JP5941435B2 - Congestion detection device, congestion detection method, congestion detection program, and program recording medium - Google Patents

Description

  The present invention relates to a congestion detection apparatus, a congestion detection method, a congestion detection program, and a program recording for detecting congestion occurring in another IP network in a communication system that provides communication services via a plurality of IP (Internet Protocol) networks. It relates to the medium.

  Currently, various types of communication services are provided to clients by connecting many networks constructed by a plurality of operators. Since the networks constructed by each business operator have different processing capabilities, if traffic in a network with a high processing capability flows into a network with a low processing capability, congestion will occur.

  When congestion occurs in a certain network, it is necessary to coordinate with other connected networks in order to limit the traffic from other networks flowing into that network and to calm down the congestion state at an early stage. . Conventionally, a maintenance person of a network in which congestion has occurred performs cooperation (hereinafter referred to as “maintenance person cooperation”) by manual work (e-mail, telephone, etc.) to a maintenance person of another network connected to the network. ing.

  On the other hand, in order to detect a failure on the network at an early stage, a technique is disclosed in which control signals are collected for VoIP (Voice over Internet Protocol) in the application layer to isolate the failure (see Patent Document 1). . Also, a technique for collecting log information of SIP (Session Initiation Protocol) signals is disclosed in order to efficiently analyze log information (see Patent Document 2).

JP 2004-248192 A JP 2011-166582 A

  However, the technology described in Patent Document 1 or Patent Document 2 is based on the premise of collecting information (control signal and log information) in the network to which the device belongs, and collects information across networks with different management units. I can't. Therefore, with the existing technology, it has been impossible to grasp the traffic situation between networks and perform appropriate congestion control without maintenance personnel cooperation.

  On the other hand, the environment (terminals, services, etc.) surrounding the network is changing greatly. For example, many operators (OTT: Over The Top) entering the M2M (Machine to Machine) service are expected. We are changing a lot.

  When a large number of OTTs enter and M2M services are started, the traffic from one network to the OTT will cause unexpected congestion due to the inability to fully understand the impact of traffic for new services on other networks. May cause.

FIG. 5 is a diagram for explaining the occurrence of network congestion associated with the M2M service. In FIG. 5, it is assumed that a large amount of traffic is transmitted from the client 4 accommodated in the IPNW_A to the service provided by the OTT 5 connected to the IPNW_C via the IPNW_A, IPNW_B,. 5 # 1).
Various types of OTT services have been proposed. A specific example is an automobile emergency call system. In the automobile emergency call system, when a car registered in the system breaks down due to a large-scale disaster (earthquake, tsunami, etc.), its emergency information (the emergency situation and vehicle location information, etc.) is automatically reported to the server. To do. In such a case, it is assumed that the traffic directed to the OTT server in the area increases rapidly.

  Subsequently, in IPNW_C connected to OTT5, it is assumed that congestion has occurred due to the concentration of traffic from IPNW_A to OTT5 (# 2 in FIG. 5).

  In the prior art, when congestion occurs, congestion regulation is performed between adjacent networks using maintenance personnel cooperation. However, as shown in FIG. 5, when the traffic is spread over a plurality of networks, a request from the IPNW_C maintenance person (maintenance person terminal 3 (3C)) to the IPNW_A maintenance person (maintenance person terminal 3 (3A)) It cannot be performed promptly (# 3 in FIG. 5). In this case, the maintenance person collaborates with the network (IPNW_C) where the congestion has occurred and the other network connected to the network, and further, the congestion restriction request is issued to the maintenance person of IPNW_A via the connected network sequentially. Will be made.

  Therefore, in IPNW_A, the congestion occurrence information in IPNW_C cannot be acquired at an early stage, and congestion control is delayed, resulting in a decrease in service quality (# 4 in FIG. 5).

  As described above, in order to quickly calm down congestion, it is necessary to perform congestion control promptly. However, in cases such as spanning multiple networks, it is necessary to perform maintenance operator cooperation multiple times in sequence, so congestion Regulations may not be made promptly. Therefore, it is necessary to quickly grasp congestion not only in the own network but also in other networks.

  The present invention has been made in view of such a background, and the present invention provides a congestion detection device, a congestion detection method, a congestion detection program, and a program recording medium that can detect congestion in other networks at an early stage. The task is to do.

