JP2005167769A - Congestion control method and apparatus, program and totaling apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce a load of an incoming side session processing server that is turned into congestion state, by loading a regulation step by step in an outgoing side session processing server in the case of congestion. <P>SOLUTION: The congestion control apparatus monitors that a session processing server becomes a congestion server, first limits, based on a result of the monitor, the number of processes due to the congestion server relating to session setting directly issued from a terminal to the congestion server at a first regulation level via a regulation implementation section using a regulation control section, and then outputs a regulation instruction to the outgoing side session processing server that issues processing relating to session setting to the congestion server, to limit the number of processes to the congestion server at a second regulation level. A regulation request receiving section receives the regulation instruction from the other congestion server to the session processing server of a controlled system, the regulation implementation section limits the number of processes relating to session setting from the session processing server of the controlled system to the congestion server according to the received regulation instruction. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a congestion control technique for an application that performs session setting including an Internet telephone using, for example, VoIP (Voice over Internet Protocol) in an IP network, and more particularly, congestion control suitable for efficiently performing load distribution at the time of congestion. It is about technology.

  As a conventional congestion control technique, a congestion control technique in a public telephone exchange network will be described below. The network configuration of the public telephone exchange network is associated with a region and further has a hierarchical structure. The association with a region means that the subscribers accommodated in a certain subscriber line switch are all the subscribers in a certain region.

  The network configuration has a hierarchical structure in which subscribers are accommodated in subscriber line exchanges, and the subscriber line exchanges are accommodated in higher order exchanges. This is also reflected in the configuration of the telephone number, and the area can be specified by the telephone number. Based on this number, each exchange finally determines which destination exchange should be connected, and connects the session to the next stage to finally establish the session.

  Centralized management is suitable for the control of a network having such characteristics. This is because, since a hierarchical structure is used, in order to apply control in a larger range, it is only necessary to perform control in a higher hierarchy.

  For this reason, the congestion control technique in the public switched telephone network performs centralized management as disclosed in Non-Patent Documents 1 to 3, for example, and each switch notifies the centralized management system when congestion is detected, and the centralized management system The amount of regulation at each switch is decided on the side and notified to each switch.

  In such a public switched telephone network, the destination telephone number, the exchange in which the number is accommodated, and the relationship with the area are clearly shown. From normal AC traffic measurement, it is possible to change from a certain originating exchange to a congested exchange. Since it is easy to estimate the ratio of session volume to that of all switches, it is easy to determine the amount of regulation. For this reason, in the public switched telephone network, it is possible to receive a notification of congestion from the congestion switch and to instruct the other switch for the restriction amount.

  However, an application (program) for setting a session in an IP network, such as an Internet telephone, clearly shows a telephone number, a URI (Uniform Resource Identifier), a SIP (Session Initiation Protocol) server, a location server, and a relationship with a region. Not necessarily. Also, the telephone number and URI may not change even if the user moves or the user moves. For this reason, it is very difficult to determine an appropriate amount of regulation even when measuring AC traffic.

  It is conceivable that the server in a congested state performs a regulation such as automatically rejecting the amount exceeding the processing load. However, there are cases in which multiple servers for session establishment are routed, and in this case, the routed server must perform wasteful processing, and resources are wasted. This waste causes additional congestion. There is.

Shimogaki, Kobayashi, Togawa, "Operation system to prevent telephone network congestion-Advanced traffic control system (TCS-V2)", NTT Technical Journal, 1-8 (1989) 61-65 Shimogaki, Hasegawa, “Enhancement of Traffic Control System (TCS-V2)”, NTT Technical Journal, 3-2 (1991), 40-43 Information and Communication Technology Handbook Editorial Committee, “Information and Communication Technology Handbook”, Ohmsha (1987), 87-89

  The problems to be solved are that it is very difficult to determine an appropriate amount of restriction for an application for setting a session in an IP network such as an Internet telephone, and the processing load on the server side in a congested state is difficult. In the technology that regulates such as automatically rejecting the amount exceeding the limit, if you go through multiple servers for session establishment, the server that goes through has to perform wasteful processing, and resources are wasted. Waste is another point of congestion.

