JP2003273917A - Traffic control method and its system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To maintain service quality between pieces of equipment on a network. <P>SOLUTION: Between at least one client 4 and at least one server 1 (2, 3) that offers service to the client 4 via the network, a traffic controller 5 monitors line usage rate of a plurality of lines existing between the client 4 and the server 1 (2, 3) and data communication processing ability between them. Then, when the data communication processing ability is lowered, the traffic controller 5 specifies a line to be a bottleneck among the respective lines based on monitoring information of the line usage rate and performs traffic control to the bottleneck line. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traffic control method and system for improving the processing capacity between devices that perform data communication on a network.

[0002]

2. Description of the Related Art Conventionally, between devices performing data communication on a network, a processing capacity such as a response delay during data communication is deteriorated due to an increase in traffic amount due to an increase in data amount. there were. Therefore, as a method for improving such inconvenience, a network traffic control method is disclosed in, for example, Japanese Patent Application Laid-Open No. 7-28733. This Japanese Patent Laid-Open No. 7-2873
According to the network traffic control method disclosed in Japanese Patent Laid-Open No. 3, an intermediate device is provided on a network having a plurality of clients that are data transmission sources and a plurality of servers that are data transmission destinations. To reduce the traffic volume on the network and improve the processing capacity during data communication. Specifically, once the transmission data sent from the client to the server is put in the queue of the intermediate device,
When the CPU load of the server is low, the transmission data is transmitted from the queue in ascending order of the data size to the server to improve the above inconvenience.

[0003]

When a decrease in processing capacity (a problem of service quality) such as a response delay between the client and the server, that is, End-to-End occurs, the End-
It can be seen that not all of the lines passing between the to-End nodes are bottlenecks, but only the lines between some nodes. In such a case, it is possible to improve the processing capacity by securing a band for data in which the quality of service is deteriorated, at least for the bottleneck line causing the deterioration of the processing capacity. Then, in order to realize this, deterioration of service quality is detected. Next, it is necessary to detect a bottleneck line and perform traffic control for the bottleneck line to secure a bandwidth for data in which service quality is deteriorated.

However, even if such control is applied to the above-described conventional network traffic control method, in this network traffic control method, the above events are used as a trigger to coordinate the above and the network traffic. Since a method of dynamically performing control has not been established and the identification of the bottleneck line to be controlled and its control timing is all entrusted to the knowledge and experience of the administrator, the burden on the administrator is increased. There was an inconvenience. After all, in the network traffic control method of this conventional example, data cannot be transmitted unless the CPU load of the transmission destination becomes low, and the processing capacity is improved depending on the load of the CPU. There was an inconvenience that it was difficult to plan.

The present invention improves the inconvenience of the conventional example, maintains the quality of service between devices on the network, that is, suppresses the deterioration of the data communication processing capability between the devices, and the system thereof. The purpose is to provide.

[0006]

[Means for Solving the Problems] To achieve the above object,
According to the present invention, between at least one client and at least one server that provides a service to the client via the network, the line usage rate of a plurality of lines existing between the client and the server,
And the data communication processing capacity between them. Then, when the data communication processing capability decreases, the bottleneck line is identified from among the respective lines based on the line utilization rate monitoring information, and traffic control is performed for this bottleneck line.

In order to achieve the above object, according to the present invention, the line usage rate of a plurality of lines existing between at least one client and at least one server that provides the service to the client via a network is calculated. A line usage rate monitoring means for monitoring, a data communication processing capacity monitoring means for monitoring the data communication processing capacity between the client and the server, and this data communication processing capacity monitoring means determines that the data communication processing capacity has decreased. In this case, the bottleneck line identifying means for identifying the bottleneck line from each line based on the line utilization rate monitoring information of the line utilization rate monitoring means and the bottleneck line identifying means Traffic control system including traffic control means for controlling traffic to a bottleneck line It is building.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of a traffic control system according to the present invention will be described with reference to FIGS.

