JPH11252075A - Traffic estimation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a traffic estimation device which can quickly estimate the traffic information with high accuracy from the using situations of applications. SOLUTION: An application traffic characteristic information storage part 4 stores application traffic characteristic information 5 as the characteristic information of a traffic occurrence pattern for every application. An application use information storage part 6 stores application use information 7 as parameters (data size (bytes), data transmission/reception frequency) in application units among respective bases. A traffic information estimation part 8 estimates traffic information by using information 5 and 7 which are stored in the parts 4 and 6 respectively.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traffic estimating apparatus at each site of a communication network comprising a plurality of communication sites connected to each other via a communication line.

[0002]

2. Description of the Related Art When computer communication such as the Internet and an intranet is introduced into a corporate network, an average rate [bit / s] and a peak rate [bit / s] at each site of the corporate network.
It is necessary to estimate traffic information such as Corporate network design tools that actually support the design of corporate networks are also designed based on traffic information such as average rate [bit / s] and peak rate [bit / s] [Reference: Japanese Patent Application No. Hei 7 (1996)]. 324
790 Private line delay time evaluation method and apparatus].

[0003]

However, conventionally,
For estimating traffic information, it was necessary to estimate the traffic information manually from the usage status (or usage request) of the application, or to estimate the traffic information by actually measuring the traffic.

(Problem 1) When a user of a company requests a design of a corporate network based on the use state of an application, or a case where a company network is required to be designed in accordance with a change in the use state of an application. In many cases, it is necessary to obtain traffic information from the usage status of the application. However, it is often necessary to manually determine the traffic information from the usage status (or usage request) of the conventional application, or to measure the traffic information by actually measuring the traffic. There is a problem that the estimation work takes time, which hinders rapid construction of a corporate network, design of a corporate network, and the like.

(Problem 2) In addition, the conventional traffic information estimation work does not take into account different traffic fluctuation characteristics for each application, and in order to obtain more accurate traffic information from the usage status of the application,
There is a problem that a method for estimating traffic information in consideration of the fluctuation characteristics is required.

[0006] The present invention has been made in view of the above-mentioned circumstances, and enables rapid estimation of traffic information from the usage status of an application at each site of a corporate network, and furthermore, the traffic information corresponding to different traffic fluctuation characteristics for each application. It is an object of the present invention to provide a traffic estimation device capable of estimating traffic information with high accuracy.

[0007]

In order to solve the above-mentioned problems, according to the present invention, there is provided a traffic estimating apparatus for estimating a traffic length at each site of an enterprise network, comprising: Application traffic characteristics that store daily traffic coefficients that indicate traffic fluctuation characteristics on a daily basis and short-term fluctuation coefficients that indicate traffic fluctuation characteristics on a time basis as characteristic information. The information storage unit, the names of the applications used at each site of the corporate network, the average data size [bytes] that can be transmitted and received by the applications, and the busy hour 1
An application use information storage unit having a function of storing the number of times of transmission / reception in time as application use information; an application name used at each base of the corporate network obtained from the application use information storage unit; average data transmitted / received by the application Using the size [bytes] and the number of times of transmission / reception in one hour during the busy hour, the average rate [bit / s] of each application at the base of the corporate network is averaged [bit / s] =
An average rate calculation unit is obtained by: average data size [bytes] × number of times of transmission / reception during busy times ÷ 3600 [sec] × 8.

