change beyond the range assumed in the probability model. The traffic trend change model 15E expresses an event when a system that generates a large amount of traffic is introduced, or an event when traffic usage on a contract line owned by users having a large number of specific contract lines starts simultaneously. As a specific example, there is a model that estimates the trend change traffic amount due to the change in the increase or decrease tendency of the observed traffic amount according to the event corresponding to the condition specified by the analysis target condition 52.

[Traffic analysis operation]
Next, the traffic analysis operation of the traffic analyzer according to the embodiment of the present invention will be described with reference to FIGS. FIG. 5 is a flowchart showing a traffic analysis process of the traffic analyzer according to the embodiment of the present invention. FIG. 6 is a flowchart showing traffic analysis processing. FIG. 7 is a flowchart showing the traffic counting process.

The arithmetic processing unit 14 of the traffic analysis device 10 is based on a traffic analysis instruction from an external device such as the network status management terminal 20 or the traffic management terminal 30 or a traffic analysis instruction from an operator detected by an operation input unit of the device. Thus, the traffic analysis process of FIG. 5 is executed.
First, the arithmetic processing unit 14 performs a variable element traffic amount estimation process as shown in FIG. 6 by the traffic analysis unit 14C (step 110).

  In this variable element traffic amount estimation processing, the traffic analysis unit 14C receives the analysis target condition 52 from the traffic management terminal 30 by the communication I / F unit 11, and reads the traffic model 12D from the data storage unit 12, and these Based on the analysis target condition 52 and the traffic model 12D, for each of a plurality of variable elements that change the traffic volume of the target communication path, the variable element traffic volume corresponding to the variable element is estimated, and the analysis result of the data storage unit 12 Store in the storage DB 12E.

  The traffic analysis unit 14C includes, as an analysis processing unit, a long-term traffic usage trend change analysis unit 17A, an observed traffic fluctuation analysis unit 17B, a short-term traffic fluctuation analysis unit 17C, and a traffic at the start of use as an analysis processing unit in order to estimate individual variable element traffic amounts. An analysis unit 17D and a traffic tendency change analysis unit 17E are provided.

Based on the long-term traffic usage trend change model 15A, the long-term traffic usage trend change analysis unit 17A estimates the long-term traffic amount generated at the target time as one of the variable element traffic amounts.
Based on the observed traffic fluctuation model 15B, the observed traffic fluctuation analysis unit 17B estimates an observed traffic fluctuation amount indicating a fluctuation in the observed traffic amount as one of the variable element traffic amounts.
Based on the short-term traffic fluctuation model 15C, the short-term traffic fluctuation analysis unit 17C estimates a bursty short-term traffic quantity that occurs in a short time as one of the variable element traffic quantities.

Based on the use start time traffic model 15D, the use start time traffic analyzing unit 17D estimates the use start time traffic amount generated from the line that newly uses the target communication path as one of the variable element traffic amounts.
Based on the traffic trend change model 15E, the traffic trend change analysis unit 17E estimates a trend change traffic amount due to a change in the increase / decrease trend of the observed traffic amount as one of the variable element traffic amounts.

Subsequently, the arithmetic processing unit 14 performs a traffic totaling process as shown in FIG. 7 by the traffic totaling unit 14D (step 111).
In this traffic totaling process, the traffic totaling unit 14D reads out each variable element traffic amount stored in the analysis result DB 12E of the data storage unit 12 and totals these variable element traffic amounts to thereby obtain the target of the target communication path. The traffic amount is calculated, and the analysis result 53 including the target traffic amount is transmitted to the traffic management terminals 3 and 0 via the communication I / F unit 11.

[Effects of the present embodiment]
As described above, in the present embodiment, in the traffic analyzer, for each of a plurality of variable elements that change the traffic volume of an arbitrary communication path based on the network status data, the variable element traffic volume corresponding to the variable element is calculated. By configuring the traffic model to be estimated, estimating the variable element traffic volume in the target communication path based on the input analysis target conditions with the traffic model, and then adding up the estimated variable element traffic volume The target traffic amount of the target communication path is calculated.