  In order to solve the above-described problem, the invention according to claim 1 is directed to a plurality of IP (Internet Protocol) networks connected to provide a communication service, and the own network that is the IP network to be controlled by itself. Congestion detection of a congestion control system, comprising: a congestion control device that executes congestion control of a network; and a congestion detection device that is connected to the congestion control device and detects congestion of another network other than the IP network. In the POI (Point Of Interface) between the own network and the other network adjacent to the own network, a control signal directed from the own network to each of the other networks, and the control signal A message collection unit that collects response signals to the message, the number of the collected control signals, and a message classified based on a response code indicating a response content of the response signal. The number of response signals for each element is totaled as the latest message number information for each other network as a destination of the control signal and every predetermined time, and stored in a storage unit, the message number totaling unit, Based on the latest message number information, the ratio of the number of response signals for each message element to the total number of messages, which is the number of control signals, is calculated for each other network, and the calculated ratio in the other network is A congestion determination unit that determines that congestion has occurred with respect to the other network that satisfies a first congestion determination condition that determines occurrence of congestion in comparison with a predetermined threshold, and identification of the other network that has determined that congestion has occurred A congestion detection apparatus comprising: a congestion information transmission unit that generates congestion determination information including information and transmits the congestion determination information to the congestion control apparatus.

  According to a third aspect of the present invention, there is provided a plurality of IP networks connected to provide a communication service, and a congestion control device that executes congestion control of the own network that is the IP network to be controlled by the network. A congestion detection method of the congestion detection device in a congestion control system, comprising: a congestion detection device connected to the congestion control device and detecting congestion of another network which is the other IP network than itself, In the POI between the own network and the other network adjacent to the own network, the congestion detection device sends a control signal directed from the own network to each of the other networks and a response signal to the control signal. A response signal for each message element classified based on the collecting step, the number of the collected control signals, and the response code indicating the response content of the response signal The number is counted as the latest message number information for each other network that is the destination of the control signal and every predetermined time, and stored in a storage unit, and the control is performed based on the latest message number information. Calculating a ratio of the number of response signals for each message element with respect to the total number of messages, which is the number of signals, for each other network, and comparing the calculated ratio for the other network with a predetermined threshold value. A step of determining that congestion has occurred for the other network satisfying a first congestion determination condition for determining occurrence of congestion, and generating congestion determination information including identification information of the other network determined to have generated congestion, And a step of transmitting to the congestion control device.

In this way, the congestion detection device transmits a control signal directed from the own network to each other network and a response signal to the control signal in the POI between the own network and another network adjacent to the own network. Collect every net. Then, the congestion detection apparatus performs the congestion determination by comparing the ratio of the number of response signals for each message element in the whole with a predetermined threshold value.
Therefore, according to the present invention, even when the maintenance person cooperation is not quickly performed due to a plurality of IP networks, by analyzing the control signal and the response signal directed to another network, It is possible to detect congestion in other networks at an early stage.

  In the invention according to claim 2, the number of control signals and the number of message elements for each of the other networks stored in the storage unit as the latest message number information for each predetermined time are stored in the storage unit. The average of the most recent predetermined number of times for each response signal is calculated and stored in the storage unit as normal message number information, and the congestion determination unit is based on the normal message number information. The ratio of the number of response signals for each message element to the total number of messages that is the number of control signals is calculated for each of the other networks, and a second congestion determination is performed instead of the first congestion determination condition. As a condition, the ratio of the number of response signals of the message element calculated from the message number information in the normal time, and the response signal of the message element calculated from the latest message number information The congestion detection apparatus according to claim 1, wherein the congestion detection apparatus determines that congestion has occurred in the other network when an index based on a magnitude of the difference exceeds a predetermined threshold. did.