[Configuration of Traffic Control System] First, the configuration of the traffic control system will be described with reference to FIG. This traffic control system responds to a server that provides various services on the network, a client machine 4 that receives the various services from this server on the network, and a network congestion status (status of deterioration of service quality). Traffic control device 5 for controlling traffic of the network. Here, in the present embodiment, the Web server 1, the file server 2, and the database server (hereinafter referred to as “DB server”) 3 are illustrated as the servers, but they are not necessarily limited to these. Hereinafter, each of these configurations will be described.

[Server] First, the Web server 1 requests the W requested by an instruction from the client machine 4, for example.
A controller (not shown) stores the eb content in a storage (not shown).
It provides a service such as reading from and sending to the client machine 4. Also, the file server 2
For example, the control unit (not shown) reads a file requested by an instruction from the client machine 4 from a storage unit (not shown) and sends the file to the client machine 4. Furthermore, the DB server 3 receives data to be processed, which is read by a control unit (not shown) from a storage unit (not shown) when a transaction process such as data rewriting, addition or deletion is requested from the client machine 4, for example. The service required is to perform the processing requested by the above and store the data in the storage unit.

[Client Machine] Next, the aforementioned client machine 4 will be described. The client machine 4 has a measurement program for measuring the service provision time with the server and transmitting the measurement result to the traffic control device 5. Here, the service providing time is, for example, the time until the client machine 4 issues a web content download request instruction from the client machine 4 to the web server 1 and the client machine 4 receives the web content. Say. Further, when it is between the file server 2 and the file server 2, it means the time until the client machine 4 issues a file transfer instruction to the file server 2 and the client machine 4 receives this file. Furthermore, when it is between the DB server 3 and the DB server 3, it means the time until the client machine 4 instructs the DB server 3 to perform transaction processing and the client machine 4 receives the execution result of this transaction processing.

Specifically, as shown in FIG. 1, the client machine 4 includes a program storage unit 4a storing the measurement program, and a management table storage unit 4b storing a measurement information management table 4B shown in FIG. , A control unit 4c in which a central processing unit (CPU) (not shown) performs processing described later based on the measurement program read in a main storage device (memory) (not shown), and an operator sends a desired instruction such as service provision to a server. For example, a keyboard for an electronic computer or the like and an instruction unit 4d such as a mouse for giving are provided.

Here, the measurement information management table 4B of the management table storage unit 4b includes identification information such as the IP address of the client machine 4, information on the measured value X of the service provision time, and a server to be measured (in other words, a service). Of I
It is a table for storing measurement information composed of identification information such as P address, and as shown in FIG. 2, a client identification information column 4B ₁ , a measurement value information column 4B ₂ and a server identification information column 4B ₃ are correspondingly provided. Have. For example, if the measured value X between a certain client machine and the server is 6 seconds, “6s” is displayed in the measured value information column 4B ₂ as shown in FIG.
Is stored. Here, in the present embodiment, the client identification information column 4B ₁ and the server identification information column 4B.
The ₃ shall identification information of the identification information and the server each client machine 4 is stored in advance. In addition, the measurement information management table 4B corresponding to the combination with a plurality of servers is stored in the management table storage unit 4b.

[0014] Subsequently, the control unit 4c of the client machine 4, as shown in FIG. 1, and measure the service provision time described above, the measurement functions 4c ₁ for recording the measurement results in the measurement information management table 4B, the It has a transmission function 4c ₂ for transmitting the data of the measurement information management table 4B to the traffic control device 5. Here, this transmission function 4c
₂ transmits the data of the measurement information management table 4B, for example, at predetermined intervals or at predetermined times that are arbitrarily set in advance.
The transmission time may be set for each server identification information, or may be set regardless of the server identification information (that is, all the measurement information management tables 4B are transmitted when the time comes. Thing).

