WO2008049347A1

WO2008049347A1 - A method and device of controlling congestion in a service network

Info

Publication number: WO2008049347A1
Application number: PCT/CN2007/070256
Authority: WO
Inventors: Haopeng Zhu
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2006-10-25
Filing date: 2007-07-09
Publication date: 2008-05-02
Also published as: CN101170488A; CN101170488B

Abstract

A method of controlling congestion in a service network is disclosed, the method includes: monitoring the congestion status of the link in the process of exchanging service messages between local terminal and remote terminal; starting or closing flow control according to the congestion status of the link; in the process of flow control, discarding the first service message to remote terminal in ratio of flow control. A device of controlling congestion in a service network is also disclosed, the device includes: unit of monitoring congestion, unit of controlling flow, unit of executing flow control. Flow of wireless link congestion can be automatically controlled and the influence to the service can be reduced.

Description

Business Network Congestion Control Method and Apparatus The present application claims priority to Chinese Patent Application No. 200610149982.6, entitled "Business Network Congestion Control Method and Apparatus," filed on October 25, 2006, The content is incorporated herein by reference.

The present invention relates to the field of communications technologies, and in particular, to a service network congestion control method and apparatus. Background technique

At present, with the development of communication and network technologies, the demand for mobile information services is also growing. The rapid growth of mobile users has made the impact of large traffic on network elements more and more difficult to ignore. For example, during traditional holidays and peak hours of commuting to work, it is often the case that you can't get through the phone. A large part of the reason is that the traffic volume during the peak period exceeds the load capacity of the communication equipment system design, causing large-scale congestion of the communication equipment, a decrease in the connection rate, and even a failure of the communication equipment. Since the congestion and overload of the network will seriously affect the throughput of the system, the flow control technology has become a research hotspot.

The existing flow control technologies are mostly based on the underlying link congestion, such as the MTP3 (the third level of the message delivery part) layer, and simply use the leaky bucket technology to simply drop the message to alleviate the link congestion. This method of packet loss has strong randomness and arbitrariness. Since the business interaction process between network elements requires a series of message interactions, it is obvious that this random and arbitrary packet loss policy will affect the normal operation of the service. Conducting, even causing business interruptions, network problems and other serious problems. Summary of the invention

The embodiment of the invention provides a congestion control method in a communication system, which implements automatic flow control on the congestion of the link and reduces the impact on the service.

The embodiment of the invention provides a service network congestion control method, including:

Monitoring the link congestion status during the service message exchange between the local device and the peer device; and starting or closing the flow control according to the link congestion status;

In the flow control process, the service message sent to the peer device is discarded according to the flow control ratio.

The embodiment of the invention further provides a service network congestion control device, including: The congestion monitoring unit is configured to monitor a link congestion condition during a service message exchange between the local device and the peer device.

a flow control unit, configured to start or close the flow control according to the link congestion condition; the flow control execution unit, the flow control is performed according to the command of the flow control unit, and in the flow control process, according to the flow The first message that discards the traffic sent to the peer device.

It can be seen from the technical solution provided by the foregoing embodiments of the present invention that the present invention uses a transaction processing delay to monitor an HTR (Hard to Reach) of an entity, and initiates a service message exchange between the local device and the peer device. The congestion monitoring timer corresponding to the service performs the link flow control according to the timeout ratio of the congestion monitoring timer corresponding to all the services on the local device, thereby shielding the complex networking mode such as the underlying link switching. In the technical solution of the example, the failure ratio is used to characterize the link congestion condition, so that the network device implements automatic flow control according to the congestion condition of the link. It ensures that the network equipment can still operate normally under the condition of large traffic impact link, and there is no accident such as downtime, which effectively improves the reliability of the network. In addition, when the message is discarded, the present invention, in combination with the business process, can selectively discard the service message according to the business process, thereby minimizing the cost of the business loss. DRAWINGS

1 is a flow chart of an implementation method of a preferred embodiment of the present invention;

2 is a flow chart of discarding a service message according to a link congestion ratio according to an embodiment of the present invention; FIG. 3 is a flowchart of an application of the method of the present invention in a MAP location update;

4 is a schematic block diagram of a congestion control apparatus according to a first embodiment of the present invention;

Figure 5 is a block diagram showing the principle of a congestion control apparatus according to a second embodiment of the present invention. detailed description

The core of the present invention is to use a transaction processing delay to monitor an entity HTR (Hard to Reach), and monitor the link congestion status during the service message exchange between the local device and the peer device; Or the flow control is disabled. In the flow control process, the first or higher message of the service sent to the peer device is discarded. To further ensure the reasonable and effective flow control, adjust the current flow control ratio of the link in real time, and determine the proportion of discarded messages according to the current flow control ratio of the link. The larger the flow control ratio, the discarded message. The bigger the ratio.