  According to a fourth aspect of the present invention, the congestion detection device further includes the number of the control signals for each of the other networks stored in the storage unit as the latest message number information for each predetermined time. For the number of response signals for each message element, calculate the average for the most recent predetermined number of times, and store it in the storage unit as normal message number information, based on the normal message number information, Calculating a ratio of the number of response signals for each message element to the total number of messages, which is the number of control signals, for each of the other networks; and determining occurrence of congestion using the first congestion determination condition Instead of the step, as the second congestion judgment condition, the ratio of the number of response signals of the message element calculated from the message number information in the normal time and the latest message Comparing the ratio of the number of response signals of the message element calculated from the number information, and determining that congestion has occurred in the other network when an index based on the magnitude of the difference exceeds a predetermined threshold; The congestion detection method according to claim 3, wherein:

  As described above, the congestion detection apparatus compares the ratio of the number of response signals for each message element with respect to the latest message number information and the normal message number information, so that an index based on the difference is predetermined. If the threshold is exceeded, it can be determined that congestion has occurred in another network. Therefore, according to the present invention, congestion in another network can be detected at an early stage.

  The invention according to claim 5 is a congestion detection program for causing a computer to execute the congestion detection method according to claim 3 or claim 4.

  According to such a congestion detection program, the congestion detection method according to claim 3 or claim 4 can be executed by a general computer.

  The invention described in claim 6 is a computer-readable recording medium on which the congestion detection program according to claim 5 is recorded.

  By doing so, the congestion detection program according to claim 5 can be recorded on a recording medium and read by a general computer.

  ADVANTAGE OF THE INVENTION According to this invention, the congestion detection apparatus, the congestion detection method, the congestion detection program, and program recording medium which detect the congestion in another network at an early stage can be provided.

It is a figure for demonstrating the structure and process outline | summary of a congestion control system containing the congestion detection apparatus which concerns on this embodiment. It is a functional block diagram which shows the structural example of the congestion detection apparatus which concerns on this embodiment. It is a figure which shows the example which compared the ratio of each response code (message element) to the whole with "normal" and "latest". It is a flowchart which shows the flow of a process of the congestion detection apparatus which concerns on this embodiment. It is a figure for demonstrating the congestion occurrence of the network accompanying M2M service.

  Next, a congestion control system 100 including the congestion detection device 1 in a mode for carrying out the present invention (hereinafter referred to as “this embodiment”) will be described.

<System configuration and processing overview>
Initially, with reference to FIG. 1, the structure and process outline | summary of the congestion control system 100 containing the congestion detection apparatus 1 which concern on this embodiment are demonstrated.
The congestion detection device 1 provided in the congestion control system 100 according to the present embodiment includes an IP network to be managed (hereinafter sometimes referred to as “own network”) and an IP network other than the own network (hereinafter referred to as “self network”). By collecting and analyzing a control signal (“SIP message” to be described later) transmitted / received to / from the “other network”, the congestion state in the other network is detected. Then, the congestion detection device 1 transmits the detection information (“congestion determination information” to be described later) to the congestion control device 2, so that the congestion detection device 1 is congested earlier in the local network than in the case of performing the maintenance person cooperation which is the conventional technology. Execute control.

  As shown in FIG. 1, the congestion control system 100 according to the present embodiment has its own IP network (indicated as “IPNW_A (own network A)” in FIG. 1) and other than itself as the object of congestion control. A plurality of networks including IP networks (in FIG. 1, described as “IPNW_B (other network B)” and “IPNW_C (other network C)”), and a congestion detection device 1 that detects congestion in other IP networks, And a congestion control device 2 that executes congestion control of IPNW_A (own network A).

  For IPNW_A, IPNW_B, IPNW_C, etc., the operators managing each network may be a single operator or a plurality of operators. For example, a fixed telephone network, mobile (mobile) ) Each network is managed independently by a maintenance person (maintenance person terminal 3) like a telephone network.

  The congestion control device 2 is connected to the maintenance person terminal 3 (3A) that manages its own IP network (IPNW_A (own network A)) and the congestion detection device 1, and indicates the indication information or congestion from the maintenance person terminal 3 (3A). The congestion determination information from the detection device 1 is received, and congestion control is executed in its own IP network (IPNW_A (own network A)).

  The congestion detection apparatus 1 is a POI (Point Of) which is an interconnection point between its own IP network (IPNW_A (own network A)) and another IP network (IPNW_B (other network B)) adjacent to its own network A. Interface) collects a control signal message (SIP message) from the own network to each of the other networks and a response message (response signal) to the message (S10: signal collection). Then, the congestion detection device 1 analyzes the contents of the messages collected for each other network (for example, “200 OK”, “503 Service Unavailable”, etc. in the case of SIP), and analyzes a predetermined condition (“congestion determination” described later). Congestion determination is performed under the condition “)” (S11: congestion determination). The congestion detection device 1 transmits congestion determination information to the congestion control device 2 when it is determined that congestion has occurred in any of the other networks and satisfies a predetermined condition (S12: congestion determination information transmission).