  In order to achieve the above object, in the present invention, the calling side session processing server can be regulated by instructing the calling side session processing server to regulate the calling side session processing server. To. Furthermore, the load at the time of congestion is appropriately distributed by distributed control by changing the regulation value stepwise.

  According to the present invention, at the time of congestion, the originating session processing server can be regulated in stages, and the load on the terminating session processing server that has become congested can be reduced and excessive regulation can be applied. It is possible to apply an appropriate amount of regulation.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration example of a congestion control apparatus according to the present invention, FIG. 2 is a flowchart showing an example of a processing procedure of a congestion control method according to the present invention, and FIG. 3 is a congestion control according to the present invention. FIG. 4 is a block diagram showing a hardware configuration example of a congestion control apparatus according to the present invention.

  As shown in FIG. 4, the congestion control device 1 in FIG. 1 and the congestion control devices 1a and 1b and the session processing servers 33a and 33b and the aggregation device 30 and the terminals 31a, 31b, 32a and 32b in FIG. Central processing unit), a main memory, a display device, an input device, an external storage device, and the like. The external storage device stores programs and data recorded on a storage medium such as a CD-ROM via an optical disk drive device or the like. After being installed inside, the function of each processing unit is executed by reading from the external storage device into the main memory and processing by the CPU.

  In FIG. 4, reference numeral 41 denotes a display device such as a CRT (Cathode Ray Tube) or LCD (Liquid Crystal Display), 42 denotes an input device such as a keyboard or a mouse, and 43 denotes an external storage device such as an HDD (Hard Disk Drive). , 44 is an information processing apparatus that includes a CPU 44a, a main memory 44b, an input / output interface 44c, etc., and performs computer processing. 45 is a CD-ROM (Compact Disc-Read Only Memory) or a program-data recording program or data according to the present invention. An optical disk made up of a DVD (Digital Video Disc / Digital Versatile Disc), etc., 46 is a drive device for reading programs and data recorded on the optical disk 45, and 47 is a communication device made up of a LAN (Local Area Network) card, a modem or the like. is there.

  After the program and data stored in the optical disk 45 are installed in the external storage device 43 by the information processing device 44 via the drive device 46, they are read from the external storage device 43 into the main memory 44b and processed by the CPU 44a. Each function (2 to 9) of the congestion control device 1 shown in FIG.

  In FIG. 3, 33a and 33b are session processing servers that perform processing related to session control in the IP network, and include at least a SIP server, a location server, a gatekeeper, a gateway, and a gateway control device. In each of the session processing servers 33a and 33b, a server whose processing load exceeds a predetermined processing load threshold is called a congestion server.

  In FIG. 3, reference numerals 1a and 1b denote congestion control devices that control congestion generated in the session processing servers 33a and 33b (congestion servers), and 31a, 31b, 32a, and 32b pass through the session processing servers 33a and 33b. A terminal for performing application processing such as an Internet telephone, 30 is information indicating processing load status in a congestion server (session processing servers 33a and 33b) periodically transmitted from the congestion control devices 1a and 1b and a session processing server on the originating side It is a totaling device that receives the restriction instruction information issued to the user via the network and outputs it for display.

  The congestion control devices 1a and 1b in FIG. 3 both have the configuration of the congestion control device 1 shown in FIG. 1, and the congestion control device 1 shown in FIG. 1 includes a processing load measuring unit 2, a load threshold value monitoring unit 3, a regulation control. 4 includes a server 4, an own server-generated load measuring unit 5, another server-generated load measuring unit 6, a regulation request transmitting unit 7, a regulation executing unit 8, and a regulation request receiving unit 9.

  With such a configuration, in the congestion control device 1, the processing load measurement unit 2 measures the processing load of the session processing servers (33 a and 33 b), and the load threshold monitoring unit 3 processes the processing load of the measurement result of the processing load measurement unit 2. Is compared with a predetermined processing load threshold, and the control target session processing server (33a, 33b) monitors whether the processing load exceeds the predetermined processing load threshold and becomes a congestion server.