[0008] According to a second aspect of the present invention, there is provided a traffic estimating apparatus for estimating a traffic length at each site of a corporate network, wherein the traffic characteristic information for each application indicates a day-to-day variation characteristic of traffic. An application that stores different variation coefficients, an application that stores short-term variation coefficients that indicate the variation characteristics of traffic in units of time, an application traffic characteristic information storage unit that stores traffic characteristic information, and an application that is used at each site of the corporate network An application use information storage unit having a function of storing, as application use information, a name, an average data size [bytes] that can be transmitted / received by the application, and the number of times of transmission / reception during one hour of a busy period; of The average rate [bit / s] of each application at the business network base is determined using the name of the application used at the base, the average data size [bytes] transmitted / received by the application, and the number of transmissions / receptions during one hour during a busy time zone. Average rate [bit
/ s] = average data size [bytes] × number of times of transmission / reception during busy times ÷ 3600 [sec] × 8, and an average rate calculation unit for each application obtained by the average rate calculation function. By using the daily variation coefficient and the short-term variation coefficient of each application obtained from the application traffic characteristic information storage unit, the peak rate [bit / s] of each application at the base of the enterprise network is converted to the peak rate [bit]. / s] = average rate × (1 + coefficient of daily variation) × (1 + coefficient of short-term variation).

In the present invention, the application traffic characteristic information storage section stores application traffic characteristic information as characteristic information of a traffic occurrence pattern for each application. On the other hand, the traffic information estimating unit uses the application traffic characteristic information stored in the application traffic characteristic information storage unit and parameters (data size [bytes], number of data transmission / reception) for each application between each base. The traffic information (average rate [bit / s], peak rate [bit / s]) is estimated from Expressions 1 and 2 shown below.

Formula 1 Average rate [bit / s] = Average data size [bytes] × Number of times of transmission / reception during busy times ÷ 3600 [sec] × 8

Equation 2 Peak rate [bit / s] = Average rate × (1 + Daily variation coefficient) × (1 + Short term variation coefficient)

Formula 1 obtains the name of the application used at each site of the corporate network, the average data size [bytes] transmitted / received by the application, and the number of times of transmission / reception in one hour during the busy time zone. Average rate [bit /
s].

Equation (2) is obtained by multiplying the average rate obtained by Equation (1) by “1 + coefficient of daily variation”, thereby estimating the average rate especially during high traffic times during a busy period, and further calculating “1 + short time”. The value multiplied by the coefficient of variation "means that the maximum value of the peak rate within one hour of the busy time zone is estimated. Therefore, it is possible to estimate the maximum value of the peak rate within one hour of the busy time zone from the application usage information according to the traffic fluctuation characteristics of the application. For this reason, application usage status (data size,
Traffic information (average rate [bit /
s] and peak rate [bit / s]) can be quickly estimated.

[0014]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the traffic estimation device of the present invention. In the figure, a traffic estimating apparatus 1 includes a control unit 2, an input unit 3, an application traffic characteristic information storage unit 4, an application use information storage unit 6, a traffic information estimation unit 8, and a traffic information storage unit 9.

The application traffic characteristic information storage section 4 stores traffic characteristics (daily fluctuation characteristics, short-term fluctuation characteristics) for each application as application traffic characteristic information 5. The application usage information storage unit 6 stores an application name used for data transmission / reception between offices, the size of data that can be transmitted / received by the application [bytes], and the number of 1-hour data transmission / reception [number of times / 1 hour] when busy.
Is stored as the application usage information 7. The traffic information estimating unit 8 has a function of estimating an average rate [bit / s] and a peak rate [bit / s], which are traffic information, from the application usage information 7 stored in the application usage information storage unit 6. The traffic information storage unit 9 stores the average rate [bit / s] and the peak rate [bit / s] estimated by the traffic information estimation unit 8 as traffic information 10. The control unit 2 includes an input unit 3 connected thereto, an application traffic characteristic information storage unit 4, an application use information storage unit 6,
The traffic information estimation unit 8 and the traffic information storage unit 9 are controlled.

Next, FIG. 2 is a conceptual diagram showing the data structure of the traffic information 10 stored in the traffic information storage unit 9 in the above-described traffic estimation device 1.
In the figure, traffic information 10 includes a business section section 11 in which business sections in which traffic occurs are stored, and a column 1 of application names that can be used in the business section.
2. An average rate column 13 in which an estimated average rate [bit / s] per second of traffic generated by the application is stored, and an estimated peak rate [bit / s] per second of the traffic are stored. Column 1 of peak rate
4.