[Traffic Control Device] Next, the above-mentioned traffic control device 5 will be described. As described above, the traffic control device 5 controls the traffic of the network according to the congestion status of the network. In the present embodiment, as shown in FIG. 1, a congestion status management device that manages the congestion status between the client machine and the server based on each information of the measurement information management table 4B received from the client machine 4 via the network. 51 and the usage rate of the line between each node existing between the client machine and the server via the network, and the traffic control of the bottleneck line between the client machine and the server instructed by the congestion status management device 51. The traffic control device 5 is configured with the main device 52 that performs the traffic control.

[Congestion Status Management Device] Specifically, the congestion status management device 51, as shown in FIG. 1, includes a program storage unit 51a storing a congestion status management program and a client machine-server connection shown in FIG. Of the first congestion status determination table storage unit 51b for storing the congestion status determination table (hereinafter, referred to as "first congestion status determination table") 51B and the measurement information management table 4B received from the client machine 4 via the network. A reception table storage unit 51c for storing data, and a control unit 51d for performing processing described later by a central processing unit (CPU) not shown based on a congestion status management program read in a main storage device (memory) not shown. .

Here, the first congestion situation determination table 51B of the first congestion situation determination table storage unit 51b will be described. This first congestion status determination table 51B is the same as the above-mentioned measurement information management table 4B, identification information of the client and server, and the measurement information management table 4
FIG. 3 is a table that stores information on threshold values for determining the presence or absence of congestion in comparison with the measured value X of B, as shown in FIG.
A client identification information column 51B ₁ , a server identification information column 51B ₂ and a threshold information column 51B ₃ are provided corresponding to each. For example, the first congestion situation determination table 5 shown in FIG.
In 1B, if the measured value X exceeds a threshold value (5 seconds) between a certain client machine and the server, it is determined to be congested. Here, in the present embodiment, it is assumed that the client identification information column 51B ₁ and the server identification information column 51B ₂ respectively store the identification information of the client machine 4 and the identification information of the server in advance. Further, the first congestion situation determination table storage unit 51b stores the first congestion situation determination table 51 corresponding to each combination of the plurality of client machines 4 and the plurality of servers.
B is stored.

Next, the control of the congestion status management device 51
The section 51d will be described. This control unit 51d is
Measurement information management table 4 received from the client machine 4
Information of B and information of the first congestion situation determination table 51B
Contrast situation with congestion between client machine and server
Judgment function 51d for judging₁And this judgment function 51d ₁
If it is judged as “crowded” in
The fact that there is congestion between the Ant machine and the server, and this
The identification information of the client machine and the server is used as the main unit 52.
Function 51d to notify_TwoHave and. Where this
Transmission function 51d _TwoIs the determination function 51d, for example.₁At
After the determination of "congestion" is made, the above-mentioned information is sent to the main unit 5
Send to 2. In addition, every predetermined interval set in advance or
It may be transmitted at a predetermined time.

[Main Device] The main device 52 will be described. As shown in FIG. 1, the main device 52 includes a monitoring program storage unit 52a storing a line usage rate monitoring program for monitoring the line usage rate of each line existing between the client machine and the server, and the client machine and the server. A traffic control program storage unit 52b storing a traffic control program for instructing traffic control for a bottleneck line, and a line identification table storage unit 5 storing a line identification table 52C shown in FIG. 4 described later.
2c, a second congestion status determination table storage unit 52d for storing a congestion status determination table between nodes (hereinafter, referred to as “second congestion status determination table”) 52D shown in FIG. 5, which will be described later, and FIG. A monitoring table storage unit 52e for storing the line utilization rate monitoring table 52E shown, and control for performing processing described later by a central processing unit (CPU) not shown based on the programs read into a main storage device (memory) not shown And a portion 52f. The line utilization rate in the present embodiment is the ratio of the amount of data flowing through the line (the amount of data flowing through the line / the total bandwidth of the line) to the total bandwidth of the line. In other words, it is the ratio of the total amount of data that can be flowed to the line and the amount of data that is flowing through this line (the amount of data that is flowing through the line / the total amount of data that can be flowing through the line).