  The congestion control device 2 receives the congestion determination information from the congestion detection device 1 and, in its own IP network (IPNW_A (own network A)), other IP networks (for example, IPNW_C ( Congestion control for signal transmission for other network C)) is executed (S13: congestion control).

  By doing in this way, the congestion control system 100 including the congestion detection device 1 according to the present embodiment allows the congestion detection device 1 to detect congestion in other IP networks at an early stage. Compared to cooperation, congestion regulation can be performed quickly, and deterioration in service quality can be prevented.

<Configuration of congestion detection device>
Next, the congestion detection apparatus 1 according to the present embodiment will be specifically described.
FIG. 2 is a functional block diagram illustrating a configuration example of the congestion detection apparatus 1 according to the present embodiment.
The congestion detection device 1 collects a control signal message from its own network to another network and a response message (response signal) to the message in the POI (see FIG. 1), and detects a congestion state in the other network. It is a device to perform.
As shown in FIG. 2, the congestion detection device 1 includes a control unit 10, an input / output unit 20, and a storage unit 30.

  The input / output unit 20 acquires a message from the POI and inputs / outputs information to / from the congestion control device 2 and the like. The input / output unit 20 also performs input / output between a communication interface that transmits and receives information via a communication line and an input unit such as a keyboard (not shown), an output unit such as a monitor, and the like. It consists of.

The control unit 10 controls the entire congestion detection apparatus 1 and includes a message collection unit 11, a message count totaling unit 12, a congestion determination unit 13, and a congestion information transmission unit 14.
The control unit 10 is realized by, for example, a program (congestion detection program) stored in the storage unit 30 being developed and executed in a RAM (Random Access Memory) by a CPU (Central Processing Unit) (not shown). Is done.

In the POI between the own network and another network adjacent to the own network, the message collection unit 11 sends a control signal message from the own network to each of the other networks and a response message (response signal) to the message. collect.
For example, when SIP is used as the communication control protocol, the message collection unit 11 moves from its own network (IPNW_A (own network A) in the example of FIG. 1) to another network (IPNW_C (other network C) in the example of FIG. 1). Collect INVITE messages sent to. Further, the message collection unit 11 collects response messages from the other network (IPNW_C (other network C)) to the own network (IPNW_A (own network A)) in response to the INVITE message.

  The session is established when there is a response message indicating communication such as “200 OK” with respect to “INVITE” in the SIP message, so that the network is in a congested state or congestion control. It is presumed that the session cannot be established during the execution of Therefore, by analyzing the number of INVITE messages and the number of response messages by the message count totaling unit 12 and the congestion determining unit 13 described below, it is possible to detect the congestion state of other networks.

The message count totaling unit 12 totals the control signals (messages) collected by the message collecting unit 11 at predetermined time intervals for each other network (other IP network).
Specifically, the message count totaling unit 12 counts the number of INVITE messages from its own network to other networks for each other destination network. Further, the message count totaling unit 12 counts the number of response messages from other networks toward the own network by response code (code value). For example, the message count totaling unit 12 classifies the response messages based on the response code as follows, for example, and counts the number of response messages. In the following description, each response message classification may be referred to as a message element. This message element (classification for collecting messages based on response codes) is arbitrarily set by a maintenance person or the like.

-“200 OK”: Successful request.
“2xx Other”: Indicates a response code for all successful responses (in the 200s) (excluding “200 OK”).
“404 Not Found”: No match with the request URI.
“4xx Other”: Indicates a response code of a request error response (400 range) in general (excluding “404 Not Found”).
・ "503 Service Unavailable" ... Service unavailable.
“5xx other” indicates a general response code of the server error response (500 range) (excluding “503 Service Unavailable”).
・ "Other" ... Response codes other than the above. For example, the 300s and 600s.
-“No response”: When there is no response to INVITE within a predetermined time.