Next, FIG. 3 is a conceptual diagram showing a data structure of the application traffic characteristic information 5 in the enterprise network traffic estimating apparatus 1 described above. In the figure, the application traffic characteristic information 5 includes an application name column 15 in which an application name is stored, and a daily variation coefficient in which an application daily variation coefficient stored in the same row of the application name column 15 is stored. And a short-term variation coefficient column 17 in which the short-term variation coefficient of the application stored in the same row of the application name column 15 is stored. In addition, the same application name is not stored in the application column 15 repeatedly, and the daily variation coefficient and the short-term variation coefficient of the application are uniquely determined from the application name.

The daily variation coefficient for each application stored in the daily variation coefficient column 16 is particularly high among the average rate (α) in one hour, which is the busy time of the application, and the busy time. It is assumed that the ratio is represented by the following Expression 3 using the ratio of the average rate (β) during traffic and is determined in advance for each application. Formula 3 Daily variation coefficient = β / α-1

Similarly, the short-term variation coefficient of each application stored in the short-term variation coefficient column 17 is calculated for each short time (for example, every minute) within one hour, which is a busy time period of the application. It is assumed that the ratio is represented by the following Expression 4 using the ratio between the standard deviation (σ) of the average rate and the average rate (λ) during the busy hour, and is predetermined for each application. Equation 4 Short-term variation coefficient = 2 · σ / λ

Next, FIG. 4 is a conceptual diagram showing the data structure of the application use information 7 in the enterprise network traffic estimating apparatus 1 described above. In the figure, the application usage information 7 includes a business section section column 18 in which the business section is stored, an application name column 19 in which the application name is stored, and an average per data transmission / reception by the application in the business section. A column 20 of the data size in which the typical data size [bytes] is stored and a column 21 of the number of times of transmission / reception storing the number of times of data transmission / reception [number of times / 1 hour] in the busy period of data transmission / reception.

Next, FIGS. 5 and 6 are flow charts for explaining the traffic estimation processing in the traffic estimation device of the present invention. First, when the traffic estimation process is started (S10), the control unit 2 enters a state of waiting for input of application usage information from the input unit 3 (S1).
2). In the input unit 3, the application usage information, that is, the section of the office where the traffic occurs (office 1, office 2), the application name "A" to be used, and the data transmission / reception by the application in the office section Average data size “DATA1” [by
tes] and the number of data transmission / reception "N1" [number of times / 1 hour] in the data transmission / reception busy time zone (S)
14).

When the application use information is input from the input unit 3, the control unit 2 requests that the application use information be stored in the application use information storage unit 6 (S16). Application use information storage unit 6 that has received the application use information storage request
Sequentially stores the application usage information in each column of the same row of the application usage information 7 (S18). For example, as shown in FIG. 4, “business establishment 1 and business establishment 2” are stored in the business establishment section column 18 in the row 7-1 of the application usage information 7, and the application name column 19 is “
“A” is stored, and “N1” is stored in the transmission / reception count column 21. The processes of steps S14 to S18 are repeatedly executed until the input of the application usage information is completed.

When the input end of the application usage information is input (S20), the control unit 2 requests the traffic information estimating unit 8 to create the traffic information 10 stored in the traffic information storage unit 9 (S22). Upon receiving the traffic information creation request, the traffic information estimating unit 8 requests the application usage information storage unit 6 for the application usage information 7 (S24). Application usage information storage unit 6 receiving a request for application usage information 7
Responds with the application usage information 7 (S2
6).

The traffic information estimating unit 8 having received the response to the application usage information 7 performs a process of estimating the traffic information sequentially from the first row 7-1 of the application usage information 7 (S28). The traffic information estimating unit 8 calculates the office section “Office 1” and “Office 2” stored in the office section column 18 of the first row 7-1 of the read application usage information 7 and the application usage information. 7
The application name “A” stored in the application name column 19 of the first row of the
Is requested to be stored (S30).