Here, the line identification table storage 52
The line identification table 52C of c will be described. The line identification table 52C is a table for identifying the line between the nodes existing between the client machine and the server based on the identification information of the client machine and the server sent from the congestion management apparatus 51, and FIG. As shown in FIG. 5, it has a client identification information column 52C ₁ , a server identification information column 52C ₂ and a line information column 52C ₃ . Here, it is assumed that information corresponding to each column is stored in advance in each column of the present embodiment. The line identification table storage unit 52c stores the line identification table 5 corresponding to each combination of the plurality of client machines 4 and the plurality of servers.
2C is stored. For example, the line identification table 52C shown in FIG. 4A is the client machine 4 shown in FIG.
Is a table for identifying a line existing between the server and the Web server 1, and is a line identifying table 52C shown in FIG. 4B.
Is the client machine 4 and the file server 2 shown in FIG.
It is a table that identifies the lines existing between and.

The second congestion status determination table 52D of the second congestion status determination table storage unit 52d will be described. The second congestion situation determination table 52D is a table that stores threshold value information of the line utilization rate for each line on the network, and the threshold value information is preset and the threshold value information column 52D _{1 as} shown in FIG. It is stored in.

Furthermore, the monitoring table storage section 52e is also provided.
The line usage rate monitoring table 52E will be described. The line utilization rate monitoring table 52E is used in the control unit 5 described later.
It is a table for storing the information of the line usage rate of each line acquired at a predetermined interval or at a predetermined time by the monitoring function 52f ₁ of 2f, and the line usage rate information is a line usage rate information column as shown in FIG. 52E ₁ .

Next, the control unit 52f of the main unit 52 will be described. The control unit 52f has a processing function that operates based on the monitoring program and a processing function that operates based on the traffic control program.

First, as a processing function that operates based on the monitoring program, the line usage rate shown in FIG. 1 is used to acquire the line usage rate information of the lines between nodes existing on the network and update the line usage rate monitoring table 52E. Monitoring function 5
There is 2f ₁ . For example, the information on the line usage rate is acquired from the node at predetermined intervals or at predetermined times. Here, the line to be acquired by the monitoring function 52f ₁ may be a line existing between the client machine and the server designated in advance.

Further, as a processing function which operates based on the traffic control program, the line existing between the client machine and the server, which has been notified from the congestion state management device 51 that congestion has occurred, is based on the line identification table 52C. A line for which the line usage rate exceeds a threshold value by specifying and comparing the respective information of the second congestion status determination table 52D and the line usage rate monitoring table 52E (that is, the bottleneck line between the client machine and the server)
Traffic control to the line identification function 52f ₂ shown in FIG. 1 for identifying the node and each node (router when the network is the Internet) located at both ends of the bottleneck line identified by the line identification function 52f ₂ . There is an instruction function 52f _{3 for} instructing to execute.

Here, the traffic control of this embodiment will be described. The traffic control of the present embodiment is control for securing a band for data in which the quality of service is deteriorated in the bottleneck line specified by the line specifying function 52f ₂ . In such a case, the bandwidth for data in which the quality of service has deteriorated is basically at least the transmission of the Web content to be transmitted to the client machine 4 when the service with the Web server 1 is taken as an example. It means the bandwidth for communication (comparable to the data capacity of the Web content) as much as possible.

In the present embodiment, in order to secure the above band, the instruction function 52f ₃ of the control unit 52f of the main unit 52.
Performs the operation of Expression 1 or Expression 2 according to the following conditions,
Based on the result of this calculation, execution of traffic control is instructed for the bottleneck line between the client machine and the server that is the target of the instruction. The calculation here is the instruction function 5
One client machine to be instructed by 2f ₃
It is a calculation of a bandwidth Y (kbps) allocated to communication of service data between servers. Instruction function 5
2f ₃ gives an instruction to secure the bandwidth Y to each node located at both ends of the bottleneck line identified by the line identification function 52f ₂ .