  For example, the message count totaling unit 12 counts the number of response messages for each message element classified as described above at a predetermined time interval for each other network that is the destination of the INVITE message, and the count result (the latest count result). ) Is stored in the latest message storage unit 312 of the other network message storage unit 31 (31B,..., 31N) provided for each other network in the storage unit 30 as the latest message number information.

  In addition, the message count totaling unit 12 determines that the message count information for each response code (message element) collected at a predetermined time interval is the last past X times (predetermined times) (abnormal time (congestion occurrence)). The average message is excluded), and the normal message storage unit 311 of the other network message storage unit 31 (31B,..., 31N) provided for each other network in the storage unit 30 stores the normal message count information. Remember.

  For example, the congestion determination unit 13 determines whether or not a predetermined congestion determination condition shown below is satisfied based on the message number information that the message count totaling unit 12 counts for each response code (message element) of the response message for each other network. If the congestion determination condition is satisfied, it is determined that congestion has occurred in the other network (other IP network).

In the following description, the congestion detection device 1 will be described as determining congestion using message number information based on a response code (code value) of a response message to the INVITE message collected by the message count totaling unit 12. However, the congestion detection apparatus 1 uses (1) the cause value of the response message for INVITE, and (2) the SDP (included in the response message for INVITE) as information used for congestion determination, in addition to the response code (code value) of the response message. Session Description Protocol) may be used.
There are cases where congestion occurrence information is set according to the cause value of the response message, or congestion guidance information on other networks using SDP voice media may be included in the response message. In addition to the congestion determination based on the congestion determination unit 13, the congestion determination unit 13 can also perform congestion determination of another network. Specifically, for example, in the congestion determination based on the response code, even if it is not determined that congestion occurs, a response message including a cause value indicating the occurrence of congestion exceeds a predetermined ratio with respect to all response messages. If a response message including SDP audio media that indicates congestion occurrence (cannot communicate) is detected exceeding a predetermined ratio with respect to all response messages, it is determined that congestion has occurred. To do.

<< Congestion criteria >>
Hereinafter, conditions (a) to (c) in the case where the congestion determination is performed using the message number information collected based on the response code (code value) of the response message will be described as an example of the congestion determination condition.
(A) When the latest number of messages by response code (message element) is compared with the number of normal messages by response code (message element), and the difference index (D) exceeds the threshold T ₁ Judged as congestion.
(B) if the ratio of "200 OK" in the reply message is below the threshold value _{T 2,} it determines the congestion.
(C) of the response message, (set in advance, for example, "404 Not Found" and "503 Service Unavailable") message interruption of fact determines if the ratio of exceeds the threshold value T _3, and congestion.
Note that the congestion determination condition (a) corresponds to the “second congestion determination condition” in the claims, and the congestion determination conditions (b) and (c) correspond to the “first congestion determination condition” in the claims. To do.
Subsequently, the congestion determination conditions (a) to (c) will be described in detail.

The congestion determination unit 13 performs a process described below when executing the congestion determination.
The congestion determination unit 13 refers to the other network message storage unit 31 for each other network stored in the storage unit 30, and each response code (message element) is obtained from the message number information stored in the normal message storage unit 311. While calculating the ratio to the whole, the ratio to which each response code (message element) occupies the whole is calculated from the message number information stored in the latest message storage unit 312. FIG. 3 is a diagram illustrating the ratio of each response code (message element) to “normal” and “latest”.

(Congestion criteria (a))
In the congestion judgment condition (a), for each response code (message element), the number of messages of the response code (message element) occupies the whole (the number of control signal messages (INVITE)) and the latest The difference index (D) is calculated by comparing with the ratio of. Then, the congestion determination unit 13 determines if the index difference (D) exceeds the threshold value T _1, that congestion has occurred. In order to calculate the difference index (D), the congestion determination unit 13 performs the following setting for each response code (message element).
[A1]... Regarding “200 OK”, d ₂₀₀ = m when the latest rate is reduced by x% or more compared to the normal rate.
[A2] ... With respect to message elements other than “200 OK”, when the latest rate is ± y% or more compared to the normal rate, that is, the magnitude of the difference between the normal rate and the latest rate Is equal to or greater than a predetermined ratio, d _xxx = n (xxx indicates a code value). For example, when the response code is in the 200s or the like and the message indicates a successful response, d _2xx = n when the latest rate has decreased by y% or more compared to the normal rate. If the response code is 400 or 500 and the message indicates an error response, d _4xx = n, d _5xx when the latest rate has increased by y% or more compared to the normal rate. = N etc.
Then, as in (Equation 1) below, it is determined that congestion has occurred when the index (D) based on the difference between the normal rate and the latest rate for each response code exceeds a predetermined threshold value. To do.
Index (D) = d ₂₀₀ +... + D _xxx +...> Threshold T ₁ (Expression 1)