  Based on the monitoring result that the session processing servers (33a, 33b) in the load threshold value monitoring unit 3 become congestion servers, the regulation control unit 4 first passes the congestion server (session processing server) via the regulation execution unit 8. 33a, 33b) restricts the number of processes by the congestion server (session processing servers 33a, 33b) relating to the session setting directly issued from the terminals (31a, 31b, 32a, 32b) to a predetermined first restriction level. In addition, the congestion server (session processing server 33a) is transmitted to the originating session processing server that has issued processing related to session setting to the congestion server (session processing servers 33a and 33b) via the restriction request transmission unit 7. , 33b), a restriction instruction is issued so as to limit the number of processes to a predetermined second restriction level.

  The restriction request receiving unit 9 receives a restriction instruction from another session processing server (congestion server) for the session processing server (33a, 33b) to be controlled, and according to the received rule instruction via the restriction execution unit 8. Limit the number of processes related to session settings from the session processing server to be controlled to other session processing servers (congestion servers).

  The local server load measurement unit 5 measures the processing load related to the session setting directly issued from the terminal (31a, 31b, 32a, 32b) to the session processing server (33a, 33b) to be controlled, and measures the load from other servers. The unit 6 measures the processing load related to the session setting issued from the other session processing servers to the session processing servers (33a, 33b) to be controlled, and the regulation control unit 4 determines the own server-generated load measuring unit 5 and the other servers. Based on the measurement result in the generation load measuring unit 6, the restriction level (first) for the process related to the session setting directly issued from the terminal (31a, 31b, 32a, 32b) to the session processing server (33a, 33b) to be controlled. Control level) and the session processing servers (33a, 3) to be controlled from other session processing servers. Determining the respective restriction level (second restriction level) to the processing for the session setting emitted in b).

  Further, the restriction control unit 4 measures the time when the processing load in the session processing server (33a, 33b) that has become the congestion server exceeds the processing load threshold, and each time the predetermined time elapses, the first restriction is performed. Session processing server which has a function of strengthening the regulation by gradually increasing the level and the second regulation level, and further becoming a congestion server after strengthening the first regulation level and the second regulation level once The time when the processing load at (33a, 33b) falls below a predetermined restriction release threshold is measured, and the first restriction level and the second restriction level are gradually reduced each time the predetermined time elapses. It has a function.

  In addition, the restriction control unit 4 receives a restriction instruction from another session processing server (congestion server) to the session processing server (33a, 33b) which is a control target and is a congestion server in the restriction request reception unit 9. Then, the restriction level by the restriction instruction from the other session processing server received by the restriction request receiving unit 9 is compared with the restriction level for the congestion server that is the session processing server (33a, 33b) to be controlled. It has a function to select a strict regulation level and use it for regulation control.

  Furthermore, the restriction control unit 4 includes information indicating the processing load state in the session processing servers (33a, 33b) to be controlled that are the congestion servers, and restriction instruction information issued to the originating session processing server; Is periodically transmitted to the aggregation device 30 of FIG. 3 connected to the network.

  In the aggregation device 30 in FIG. 3, information indicating the processing load state in the congestion server and the restriction instruction information issued to the originating session processing server, which are periodically transmitted from the congestion control devices 1a and 1b, are transmitted via the network. The received information is displayed on the display of the display device.

  The congestion control processing procedure according to the present invention by the congestion control device 1 of FIG. 1 will be described below with reference to FIG. Here, a description will be given on the assumption that session connection requests are concentrated on the terminal 32a accommodated in the session processing server 33a in FIG. 3, thereby increasing the processing load of the session processing server 33a and becoming a congestion server.

  As shown in the flowchart of FIG. 2, the congestion control device 1 of FIG. 1 first initializes the counter value j and the timer value t (steps S201 and S202), and then the processing load measurement unit 2 measures the processing load. Is started and measurement is continued (step S203).

  Then, the processing load threshold value monitoring unit 3 determines whether the timer value t has exceeded a predetermined processing load threshold value determination time T (step S204). The load threshold values are compared (step S205), and if not exceeded, it is determined whether the counter value j is positive (step S212). Here, since the counter value j = 0, the process returns to step S202 again.