First, when there is one service to be instructed (between the client machine and the server), or even if there are a plurality of services to be instructed, the bottleneck line to be controlled is not shared by the respective services. The case will be described. The bandwidth Y ₁ (kbps) obtained by the arithmetic processing in such a case is Y _all (kbps) as the total bandwidth of the bottleneck line to be controlled, and the service (between the client machine and the server) to be instructed in the bottleneck line. Assuming that the ratio of the bandwidth that can be allocated for use is a (%), it can be obtained by an arithmetic expression of Y ₁ = a · Y _all / 100 (Equation 1).

Next, the case where there are a plurality of services (between the client machine and the server) to be instructed and the bottleneck line to be controlled is common to each service will be described. In such a case, each service is weighted in advance, and the weight of the service determines the bandwidth of the service in which data can be preferentially passed (that is, the bandwidth can be secured). The bandwidth Y ₂ (kbps) obtained by the arithmetic processing in such a case is the total bandwidth of the bottleneck line to be controlled, Y _all (kbps), and the occupancy rate that can be occupied by the bottleneck line for the target service. (For example, in the present embodiment, We
b (%) allocated to the services related to the server 1, the file server 2 and the DB server 3) is b (%), and the bandwidth that can be preferentially allocated for the service that is one instruction target in the bottleneck line When the ratio is c (%), Y ₂ = (b / 100) × (c · Y _all / 100) (Equation 2) is obtained.

Here, in the equation (2), weighting is performed by the bandwidth ratio c. For example, for a service related to the Web server 1, this bandwidth ratio is c
₁ (%), c ₂ (%) for the service related to the file server 2, and c ₂ for the service related to the DB server 3
₃ (%), c ₁ = 50 (%), c ₂ = 20 (%),
Set c ₃ = 30 (%). As a result, the service related to the file server 2, the service related to the DB server 3,
It is possible to perform weighting such that the priority of data communication becomes higher in the order of services related to the Web server 1. In this way, it is possible to set the priority order of services when there are a plurality of services to be instructed and the bottleneck lines to be controlled are common to the respective services.

FIG. 7 shows a conceptual diagram of the present system described above. In FIG. 7, the client machine 4 and We
b The traffic control between the server 1 and between the client machine 4 and the file server 2 will be illustrated. Here, the client machine 4 and the Web server 1
Are connected via four nodes N1 to N4, and three lines 1 to 3 exist between these nodes. Further, the client machine 4 and the file server 2 are connected via four nodes N1, N2, N5 and N6, and three lines 1 are provided between these nodes.
There are 4,5.

The main unit 52 is, as shown by the arrow in FIG.
For example, the information of the line utilization rate is acquired from the lines 1 to 5 at predetermined intervals or at predetermined times (in FIG. 7, the drawing from the lines 4 and 5 is omitted because the drawing is complicated). Here, when the quality of service is deteriorated (congested) in one or both of the client machine and the Web server, the client machine and the file server, the congestion status management apparatus 51 causes the arrow shown in FIG. As described above, the fact that the client machine and the server are congested and the identification information of the client machine and the server are notified to the main device 52.

With this notification, the main unit 52 detects the bottleneck line from the lines existing between the client machine and the server, and notifies the nodes located at both ends of this bottleneck line to the nodes. It gives a traffic control instruction to secure a bandwidth for data whose service quality has deteriorated at the bottleneck line. For example, when the processing capacity of data communication between the client machine and the Web server is low, the main device 5
2 detects a line which is a bottleneck from lines 1 to 3. Here, it is assumed that the line 2 is a bottleneck line. In such a case, the main device 52 gives a traffic control instruction to the nodes N2 and N3 located at both ends of the line 2.

The bandwidth for data communication in the bottleneck line is secured by the traffic control of the node to which the instruction is given as described above, so that the data communication processing capacity between the client machine and the server can be improved. You can

[Operation of Traffic Control System] The operation of the traffic control system of this embodiment will be described in detail below with reference to FIGS. 8 and 9. First, the measurement processing operation on the client machine 4 side will be described with reference to the flowchart of FIG.