(Congestion judgment condition (b))
In the congestion determination condition (b), the congestion determination unit 13 determines that, based on the latest message number information stored in the latest message storage unit 312, the ratio of “200 OK” in the response message falls below the threshold T _2. It is determined that congestion has occurred. For example, as shown in FIG. 3, the ratio of “200 OK” of “latest” is decreased compared to “normal”, and the ratio of “200 OK” in the entire latest message number information is a predetermined value (threshold value). If it falls below (T ₂ ), it is determined that congestion has occurred in the other network (other IP network).

(Congestion criteria (c))
In congestion determination condition (c), the congestion determining unit 13 determines that if the proportion of the non-delivery of a message indicating that stored in the most recent message storing portion 312 exceeds the threshold value T _3, congestion occurs. For example, “404 Not Found”, “503 Service Unavailable”, or the like is set in advance as the message indicating the disconnection. For example, as shown in FIG. 3, the ratio of “503 Service Unavailable” of “latest” is increased compared to “normal”, and the ratio of “503 Service Unavailable” in the entire latest message number information is a predetermined value. When the value exceeds (threshold value T ₃ ), it is determined that congestion has occurred in the other network (other IP network).

  In the congestion determination unit 13, when at least one of the congestion determination conditions (a) to (c) is set and the congestion determination condition is satisfied, it is determined that congestion has occurred in the corresponding other network. To do. Also, congestion is generated when two or more sets of the congestion determination conditions (a) to (c) are set in the congestion determination unit 13 and any one of the congestion determination conditions is satisfied. May be determined.

Furthermore, the congestion determination unit 13 may combine the congestion determination conditions (a) to (c) after adding a congestion generation preliminary condition that is a preliminary congestion determination condition.
For example, in the congestion determination condition (b), a threshold T _2b (> T ₂ ) is provided as a congestion generation preliminary condition in which the determination condition for the occurrence of congestion is less than the threshold T ₂ . Then, the congestion determination unit 13 uses the congestion determination condition (a) when the ratio of “200 OK” in the entire latest message number information decreases and falls below the threshold value T _2b of the congestion occurrence precondition. Execute congestion judgment. Even if the congestion determination condition (a) is not satisfied in this congestion determination, the congestion determination is again made by determining whether or not at least one of the congestion determination conditions (a) to (c) is satisfied after a predetermined time has elapsed. You may do it.
Further, in the congestion determination condition (c), a threshold T _3c (<T ₃ ) is provided as a congestion generation preliminary condition in which the determination condition for the occurrence of congestion is less than the threshold T ₃ . Then, the congestion determination unit 13 uses the congestion determination condition (a) when the ratio of “503 Service Unavailable” in the entire latest message number information increases and exceeds the threshold value T _3c of the congestion occurrence precondition. Perform the original congestion determination. Even if the congestion determination condition (a) is not satisfied in the congestion determination, the congestion determination unit 13 again satisfies at least one of the congestion determination conditions (a) to (c) after a predetermined time has elapsed. The congestion may be determined based on whether or not.
In this way, the congestion determination condition (a) may be executed only when the congestion occurrence preliminary condition is satisfied. Since the determination of the congestion determination condition (a) is executed, all response codes (messages) are always used. Since it is not necessary to perform a comparison between the normal state and the latest of the element), the processing load of the congestion detection apparatus 1 can be reduced. In addition, the congestion determination can be performed more accurately by considering parameters of a plurality of response codes (message elements) as compared with the case where only the congestion determination conditions (b) and (c) are used.

  When the congestion determination unit 13 determines that the congestion determination condition is satisfied, the congestion information transmission unit 14 generates congestion determination information including identification information of another network (other IP network) that satisfies the congestion determination condition, Transmit to the congestion control device 2. The congestion information transmission unit 14 transmits the congestion determination information to the congestion control device 2 and also transmits the congestion determination information to the maintenance person terminal 3 (3A) (see FIG. 1) of the own network. A person may be notified of the occurrence of congestion.