  In step S205, when the measurement result of the processing load exceeds the processing load threshold value, the regulation control unit 4 executes the processing of steps S206 to S208. That is, it is determined whether the counter value j exceeds the maximum processing load restriction level J (step S206). If it does not exceed, the counter value j is incremented by 1 (step S207), and if it exceeds, no processing is performed. Then, the process proceeds to step S208.

  In step S208, the processing load distribution of both is determined using the measurement results of the own server-generated load measuring unit 5 and the other server measuring unit 6. It is also possible to determine the distribution ratio in advance for the distribution of both of them and to omit the processing in step S208. Thereby, the measurement by the own server generated load measuring unit 5 and the other server measuring unit 6 can be omitted. The regulation at each regulation level is determined in advance for each regulation level as “x (j)%” of the current load.

  When the restriction level is determined, the restriction execution unit 8 first issues a processing request by excluding session requests for other congestion servers from direct processing requests to the congestion server itself, that is, processing requests not passing through other session processing servers. Restrict (step S209).

  There is another congestion server. For the session request for the congestion server, the restriction value instructed by the congestion server is compared with the restriction value in step S209, and the processing request is restricted according to the restriction value that is more stringent. (Step S210).

  Next, the restriction request transmission unit 7 transmits a restriction execution request to another session processing server (step S211). Thereafter, the process returns to step S202 again, the timer value t is initialized, and the process proceeds according to the flowchart.

  In the process of step S205, once the processing load threshold value is exceeded (Y), and then in the processing of step S212, the processing load threshold value is exceeded (Y: counter value j ≦ 0), the load threshold value monitoring unit 3 Then, it is determined whether or not the regulation release threshold value is below (step S213).

  If it is not lower, the process returns to step S202. If it is lower, the regulation level (counter value j) is lowered by one (step S214), and the measurement result by the own server load measuring unit 5 and the other server measuring unit 6 is obtained. Is used to determine the processing load distribution between the two (step S215).

  Also in this case, it is possible to determine the distribution ratio in advance and to omit the processing in step S215. The restriction release threshold is generally set to be smaller than the processing load threshold, but may be the same value.

  Thereafter, the restriction enforcement unit 8 performs processing requests at the current restriction level except for direct processing requests to the congestion server itself, that is, processing requests not via other session processing servers, except for session requests for other congestion servers. Are restricted (step S216).

  There is another congestion server, and for the session request for the congestion server, the restriction value instructed by the congestion server is compared with the restriction value in step S216, and the processing request is restricted according to the restriction value that is more stringent. (Step S217).

  Next, the restriction request transmission unit 7 transmits a restriction execution request to another session processing server at the current restriction level (step S218), and the process returns to step S202.

  When the session processing server other than the congestion server receives the restriction request from the congestion server (by the restriction request receiving unit 9), the session processing server restricts the process to be transmitted to the congestion server at the received restriction level.

  For example, if the regulation is 20%, the regulation implementation unit 8 regulates the regulation when receiving the regulation request from the congestion server. (* 20 ÷ 100) is discarded, and a restriction such as processing of four session processing requests can be considered.

  Each congestion control device periodically transmits the processing load status and the current regulation status to the counting device 30 in FIG. 3 while the congestion state continues if it is determined that the congestion is present. The totaling device 30 records and displays information sent from each congestion control device.

  As described above with reference to FIGS. 1 to 4, in this example, when performing congestion control of an application that sets a session including at least an Internet phone in an IP network, at least a SIP server, a location server, a gatekeeper, and a gateway When the processing load of the session processing server (33a, 33b) that performs processing related to session setting and control including the gateway control device exceeds a predetermined threshold (processing load threshold) and becomes a congestion server, the terminal ( 31a, 31b, 32a, 32b) for the processing related to the session setting directly issued to the congestion server, the congestion server restricts the number of processes at a certain rate, and via a session processing server other than the congestion server For processing related to session settings sent to the congestion server By issuing the instruction to the restriction on the originating session handling server, to allow regulation at calling side, further, by subjecting the regulation value stepwise, to distribute the load during congestion.