First, the process of measuring the congestion status with the Web server 1 will be described. The operator gives a service providing instruction (in this case, a download request instruction of a desired Web content) from the instruction unit 4d of the client machine 4 to the Web server 1 (step ST1). here,
At the same time that the instruction is sent to the Web server 1, the control unit 4c of the client machine 4 starts measuring the service providing time by the measuring function 4c ₁ (step ST
2).

Subsequently, W receiving the download request instruction
The eb server 1 sends the Web content to the client machine 4 and provides the client machine 4 with a service. Then, the client machine 4, which has received the Web content (step ST3), stops the measurement at the same time as this reception by the measurement function 4c ₁ , calculates the service provision time, and acquires the measurement value data (step ST4). At that time, the identification information of the server (for example, Web
The IP address of the server 1) is also acquired. The server identification information may be obtained by recognizing the instruction destination at the time of the service providing instruction in step ST1 described above, or may be input by the operator.

The control unit 4c of the client machine 4 that data acquired with the measurement by the measurement function 4c _1, searches the presence or absence of the measurement information management table 4B having a server identification information described above acquired from the management table storage unit 4b (Step ST5) If there is a corresponding measurement information management table 4B, this is read. If there is no corresponding measurement information management table 4B, this measurement process ends. Subsequently, the control unit 4c, the measurement information column 4B ₂ of the measurement information management table 4B read, record information of the acquired measured values (if recording to the measurement information management table 4B is a second time or later The measurement information management table 4B is stored in the management table storage unit 4b (step ST6).

Next, the process of measuring the congestion status with the file server 2 will be described. Even in this case,
Similar to the case of the Web server 1 described above, the measurement process is performed as in steps ST1 to ST6 of the flowchart of FIG. 8, and the measurement information management table 4B is stored in the management table storage unit 4b. In such a case, "Web server 1"
Is replaced with "file server 2", "Web content" is replaced with "file", and "download request instruction" is replaced with "file transfer instruction".

The processing of measuring the congestion status with the DB server 3 is performed as follows. First, as shown in FIG. 8, the operator issues a service providing instruction (in this case, a desired transaction processing instruction) from the instruction unit 4d of the client machine 4 to the DB server 3 (step ST1).
Here, at the same time that the instruction is sent to the DB server 3,
The control unit 4c of the client machine 4 uses the measurement function 4
start measurement of the serving time by c ₁ (step ST2).

Next, the DB server 3 which has received the transaction processing instruction performs the requested processing such as rewriting the data in the database, sends the result to the client machine 4, and services the client machine 4. I will provide a. Then, the client machine 4 receiving the result (step ST3), the measurement function 4c
_{By 1} , the measurement is stopped at the same time as this reception, the service provision time is calculated, and the measured value data is acquired (step S
T4). After that, the measurement process is performed in the same manner as in the case of the Web server 1 described above, and the measurement information management table 4B is stored in the management table storage unit 4b. In such a case, "W
The "eb server 1" is read as "DB server 3".

Here, the measurement interval in the above-mentioned measurement process can be arbitrarily set in advance by the operator.

At the transmission time of the measurement information management table 4B in the client machine 4 performing the measurement as described above, the control unit 4c controls each measurement created or updated from the management table storage unit 4b by the transmission function 4c _2. The information management table 4B is read and the congestion status management device 51
(Step ST7).

Next, the processing operation on the side of the traffic control device 5 will be described with reference to the flowcharts of FIGS. 9 and 10.