  Therefore, it is possible to perform modeling in consideration of a life cycle such as a use start period, a normal period, and a maximum bandwidth use period regarding the traffic occurrence state for each traffic for each contracted line accommodated in the communication path. For this reason, it is possible to calculate the target traffic amount more accurately compared to the traffic prediction based on the environment information surrounding the target communication path as in the past, and for example, it occurs in a future design period preset at the time of design. A target communication path capable of stably satisfying a possible traffic volume can be designed with high accuracy.

  Further, in this embodiment, the case where target traffic is estimated as traffic analysis has been described as an example. However, the analysis contents in the traffic analysis unit 14C and the traffic totaling unit 14D are not limited to target traffic estimation. In addition, multifaceted analysis can be performed on the traffic of an arbitrary communication path constituting the network.

  For example, the calculated individual variable element traffic amounts are compared with the respective reference values preset in the data storage unit 12 by the traffic analysis unit 14C, and the individual variable elements in the target communication path are determined at any point in time. You may analyze whether it becomes the situation. Further, when the traffic analysis unit 14C or the traffic totaling unit 14D calculates the transition of the variable element traffic amount or the target traffic amount and maintains the current communication band of the target communication path, the individual variable element traffic amount is It is also possible to analyze the traffic satisfaction period, that is, whether the target traffic amount falls within the reference value or to what point the target traffic amount falls within the communication band of the target communication path.

Further, in the present embodiment, the case where the target traffic amount of the target communication path is calculated by totaling all the variable element traffic amounts has been described as an example. However, the target traffic amount is calculated by adding the specific variable element traffic amounts. May be calculated.
For example, when the target communication path is a newly designed communication path, the above-described existing line does not exist. For this reason, in such a case, the target traffic amount may be calculated by adding up the variable element traffic amounts other than the long-term traffic amount and the observed traffic fluctuation amount.

On the other hand, when the target communication path is a communication path that does not add a new line, the above-described new line does not exist. For this reason, in such a case, the target communication traffic volume may be calculated by adding up the variable element traffic volume other than the usage start traffic volume.
Also, if the short-term traffic fluctuation is small, the target traffic volume may be calculated by summing the variable factor traffic volume other than the short-term traffic volume. If the traffic fluctuation due to traffic trend change is small, The target traffic amount may be calculated by totaling the variable element traffic amounts other than the trend change traffic amount.

It is a block diagram which shows the structure of the traffic analyzer concerning one embodiment of this invention. It is a flowchart which shows the model structure process of the traffic analyzer concerning one embodiment of this invention. It is a flowchart which shows a model structure pre-process. It is a flowchart which shows a model structure process. It is a flowchart which shows the traffic analysis process of the traffic analyzer concerning one embodiment of this invention. It is a flowchart which shows a traffic analysis process. It is a flowchart which shows a traffic total process.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Traffic analyzer, 11 ... Communication I / F part, 12 ... Data storage part, 12A ... Collected data storage DB, 12B ... Model structure DB, 12C ... Model structure parameter, 12D ... Traffic model, 12E ... Analysis result Storage DB, 13 ... storage unit, 13P ... program, 14 ... calculation processing unit, 14A ... model configuration preprocessing unit, 14B ... model configuration unit, 14C ... traffic analysis unit, 14D ... traffic counting unit, 20 ... network status management terminal , 21 ... network device configuration data, 22 ... user management data, 23 ... observation traffic data, 30 ... traffic management terminal, 51 ... network status data, 52 ... analysis target conditions, 53 ... analysis results.