  Alternatively, for processing related to session settings issued to a congestion server via a session processing server other than the congestion server, a session processing server other than the congestion server is restricted to limit the number of sessions for the congestion server at a certain regulation rate. The server receiving the instruction limits the number of sessions for the congestion server by a certain restriction ratio according to the instruction, and repeats the restriction ratio until the processing load of the congestion server falls below a predetermined restriction release threshold. The process related to the session setting issued directly from the terminal to the congestion server is changed by changing the value, and an instruction is issued to the session processing server other than the congestion server to limit the process by repeatedly changing the value of the restriction ratio.

  And when the load of a congestion server becomes below a regulation cancellation | release threshold value, it returns in steps in reverse order to the time of restrict | limiting the restrict | limited ratio. In addition, when there are a plurality of congestion servers, when two restrictions, a restriction by the congestion server itself and a restriction from another congestion server, are applied, the restriction with the stricter restriction is followed. Furthermore, the processing load is (1) the ratio of processing related to session setting directly issued from the terminal to the congestion server, and (2) processing related to session setting issued to the congestion server via a session processing server other than the congestion server. Sessions sent to the congestion server through the session processing server other than the congestion server and the restriction value (first restriction level) for the processing related to the session setting issued directly to the congestion server based on the measurement result A restriction value (second restriction level) for processing related to setting is determined.

  As a result, it is possible to regulate in stages at the originating session processing server during congestion, reducing the load on the terminating session processing server that has become congested, and appropriate without over-regulation. It is possible to apply a large amount of regulations.

  In addition, this invention is not limited to the example demonstrated using FIGS. 1-4, In the range which does not deviate from the summary, various changes are possible. For example, in FIG. 3, the congestion control devices 1a and 1b are installed as separate devices outside the session processing servers 33a and 33b, but the congestion control devices 1a and 1b are incorporated in the session processing servers 33a and 33b. The configuration may be such that it functions as one module.

  In this example, “ratio (%)” is used as the regulation value (first regulation level, second regulation level), but the present invention is not limited to this.

  Further, in this example, the computer configuration example of FIG. 2 is shown, but a computer configuration without a keyboard or an optical disk drive may be used. In this example, an optical disk is used as a recording medium, but an FD (Flexible Disk) or the like may be used as a recording medium. As for the program installation, the program may be downloaded and installed via a network via a communication device.

It is a block diagram which shows the structural example of the congestion control apparatus concerning this invention. It is a flowchart which shows the example of a process sequence of the congestion control method concerning this invention. It is a block diagram which shows the example of a network structure which provided the congestion control apparatus and aggregation apparatus concerning this invention. It is a block diagram which shows the hardware structural example of the congestion control apparatus concerning this invention.

Explanation of symbols

  1, 1a, 1b: Congestion control device, 2: processing load measurement unit, 3: load threshold value monitoring unit, 4: regulation control unit, 5: own server source load measurement unit, 6: other server source load measurement unit, 7: Regulation request sending unit, 8: Regulation implementation unit, 9: Regulation request receiving unit, 30: Aggregation device, 31a, 31b, 32a, 32b: Terminal, 33a, 33b: Session processing server, 41: Display device, 42: Input device 43: external storage device, 44: information processing device, 44a: CPU, 44b: main memory, 44c: input / output interface, 45: optical disk, 46: drive device, 47: communication device.

Claims (16)

