KR100350473B1 - CONGESTION CONTROLLING METHOD FOR SIGNALLING LINK IN No.7 SIGNALLING NETWORK - Google Patents

Abstract

The present invention relates to a congestion control method for a signal link of No.7 signal network, comprising: checking a state of a buffer by adding a predetermined congestion start offset from a write pointer for storing a message in a buffer in a non-congestion state; As a result of the check, if the buffer is in a usable state, it is managed as a non-congested state.

Description

No. 7 Signal congestion control method for signal link of signal network {CONGESTION CONTROLLING METHOD FOR SIGNALLING LINK IN No.7 SIGNALLING NETWORK}

The present invention relates to a congestion control method such as a message processing method for detecting congestion of signal traffic transmitted on a signal link and managing a signal network in a congested state.

In No.7 signaling system, 128 transmission buffers exist for each signal link. The buffer size must be greater than or equal to the maximum size of the message transmitted on the signal link. It is also recommended that the signal link withstand 0.2 erlang signal traffic on the signal link. Signal link congestion detection recommended in the ITU-T (International Telecommunication Union Telecommunication Standardization Sector) has different application criteria according to domestic signal networks and international signal networks, and precise standards are not specified. Only congestion detection based on buffer occupancy is defined.

Therefore, the congestion detection should be detected by the total number of buffers and the number of occupied buffers or a combination of the total buffer capacity and the buffer occupancy capacity (the number of buffers and the buffer size). The actual buffer occupancy capacity refers to traffic transmitted on a signal link, which is processed by an input / output device (I / O device), and thus, a method of using the number of buffers occupied by a central processing unit (CPU) will be described below.

Congestion detection methods include a write pointer check method and an input / output device request method.

1 is a view for explaining a congestion detection method using a write pointer check method. Referring to FIG. 1, the CPU 10 receives a signal from levels 3 and 4 and temporarily stores a signal in the buffer 12, and the buffer 12 temporarily stores a message received from levels 3 and 4. Memory. The input / output device 14 serves to transmit the message stored in the buffer to the physical layer. At this time, the write pointer indicates a location where the CPU 10 stores the message in the buffer 12, and the read pointer reads the message from the buffer 12 in the input / output device 14 and transmits the message to the physical layer. Point to location.

Referring to FIG. 1, the write pointer check method stores the message while continuously increasing from the initial position of the buffer 12 in the CPU 10. If the location to store the message is located at the end of the buffer, the message is stored again from the beginning of the buffer, and the input / output device 14 also reads the message from the beginning and transmits it. Therefore, if the buffer of the write pointer is occupied, the entire buffer 12 is occupied, and if it is available, it means that there is available memory in the buffer 12.

This method checks whether the write pointer's buffer is occupied or idle when saving a message, and if it is occupied, recognizes it as a congestion. In this method, there is an advantage that all the buffers can be used, but the buffer occupancy is not known as a problem, and the buffer occupancy can be known only when congestion occurs. In addition, if a message network management message is discarded without being transmitted on the signal link because the priority of the message is not considered, it may have a serious problem of downlink signal and wrong routing information, which may cause an error in signal traffic transmission. have.

2 is a diagram for describing a congestion detection method using an input / output device request method. Referring to FIG. 2, before the CPU 10 stores a message in the buffer 12, the CPU 10 requests a read pointer managed by the input / output device 14, receives a response thereto, and calculates the share of the current buffer, or When the device 14 stores a read pointer in a specific register, the CPU 10 reads the read pointer and calculates a buffer occupancy rate. In this method, the write pointer and read pointer can be used to know the exact buffer occupancy so that the signal link congestion can be managed at an arbitrary threshold. Therefore, even if a congestion occurs, a randomly available buffer exists so that the general user message is discarded and only the signaling network management message is transmitted to prevent the downlink. However, the overhead for receiving requests and responses in the input / output device 14 is so large that the number of messages that can be processed per second decreases, thus degrading the performance of the signal link. As a result of actual performance measurement, there is a problem that the signal link processing capacity is only about 20% than the write pointer method.