  The storage unit 30 includes storage means such as a hard disk, a flash memory, and a RAM. For example, the other network message storage unit 31 (31B, ..., 31N) is stored for each other network (other IP network). .

<Process flow>
Next, a processing flow of the congestion detection apparatus 1 according to the present embodiment will be described.
FIG. 4 is a flowchart showing a processing flow of the congestion detection apparatus 1 according to the present embodiment.
In addition, the congestion detection apparatus 1 is directed from the own network to the other network in the POI (see FIG. 1) which is an interconnection point between the own IP network (own network) and another IP network (other network). A control signal message and a response message to the message are collected at a predetermined time interval, and the message count totaling unit 12 averages the past X times and is provided in the storage unit 30 for each other network. Further, it is assumed that the normal message storage unit 311 of the other network message storage unit 31 (31B,..., 31N) has already been stored as initial information (normal message number information).

First, the message collection unit 11 of the congestion detection device 1 is connected to each other network from its own network in a POI that is an interconnection point between its own IP network (own network) and another adjacent IP network (other network). A control signal message and a response message to the message are collected (step S20).
For example, the message collection unit 11 collects INVITE messages from its own network to other networks, and collects response messages from other networks to its own network.

  Next, the message count totaling unit 12 determines whether or not a predetermined time has passed by a timer (not shown) such as a timer (step S21), and if the predetermined time has not passed (step S21 → No), continue message collection. On the other hand, when the predetermined time has elapsed (step S21 → Yes), the message count totaling unit 12 proceeds to the next step S22.

  In step S22, the message count totaling unit 12 analyzes the message collected by the message collecting unit 11, and sets a response code (message element) set in advance for each other network (other IP network) that is the destination of the INVITE message. ) Count the number of each message. Then, the message count totaling unit 12 stores the totaled message count for each response code (message element) in the latest message storage unit 312 of each other network message storage unit 31 as the latest message count information.

  Subsequently, the congestion determination unit 13 includes the latest message number information stored in the latest message storage unit 312, the normal message number information stored in the normal message storage unit 311 (average value for the past X times), and Are compared (step S23).

  The congestion determination unit 13 determines whether or not a predetermined congestion determination condition (for example, at least one of the above-described (a) to (c)) is satisfied (step S24). If the predetermined congestion determination condition is not satisfied (step S24 → No), the process proceeds to step S26. On the other hand, when a predetermined determination condition is satisfied (step S24 → Yes), the process proceeds to the next step S25.

  In step S <b> 25, the congestion information transmission unit 14 generates congestion determination information including identification information of another network (another IP network) that satisfies the congestion determination condition, and transmits the congestion determination information to the congestion control device 2. Thereby, the congestion control in the own network by the congestion control device 2 is executed.

  On the other hand, in step S26, the message count totaling unit 12 determines that the congestion determination unit 13 does not satisfy the predetermined congestion determination condition in step S24 (step S24 → No). The X message number information is averaged to generate new normal message number information, and the normal message number information stored in the normal message storage unit 311 is updated.

When the process of step S25 or S26 is finished, in step S27, the message count totaling unit 12 resets (initializes) the time counting unit determined in step S21, and in step S22, sets a message counter for counting each message. Reset (initialize).
Then, the congestion detection apparatus 1 returns to step S20 and continues the processing in order to start processing at the next predetermined time interval.

  As described above, according to the congestion detection apparatus 1, the congestion detection method, the congestion detection program, and the program recording medium according to the present embodiment, the control signal message from the own network to the other network and the response to the message By collecting messages, it is possible to determine congestion in another network and detect the occurrence of congestion at an early stage. Therefore, by transmitting congestion determination information indicating that congestion has occurred in the other network from the congestion detection apparatus 1 to the congestion control apparatus 2, it is possible to speed up congestion calming in the own network and prevent deterioration in service quality. .