Claims (5)

A data storage unit for storing network status data including observation traffic data indicating a time-series change in the observation traffic volume in an arbitrary communication path;
Based on the network status data of the data storage unit, a model configuration unit that configures a traffic model for estimating a variable factor traffic amount corresponding to the variable factor for each of a plurality of variable factors that vary the traffic amount of an arbitrary communication path When,
Based on the input analysis target condition for the target communication path, a traffic analysis unit that estimates the variable element traffic amount in the target communication path by the traffic model;
A traffic totaling unit that calculates the target traffic volume of the target communication path by counting the estimated variable element traffic volume ;
As one of the traffic models, the model configuration unit includes a short-term traffic fluctuation model for estimating a bursty short-term traffic amount that occurs in a short period of time that cannot be observed due to constraints such as network device / server function / performance. traffic analysis apparatus characterized by configuring. A data storage unit for storing network status data including observation traffic data indicating a time-series change in the observation traffic volume in an arbitrary communication path;
Based on the network status data of the data storage unit, a model configuration unit that configures a traffic model for estimating a variable factor traffic amount corresponding to the variable factor for each of a plurality of variable factors that vary the traffic amount of an arbitrary communication path When,
Based on the input analysis target condition for the target communication path, a traffic analysis unit that estimates the variable element traffic amount in the target communication path by the traffic model;
A traffic totaling unit that calculates a target traffic amount of the target communication path by aggregating the estimated variable element traffic amount;
With
The model configuration unit configures a traffic model at the start of use for estimating a traffic amount at the start of use generated from a line that newly uses the target communication path as one of the traffic models. apparatus. A data storage unit for storing network status data including observation traffic data indicating a time-series change in the observation traffic volume in an arbitrary communication path;
Based on the network status data of the data storage unit, a model configuration unit that configures a traffic model for estimating a variable factor traffic amount corresponding to the variable factor for each of a plurality of variable factors that vary the traffic amount of an arbitrary communication path When,
Based on the input analysis target condition for the target communication path, a traffic analysis unit that estimates the variable element traffic amount in the target communication path by the traffic model;
A traffic totaling unit that calculates a target traffic amount of the target communication path by aggregating the estimated variable element traffic amount;
With
The traffic analysis apparatus, wherein the model configuration unit configures a traffic trend change model that estimates a trend change traffic amount due to a change in an increase / decrease tendency of the observed traffic amount as one of the traffic models. Targeted at a future target time point based on network status data including observation traffic data that shows time-series changes in the amount of observation traffic on any communication path, used in a traffic analyzer equipped with a storage unit and an arithmetic processing unit A traffic analysis method for calculating a target traffic volume generated in a communication path,
A step of configuring a traffic model for estimating a variable element traffic amount corresponding to the variable element for each of a plurality of variable elements that change the traffic volume of an arbitrary communication path based on the network status data by the arithmetic processing unit. When,
Estimating, by the traffic model, a variable element traffic amount in the target communication path based on an analysis target condition regarding the input target communication path by the arithmetic processing unit;
Calculating a target traffic amount of the target communication path by aggregating the estimated variable element traffic amount by the arithmetic processing unit , and
The step of configuring the traffic model includes, as one of the traffic models, a short-term traffic that estimates a bursty short-term traffic amount that occurs in a short period of time that cannot be observed due to constraints such as network device / server function / performance. A traffic analysis method characterized by comprising a fluctuation model . Target traffic generated in the target communication path at a future target time point based on network status data including observation traffic data indicating a time-series change in the observed traffic volume in an arbitrary communication path. To the computer of the traffic analyzer that calculates the volume,
A step of configuring a traffic model for estimating a variable element traffic amount corresponding to the variable element for each of a plurality of variable elements that change the traffic volume of an arbitrary communication path based on the network status data by the arithmetic processing unit. When,
Estimating, by the traffic model, a variable element traffic amount in the target communication path based on an analysis target condition regarding the input target communication path by the arithmetic processing unit;
Calculating the target traffic amount of the target communication path by aggregating the estimated variable element traffic amount by the arithmetic processing unit ; and
The step of configuring the traffic model includes, as one of the traffic models, a short-term traffic that estimates a bursty short-term traffic amount that occurs in a short period of time that cannot be observed due to constraints such as network device / server function / performance. Construct a variation model
A program characterized by that .

Images