The traffic information storage unit 9 that has received the request for storing the office section stores the office section “office 1”, “office 2” and the application name “A” in the traffic information 10 of the traffic information storage unit 9. Column 11 of the business section in the first row 10-1 and the first row 10 of the traffic information 10
-1 is stored in the column 12 of the application name, and the end of storage is returned to the traffic information estimating unit 8 (S32). The traffic information estimating unit 8 that has received the notification of the end of the storage of the office section and the application name, sets the daily variation coefficient of the application “A” stored in the application name column 12 of the first row 10-1 of the traffic information 10. Then, a short-term variation coefficient is requested from the application traffic characteristic information storage unit 4 (S34).

The application traffic characteristic information storage unit 4 receiving the request for the daily variation coefficient and the short-term variation coefficient of the application “A” stores “A” in the application traffic column 5 of the application name in the application traffic characteristic information 5.
Is returned as the daily variation coefficient “V1a” and the short-term variation coefficient “V2a” stored in the row 5-1 in which is stored (S36). The traffic information estimating unit 8 that has received the response of the daily variation coefficient “V1a” and the short-term variation coefficient “V2a” of the application name “A” returns to the data size column 20 of the first row 7-1 of the application usage information 7. From the data size "DATA1" [bytes] stored in the column of the number of times of transmission / reception "N1" [number of times / hour] per busy hour stored in the column of the number of times of transmission / reception. s] is estimated using the following Expression 5 (S38). Equation 5 Average rate “AV1” [bit / s] = data size “DA”
TA1 "x number of transmissions and receptions" N1 "@ 3600 x 8

The traffic information estimating unit 8 that has obtained the average rate “AV1” [bit / s] receives the response in step S36, the daily variation coefficient “V1a” and the short-term variation coefficient “V”.
The peak rate “P1” [bit / s] is estimated from 2a ”using the following Equation 6 (S40): Equation 6 Peak rate“ p1 ”[bit / s] = average rate“ AV ”
1 ”× (1 + daily variation coefficient“ V1a ”) × (10 short-time variation coefficients“ V2a ”)

Then, the traffic information estimating unit 8 that has estimated the average rate “AV1” and the peak rate “P1” requests the traffic information storage unit 9 to store the average rate “AV1” and the peak rate “P1”. (S42).
Upon receiving the storage request for the average rate “AV1” and the peak rate “P1”, the traffic information storage unit 9 outputs the average rate “A”.
V1 "and the peak rate" P1 "are stored in the average rate column 13 and the peak rate column 14 of the first row 10-1 of the traffic information 10 in the traffic information storage unit 9, and the end of storage is transmitted to the traffic information estimation unit 8. In response to the response (S44), the traffic information estimating unit 8 that has received the storage end notification of the average rate "AV1" and the peak rate "P1" repeats the processing of steps S30 and S44 for each row of the application usage information 7 in the same manner. .

By performing the calculation up to the last line of the application usage information 7, the traffic information estimating unit 8 that has created the traffic information 10 notifies the control unit 2 of a traffic information creation end notification (S46). The control unit 2 that has received the notification of the completion of the traffic information creation ends the traffic information creation processing (S48). Thus, the traffic estimation processing ends.

Next, evaluation of the traffic information estimation method in the above-described traffic estimation device will be described. Taking three types of applications (e-mail, WWW, accounting application) as examples, the application traffic characteristic information (daily variation coefficient, short-term variation coefficient) of the application and the average rate of the busy hours of each application actually measured From which the peak rate was estimated. Then, the estimated peak rate was compared with the actually measured traffic data. FIG. 7 shows the result of the comparison.

FIGS. 7A, 7B, and 7C are graphs of distribution probabilities of traffic rates created from traffic data of electronic mail, WWW, and accounting applications, respectively. The vertical axis of each graph is the distribution probability of the actually measured traffic rate equal to or less than the traffic rate on the horizontal axis. Estimated peak rate γ1, γ2, γ for each application
3 are shown in the graphs of FIGS.