A congestion control method by a session processing server that performs processing related to session control in an IP network,
When the processing load exceeds a predetermined processing load threshold, processing related to session settings directly issued from the terminal to the server as a congestion server is processed at the first restriction level determined in advance by the server. For the processing related to session settings issued to the local server via other session processing servers, the number of processing to the local server is determined in advance from the local server to the originating session processing server. A procedure for issuing regulatory instructions to limit at a second regulatory level;
A congestion control method comprising: a step of restricting the number of processes related to session setting for another session processing server according to the restriction instruction when receiving the restriction instruction from another session processing server. The congestion control method according to claim 1,
Each time when the processing load exceeds the processing load threshold exceeds a predetermined time,
A congestion control method comprising a step of strengthening the restriction by gradually increasing the first restriction level and the second restriction level. The congestion control method according to claim 2,
After the first restriction level and the second restriction level are strengthened once, the first restriction level and the time each time when the processing load falls below a predetermined restriction release threshold exceeds a predetermined time. A congestion control method comprising a step of gradually weakening the second restriction level. A congestion control method according to any one of claims 1 to 3,
When the restriction instruction is received from another session processing server during the restriction process as the congestion server, the restriction level according to the restriction instruction from the other session processing server is compared with the restriction level of the own server, whichever is severe. A congestion control method comprising a procedure of selecting and using a restriction level. A congestion control method according to any one of claims 1 to 4, wherein
A procedure for measuring the rate of processing related to session setting issued directly from the terminal and the rate of processing related to session setting issued from the other session processing server;
And a procedure for determining each of a restriction level for processing relating to session setting issued directly from the terminal and a restriction level relating to processing relating to session setting issued from the other session processing server based on the measurement result. A congestion control method characterized by the above. A congestion control method according to any one of claims 1 to 5,
A congestion control method characterized in that a ratio is used as the first restriction level and the second restriction level, and the number of processes is limited by the ratio. A congestion control method according to any one of claims 1 to 6, comprising:
While continuing the process as the congestion server, periodically,
Congestion control characterized by having a procedure for transmitting information indicating the state of the processing load in its own server and restriction instruction information issued to the session processing server on the originating side to a network-connected aggregation device Method. A congestion control device that controls congestion generated in a session processing server that performs processing related to session control in an IP network,
Monitoring means for monitoring that the processing load of the session processing server to be controlled exceeds a predetermined processing load threshold and becomes a congestion server;
Means for limiting the number of processing by the congestion server relating to session settings directly issued from the terminal to the congestion server at a predetermined first restriction level;
Means for issuing a restriction instruction to the calling-side session processing server that has issued the process related to session setting to the congestion server so as to limit the number of processes to the congestion server at a predetermined second restriction level;
Means for receiving the restriction instruction from another session processing server for the session processing server to be controlled;
A congestion control apparatus comprising: means for limiting the number of processes related to session setting for the other session processing server from the control target session processing server according to the received rule instruction. The congestion control device according to claim 8, comprising:
A means for strengthening the regulation by gradually increasing the first regulation level and the second regulation level every time a predetermined time elapses when the processing load on the congestion server exceeds the processing load threshold. A congestion control device characterized by the above. The congestion control device according to claim 9, comprising:
After strengthening the first restriction level and the second restriction level once, each time the predetermined time elapses when the processing load on the congestion server falls below a predetermined restriction release threshold, A congestion control apparatus comprising means for gradually weakening the restriction level and the second restriction level. The congestion control device according to any one of claims 8 to 10,
When the restriction instruction from the other session processing server to the congestion server is received, the restriction level according to the restriction instruction from the other session processing server is compared with the restriction level in the congestion server, and any strict restriction level Congestion control apparatus comprising means for selecting and using The congestion control device according to any one of claims 8 to 11,
Means for measuring a ratio of processing relating to session setting issued directly from the terminal to the congestion server, and a proportion of processing relating to session setting issued from the other session processing server to the congestion server;
Based on the measurement result by the means, the restriction level for the process related to session setting issued directly from the terminal to the congestion server, and the restriction level for the process related to session setting issued from the other session processing server to the congestion server And a means for determining each of the congestion control devices. The congestion control device according to any one of claims 8 to 12,
The means for restricting at the first restriction level and the means for issuing a restriction instruction to restrict at the second restriction level use a ratio as the first restriction level and the second restriction level, and at the ratio A congestion control device that limits the number of processes. The congestion control device according to any one of claims 8 to 13,
It has means for periodically transmitting information representing the state of the processing load in the congestion server and restriction instruction information issued to the originating session processing server to a totaling device connected to the network. Congestion control device.   The program for functioning a computer as each means in the congestion control apparatus in any one of Claims 8-14.   The information indicating the state of the processing load in the congestion server and the restriction instruction information issued to the originating session processing server transmitted periodically from the congestion control device according to claim 14 are received via the network. And a means for displaying and outputting the received information.

Images