First, the line utilization rate monitoring process of the main unit 52 will be described with reference to FIG. The main unit 52 acquires the line usage rate information of the lines between the nodes existing on the network by the line usage rate monitoring function 52f ₁ of the control unit 52f. Control unit 52f of the present embodiment, in order to obtain the line utilization information, for example, be or predetermined time at predetermined intervals, the line utilization information transmission instruction to the nodes on the network by monitoring 52f ₁ Send (step ST11). As a result, each node transmits the line utilization rate information of the line connected to itself to the main unit 52, and the main unit 52 that receives this (step ST12)
Reads the line usage rate monitoring table 52E from the monitoring table storage unit 52e by the monitoring function 52f ₁ of the control unit 52f and inputs the line usage rate information into the line usage rate information column 52E ₁ (step ST13). Thereafter, the line usage rate monitoring table 52E is stored in the monitoring table storage unit 52e by the monitoring function 52f ₁ (step ST
14).

FIG. 1 shows the processing operation of the traffic control device 5 in the state where the line utilization rate information is acquired as described above.
A description will be given based on 0. First, the measurement information management table 4B is received from the client machine 4 (step ST21).
The control unit 51d of the congestion status management device 51 temporarily stores the measurement information management table 4B in the reception table storage unit 51c.
To store.

[0047] Subsequently, the control unit 51d, with its determining function 51d _1, the measurement information management table 4B received
The first congestion situation determination table 51B that matches both the identification information of the client and server is read from the first congestion situation determination table storage unit 51b (step ST22).
Then, the information of the measured value X in the measured value information column 4B ₂ in the measured information management table 4B is compared with the threshold information in the threshold value information column 51B ₃ in the first congestion situation determination table 51B, and the measured value X is the threshold value. It is determined whether or not (step ST23).

When the measured value X does not exceed the threshold value, it is determined whether or not the client machine-server congestion status determination has been performed for all the received measurement information management tables 4B (step ST24). If not, the process returns to step ST22 to read the next first congestion situation determination table 51B, and if so, the process ends. Further, in step ST23, the measurement value X exceeds the threshold value (that is, the corresponding client machine-
If the server is congested), information indicating that the client machine and the server are congested, and identification information of the client machine and the server (herein, each of these information is referred to as "congestion information"). It is stored in the temporary memory (step ST25). Then, it is determined whether or not the client machine-server congestion status determination has been performed for all the received measurement information management tables 4B (step ST26), and if not, the above step S
Returning to T22, the next first congestion situation determination table 51B is read. If it is, the transmission function 51d ₂ transmits the congestion information in the temporary memory to the main device 52 (step ST27).

Next, the control unit 52f of the main unit 52, the through line specific function 52f _2, line identification table 52C line identification table storage unit having the identification information of the client machine and the server received from the congestion state management device 51 The line information existing between the client machine and the server is acquired from the line information column 52C ₃ read from the line 52c (step ST28). Then, the threshold information and the line usage rate information corresponding to the acquired line information are used as the second congestion status determination table 52D and the line usage rate monitoring table 52.
It is read from E and the line (bottleneck line) whose line utilization rate exceeds the threshold value is specified (step ST2).
9).

[0050] Subsequently, the controller 52f may, by its instruction function 52f _3, and instructs the implementation of traffic control to a node of each located at both ends of the bottleneck channel identified by the channel specific function 52f ₂ ( Step ST3
0). As a result, the instructed node secures the above-mentioned bandwidth Y for the bottleneck line.

As described above, by using the traffic control system of the present embodiment, the band used for communicating the data of the degraded service is secured for the line identified as the bottleneck. ,
That is, dynamic traffic control is possible. And
As a result, it is possible to always maintain the quality of service. In addition, when there are a plurality of services to be instructed and the bottleneck lines to be controlled are common to the respective services by weighting as in the above-described Expression 2, each service is adjusted according to the set priority. It is possible to maintain the above service quality. In this way, it is possible to always maintain the quality of service between devices on various networks according to the policy of the network administrator, that is, to suppress the deterioration of the data communication processing capability between the devices.