If there is more traffic than recommended on the signal link, there may be a loss of messages, but there should be no down link. Accordingly, an object of the present invention is to maintain accurate signal network information by preventing the signal link from going down when excessive traffic is concentrated, and also to efficiently comply with the recommendation for congestion to the maximum capacity that the signal link can tolerate. 7. A control method for congestion of signal links in a signal network is provided.

Another object of the present invention is to propose a solution to the above-mentioned problem, that is, the signal link down problem in the write pointer check method and the method of reducing the overhead in the I / O device request method, and the signal recommended in the ITU-T. It provides a stable signal network by providing a method that can maximize the performance of the signal link without violating the link congestion control and detect the signal link congestion and process the message in the congestion state without the signal link down. The exercise of signal traffic processing is to provide a congestion control method for the signal link of the No. 7 signal network.

1 is a view for explaining a congestion detection method using a write pointer check method,

2 is a view for explaining a congestion detection method using an input-output device request method;

3 is a view illustrating a congestion detection method using a method of checking a flag of an offset position according to an embodiment of the present invention;

4 is a message processing flowchart of a congestion and non-congestion control process according to an embodiment of the present invention;

5 is a view for explaining a congestion detection method in a non-congestion state according to an embodiment of the present invention,

6 is a view illustrating a congestion reduction detection scheme in a congestion state according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, specific matters such as specific processes are shown, which are provided to help a more general understanding of the present invention, and the specific matters may be modified or changed within the scope of the present invention. It is self-evident to those of ordinary knowledge in Esau.

A hardware configuration for detecting signal link congestion according to the present invention includes a CPU for processing a message received from levels 3 and 4, and a buffer for receiving and storing a message from the CPU and reading the message from the CPU as in the related art. I / O device for transmitting to the physical layer is provided. In software, congestion onset threshold and congestion onset offset for detecting congestion conditions, congestion abatement threshold and congestion abatement for detecting congestion reduction. offset) is provided.

FIG. 3 is a diagram illustrating a congestion detection method using a method of checking a flag of an offset position according to an embodiment of the present invention. Referring to FIG. 3, the congestion initiation threshold is appropriately set larger than the congestion reduction threshold and smaller than the total buffer capacity. The congestion reduction threshold is appropriately set greater than zero and less than the congestion initiation threshold. The congestion start offset is a value that is used to determine whether the congestion start offset is occupied at the storage position of the current buffer to detect the congestion state. Therefore, this value is an extra buffer capacity for the signal management message, and is appropriately set to be larger than zero and smaller than the congestion reduction offset. Congestion Reduction Offset is a value used to determine whether the buffer at the current buffer position plus the congestion reduction offset is occupied in order to reduce the congestion state. The congestion reduction offset is larger than the congestion start offset and smaller than the total buffer capacity. It is set appropriately. The reason why the congestion reduction offset is larger than the congestion start offset is to provide hysteresis in the congestion state. Thus, the capacity between the congestion start threshold and the congestion reduction threshold is non-congestion if the current signal link state is managed in the non-congestion state, and congestion if it is managed in the congestion state.

4 is a flowchart illustrating a message processing of a congestion and non-congestion control process according to an embodiment of the present invention. FIG. 5 is a view illustrating a congestion detection method in a non-congestion state, and FIG. 6 illustrates a congestion reduction detection method in a congestion state. It is a figure for demonstrating. 4 and 5, first, the processing scheme according to the present invention in the current non-congestion state will be described. If the current state is a runaway state in step 20 of FIG. Step 30 is to check the congestion start threshold, in which the CPU adds the congestion start offset from the write pointer for storing the message in the buffer. This position is called a congestion start check pointer, and the buffer flag at this position is checked. As a result of checking the buffer flag, if the buffer is available, since the buffer occupancy has not exceeded the congestion start limit, it is in a non-congestion state, and then proceeds to step 32. The message is stored in the buffer and transmitted on the signal link. On the other hand, as a result of the check of the buffer flag, if the buffer is occupied state, the buffer occupancy exceeds the congestion start threshold value, so that the congestion state is managed and the process proceeds to step 32.