DESCRIPTION OF SYMBOLS 1 Congestion detection apparatus 2 Congestion control apparatus 3 Maintenance person terminal 10 Control part 11 Message collection part 12 Message count total part 13 Congestion judgment part 14 Congestion information transmission part 20 Input / output part 30 Storage part 31 Other network message storage part 311 Normal message Storage unit 312 Latest message storage unit

Claims (6)

Connected to a plurality of IP (Internet Protocol) networks connected to provide a communication service, a congestion control device that performs congestion control of the own network that is the IP network that is controlled by the network, and the congestion control device A congestion detecting device of a congestion control system comprising: a congestion detecting device that detects congestion of another network other than the IP network other than itself,
In a POI (Point Of Interface) between the own network and the other network adjacent to the own network, a control signal directed from the own network to each of the plurality of other networks, and a response signal to the control signal, A message collector to collect;
The number of control signals collected and the number of response signals for each message element classified based on the response code indicating the response content of the response signal are determined for each of the other networks that are destinations of the control signal, and the predetermined number. Total number of messages as the latest message number information every time, and memorize the number of messages stored in the storage unit,
Based on the latest message number information, the ratio of the number of response signals for each message element to the total number of messages, which is the number of control signals, is calculated for each other network, and the calculated ratio in the other network A congestion determination unit that determines that congestion has occurred for the other network that satisfies the first congestion determination condition for determining occurrence of congestion by comparing with a predetermined threshold value;
A congestion information transmission unit that generates congestion determination information including identification information of the other network determined to have congestion, and transmits the congestion determination information to the congestion control device;
A congestion detection apparatus comprising: The message count section
The average of the most recent predetermined number of times for the number of control signals and the number of response signals for each message element for each of the other networks stored as the latest message number information in the storage unit at predetermined time intervals Is calculated and stored in the storage unit as information on the number of messages in normal times,
The congestion determination unit
Based on the normal message number information, the ratio of the number of response signals for each message element to the total number of messages as the number of control signals is calculated for each of the other networks, and the first congestion determination condition Instead of the above, as the second congestion judgment condition, the ratio of the number of response signals of the message element calculated from the message number information in the normal time and the response signal of the message element calculated from the latest message number information The congestion detection apparatus according to claim 1, wherein the congestion ratio is compared with a ratio of numbers, and it is determined that congestion has occurred in the other network when an index based on a magnitude of the difference exceeds a predetermined threshold. A plurality of IP networks connected to provide a communication service, a congestion control device that performs congestion control of the own network that is the IP network that is controlled by itself, and a device other than itself connected to the congestion control device A congestion detection method of the congestion detection device in a congestion control system comprising: a congestion detection device that detects congestion of another network that is the other IP network;
The congestion detection device includes:
In the POI between the own network and the other network adjacent to the own network, collecting a control signal directed from the own network to each of the other networks and a response signal to the control signal;
The number of control signals collected and the number of response signals for each message element classified based on the response code indicating the response content of the response signal are determined for each of the other networks that are destinations of the control signal, and the predetermined number. Totaling the latest message count information every time and storing in the storage unit,
Based on the latest message number information, calculating a ratio of the number of response signals for each message element to the total number of messages, which is the number of control signals, for each other network;
Determining the occurrence of congestion for the other network satisfying a first congestion determination condition for comparing the calculated ratio in the other network with a predetermined threshold to determine occurrence of congestion;
Generating congestion determination information including identification information of the other network determined that congestion has occurred, and transmitting the congestion determination information to the congestion control device;
The congestion detection method characterized by performing. The congestion detection device further includes:
The average of the most recent predetermined number of times for the number of control signals and the number of response signals for each message element for each of the other networks stored as the latest message number information in the storage unit at predetermined time intervals And storing in the storage unit as normal message number information;
Calculating a ratio of the number of response signals for each message element to the total number of messages, which is the number of the control signals, for each of the other networks based on the number of messages in the normal time;
Instead of determining the occurrence of congestion using the first congestion determination condition, as a second congestion determination condition, a ratio of the number of response signals of the message element calculated from the message number information in the normal time, Compared with the ratio of the number of response signals of the message element calculated from the latest message number information, and when the index based on the magnitude of the difference exceeds a predetermined threshold, it is determined that congestion has occurred in the other network And steps to
The congestion detection method according to claim 3, wherein:   A congestion detection program for causing a computer to execute the congestion detection method according to claim 3.   A computer-readable recording medium on which the congestion detection program according to claim 5 is recorded.

Images