In the present embodiment described above, only one client machine 4 is illustrated, but the number is not necessarily limited. Further, although three types of servers 1, 2 and 3 are illustrated, the number of servers is not necessarily limited to that number. Furthermore,
Although the traffic control device 5 is configured by the congestion status management device 51 and the main device 52 as an example, the traffic control device 5 is not limited to this mode. For example, the respective configurations of the congestion status management device 51 and the main device 52 may be used. The traffic control device may be integrated.

[0053]

According to the traffic control method and the system thereof according to the present invention, the line identified as the bottleneck is the securing of the band used for communicating the data related to the service whose data communication processing capacity is deteriorated. Since it is possible to perform dynamic traffic control,
The quality of service can always be maintained. As described above, according to the present invention, it is possible to obtain an unprecedented excellent traffic control method and its system that can maintain the quality of service between devices on various networks according to the policy of the network administrator. It will be possible.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of a traffic control system according to the present invention.

FIG. 2 is an explanatory diagram illustrating a measurement information management table according to the present embodiment.

FIG. 3 is an explanatory diagram illustrating a first congestion status determination table according to this embodiment.

4A and 4B are explanatory diagrams illustrating a line identification table according to the present embodiment. FIG. 4A is a table between a client and a Web server, and FIG. 4B is a table between a client and a file server. It is a figure which illustrates the table of.

FIG. 5 is an explanatory diagram illustrating a second congestion status determination table according to the present embodiment.

FIG. 6 is an explanatory diagram illustrating a line usage rate monitoring table according to the present embodiment.

FIG. 7 is an explanatory diagram illustrating the concept of the traffic control system of the present embodiment.

FIG. 8 is a flowchart illustrating a measurement process (data communication process capability monitoring process) of the present embodiment.

FIG. 9 is a flowchart illustrating a line usage rate monitoring process according to the present embodiment.

FIG. 10 is a flowchart illustrating a processing operation of the traffic control device according to the present embodiment.

[Explanation of symbols]

1 Web Server 2 File Server 3 DB Server 4 Client Machine 4b Management Table Storage 4c Client Machine Controller 4c ₁ Measurement Function 4c ₂ Transmission Function 4B Measurement Information Management Table 5 Traffic Control Device 51 Congestion Status Management Device 51b First Congestion Status Judgment table storage unit 51d Congestion status management device control unit 51d ₁ Judgment function 51d ₂ Transmission function 51B First congestion status judgment table 52 Main device 52c Line identification table storage unit 52d Second congestion status judgment table storage unit 52e Monitoring table storage unit 52f Main Unit Control Unit 52f ₁ Monitoring Function 52f ₂ Line Identification Function 52f ₃ Instruction Function 52C Line Identification Table 52D Second Congestion Status Judgment Table 52E Line Utilization Monitoring Table N1, N2, N3, N4, N5, N6 Nodes

Claims (2)

[Claims] 1. A traffic control method at the time of data communication between at least one client and at least one server that provides a service to the client via a network, the method being present between the client and the server. A line usage rate monitoring processing step for monitoring the line usage rate of a plurality of lines, a data communication processing capacity monitoring processing step for monitoring the data communication processing capacity between the client and the server, and when the data communication processing capacity declines. A bottleneck line identification processing step of identifying a line which becomes a bottleneck from the respective lines based on the line utilization monitoring information, and a bottleneck line identified in the bottleneck line identification processing step Traffic having a traffic control processing step for controlling traffic Your way. 2. A traffic control system for data communication between at least one client and at least one server that provides a service to the client via a network, the traffic control system being present between the client and the server. Line usage rate monitoring means for monitoring the line usage rate of a plurality of lines, data communication processing capacity monitoring means for monitoring the data communication processing capacity between the client and server, and the data communication processing capacity monitoring means Bottleneck line identifying means for identifying a line which becomes a bottleneck from among the respective lines based on the line utilization rate monitoring information of the line utilization rate monitoring means when it is determined that the data communication processing capability has decreased, Traffic that controls traffic for the bottleneck line identified by the bottleneck line identification means. Traffic control system, comprising a Ikku control means.