In step 32, because congestion is detected, if the origin of the current message is the own country, the congestion status is notified by using the CONGESTION indication primitive, and if it is a large country, TFC (Transfer-Controlled) is transmitted. In step 33, the received message is analyzed. Thereafter, it is determined whether the received message is a signal network management message in step 34. If the user message is determined in step 34, the message proceeds to step 36 and the message is discarded. In step 37, the current state is managed as a congestion state. On the other hand, if the message received as a result of the determination in step 34 is a signal network management message, the process proceeds to step 35 and checks the buffer flag of the location to be stored at present, and if the buffer to be stored currently is available, stores the signal link. To be sent over the network. At this time, if the buffer to be stored is occupied, it is necessary to discard the message and readjust the congestion start threshold and the congestion start offset. After that, the process proceeds to step 37 to manage the current state as a runaway state.

Referring to FIG. 4 and FIG. 6, when the message processing process is performed in the congestion state, when the message is received in the congestion state, the congestion reduction check pointer is checked by adding the congestion reduction offset to the location of the buffer to store the message in step 22. Will be checked. As a result of checking the buffer flag, if the buffer is used, it is regarded as reaching the limit of congestion reduction, and then proceeds to step 24 to notify release of congestion, and then sends a message in step 25, and then returns the current status to normal in step 25. To manage. On the other hand, if the congestion reduction limit is not reached in step 22, the process proceeds to step 26 to analyze the message. If it is determined in step 34 that the received message is a signaling network management message, if it is a signaling network management message, the process proceeds to step 28 and transmits the message. If the message is a general message, the process proceeds to step 29 to discard the message.

On the other hand, if all buffers are occupied by the signal management message during the congestion processing, the system waits until an available buffer occurs, and does not generate a usable buffer even until a message existing in one buffer can be transmitted. If not, it is regarded as an error of the input / output device and the input / output device is initialized.

As described above, the present invention can be made, but the specific embodiments have been described in the above description of the present invention, and various modifications can be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be defined by the described embodiments, but by the claims and equivalents of the claims.

As described above, the present invention can control the user traffic by easily detecting signal link congestion without degrading or downlinking the signal link due to congestion start offset and the like, and also hysteresis when transitioning from the congestion state to the non-congestion state. By providing a stable signal link. Moreover, the present invention is designed to control traffic on its own without being influenced by the status of all users who use the signal link, so that the signal traffic can be normally processed by providing a stable signal network without degrading performance and downing the signal link. have.

Claims (7)

delete delete delete In the congestion control method for the signal link of No.7 signal network, Checking the state of the buffer by adding a predetermined congestion start offset from a write pointer for storing a message in the buffer in a non-congestion state, If the buffer is available in the check result of the buffer plus the congestion start offset, managing the non-congested state and transmitting a message; if the buffer is occupied, transitioning to the congested state; Checking the state of the buffer by adding a predetermined congestion reduction offset from the write pointer for storing messages in the buffer in the congestion state, And a congestion transition if the buffer is available as a result of the check of the buffer plus the congestion reduction offset, and if the buffer is occupied, managing the congestion. The congestion control method of claim 4, wherein the management of the congestion state analyzes the received message and discards or transmits the received message according to the priority of the received message. 5. The method of claim 4, wherein the management of the congestion status comprises the steps of analyzing the received message, discarding if the received message is a user message, and if the received message is a signaling network management message, the status of the buffer to be stored at present. And checking, and if the buffer to be stored currently is in a usable state, storing the signal network management message to be transmitted over a signal link. The method of claim 6, wherein when the buffer to be stored is occupied in managing the congestion state, the controller waits for a preset time, and if an available buffer does not occur after the waiting time, it is regarded as an error of the input / output device. Runaway control method.