FIG. 7 shows that 99.91% of the traffic rate measured below the estimated peak rate γ1, γ2, γ3 of e-mail, WWW, accounting application, 99.91%.
81% and 99.91% are distributed, and the estimated peak rates γ1, γ2, and γ3 satisfy most of the measured traffic rates. Therefore, the present traffic information estimation method is considered to be very effective as a method for estimating the peak rate according to the traffic fluctuation characteristics for each application.

[0033]

As described above, according to the present invention,
The following effects can be obtained. (1) An advantage is obtained in that traffic information can be quickly estimated from the use status of an application at each site in a corporate network. (2) In the above-described estimation of traffic information, there is obtained an advantage that it is possible to estimate traffic information with higher accuracy according to traffic fluctuation characteristics different for each application.

[Brief description of the drawings]

FIG. 1 is a block diagram of a traffic estimation device according to an embodiment of the present invention.

FIG. 2 is a data configuration diagram of traffic information stored in a traffic information storage unit in FIG. 1;

FIG. 3 is a data configuration diagram of application traffic characteristic information stored in an application traffic characteristic information storage unit in FIG. 1;

FIG. 4 is a data configuration diagram of application usage information stored in an application usage information storage unit in FIG. 1;

FIG. 5 is a flowchart of a traffic estimating process of the traffic estimating device in FIG. 1;

FIG. 6 is a flowchart of a traffic estimation process of the traffic estimation device in FIG. 1;

FIG. 7 is a graph showing a measured traffic rate distribution probability for evaluating a traffic information estimating method in the traffic estimating apparatus of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Traffic estimation apparatus 2 Control part 3 Input part 4 Application traffic characteristic information storage part 6 Application use information storage part 8 Traffic information estimation part 9 Traffic information storage part

Claims (2)

[Claims] 1. A traffic estimating apparatus for estimating a traffic length at each site of a corporate network, wherein a daily variation coefficient indicating a daily traffic variation characteristic and a An application traffic characteristic information storage unit having an application traffic characteristic information storage unit for storing application traffic characteristic information for storing a short-term variation coefficient indicating a traffic variation characteristic, an application name used in each base of a corporate network, an average transmitted and received by the application; An application use information storage unit having a function of storing the data size [bytes] and the number of times of transmission / reception during one hour of a busy time period as application use information; and an application used at each base of the corporate network acquired from the application use information storage unit. Using the application name, the average data size transmitted / received by the application [bytes], and the number of transmissions / receptions during one hour during a busy period, the average rate [bit / s] of each application at the base of the corporate network is calculated as the average rate [bit / s]. s] = average data size [bytes] × number of times of transmission / reception during busy hours / 3600 [sec] × 8. 2. A traffic estimating apparatus for estimating a traffic length at each base of a corporate network, wherein a daily variation coefficient indicating a daily traffic variation characteristic and a An application traffic characteristic information storage unit having an application traffic characteristic information storage unit for storing application traffic characteristic information for storing a short-term variation coefficient indicating a traffic variation characteristic, an application name used in each base of a corporate network, an average transmitted and received by the application; An application use information storage unit having a function of storing the data size [bytes] and the number of times of transmission / reception during one hour of a busy time period as application use information; and an application used at each base of the corporate network acquired from the application use information storage unit. Using the application name, the average data size transmitted / received by the application [bytes], and the number of transmissions / receptions during one hour during a busy period, the average rate [bit / s] of each application at the base of the corporate network is calculated as the average rate [bit / s]. s] = average data size [bytes] × number of times of transmission / reception during busy times ÷ 3600 [sec] × 8; an average rate calculation unit for each application obtained by the average rate calculation function; Using the daily variation coefficient of each application and the short-term variation coefficient obtained from the application traffic characteristic information storage unit, the peak rate [bit / s] of each application at the base in the corporate network is calculated as the peak rate [bit / s]. s] = Average rate × (1 + coefficient of daily variation) × (1 + coefficient of short-term variation) peak rate calculation And a traffic estimating device.