The invention will be further described in detail below with reference to the drawings and embodiments. Referring to Figure 1, there is shown a flow chart of an implementation of a preferred embodiment of the present invention, comprising the steps of:

Step 101: Set the monitoring period.

It can be flexibly set according to the actual network conditions. For example, 10 seconds is a period. Each period can be continuous or have a certain interval, for example, 2 seconds. In the specific implementation, it can be controlled by a monitoring timer.

Step 102: Statistics the failure rate of the service during the monitoring period.

The invention implements flow control based on the service, and performs timeout monitoring on the interaction of the service message, for example, using a counter to count the received message timeout. After the monitoring period starts, the counter starts counting. As long as the message times out, it indicates that the service has failed. At this time, the counter is incremented by 1. In addition, a counter is set to count the success of the business. This count is incremented by one each time the entire process of business processing is completed. In this way, the failure rate of the service during a monitoring period can be determined based on the statistics of the two counters. For example, calculate the failure rate of the business according to the following formula:

Failure ratio = timeout count / (timeout count + business success count)

Since the timeout of message interaction reflects the congestion status of the link, the proportion of service failures according to the above formula also reflects the congestion status of the link.

After the start of the monitoring period, the counter for counting the received message timeout starts counting, and the counter is incremented by one each time the congestion monitoring timer expires. In this way, the timeout ratio of the congestion monitoring timer corresponding to all services on the local device in each monitoring period can be accurately calculated.

Step 103: Determine whether the proportion of failures counted in the monitoring period exceeds a predetermined flow control threshold. If it has been exceeded, it proceeds to step 104; otherwise, it proceeds to step 105.

Step 104: Determine whether flow control has been started. If it has been started, it proceeds to step 106; otherwise, it proceeds to step 107.

Step 105: Determine whether the timeout ratio in the monitoring period is lower than a predetermined flow control closing threshold. If yes, go to step 108; otherwise, go to step 106.

Step 106: Adjust the flow control ratio according to the failure ratio. Then, return to step 102 to monitor the next cycle.

Step 107: Start flow control and set the default flow control ratio. Then, returning to step 102, the next cycle is monitored. In order to achieve smooth adjustment of service traffic, a flow control ratio can be set. After the flow control is started, a part of the traffic can be allowed to pass according to the flow control ratio. For example, if the flow control ratio is 50%, the first 10 are discarded in every 20 messages, and the last 10 are normally sent.

Step 108: Turn off flow control. Then, returning to step 102, the next cycle is monitored. In the above process, the start threshold and the close threshold of the flow control can be set according to system simulation or experience values.

After the flow control is started in the foregoing step 107, the first message of the service sent by the local device to the peer device needs to be discarded according to the flow control ratio. When discarding a business message, you can use a default flow ratio. That is, as long as the link is congested, the first message of the service that the local device needs to send to the peer device is discarded according to the flow control ratio. In this way, if the adjustment of the flow control does not cancel the congestion, the link is still congested in the next monitoring period. According to the monitoring result, the service message to be sent continues to be discarded.

Of course, in order to further improve the flow control efficiency, the flow control ratio may be determined according to the congestion status of the link, and the proportion of discarded traffic messages may be determined according to the size of the flow control ratio. The larger the flow control ratio, the larger the proportion of messages that are discarded. Step 106 in Figure 1 illustrates this situation. In a specific implementation, different flow control ratios may be set according to the congestion degree and the status of the link. After the flow control is started, the local device needs to be sent to the peer device according to the current flow control ratio of the link. The first message. In this way, after a monitoring cycle, the congestion can be basically relieved and the link can be restored to normal.

For example, you can adjust by following the flow control ratio adjustment strategy:

If the timeout ratio counted in two consecutive monitoring periods exceeds the predetermined flow control ratio adjustment threshold. That is to say, it is necessary to monitor the change of the timeout ratio in each monitoring period. If the timeout ratio of the previous monitoring period and the current period exceeds the predetermined flow control ratio adjustment threshold, the flow control ratio is increased;

If the timeout ratio counted in the previous monitoring period is lower than the predetermined flow control instance adjustment threshold, the flow control ratio is reduced.

The flow control proportional adjustment threshold can be set based on system simulation or empirical values.

When a service message is discarded, all services can have the same priority. That is, after the flow control is started, different service messages to be sent can be discarded according to the flow control ratio. To provide differentiated services to different types of services, you can also classify services and set priorities for each type of service. After the flow control is started, the service messages with lower priority are preferentially discarded. In this way, the quality of service for important services can be guaranteed.

FIG. 2 shows a flow of discarding a service message according to a service failure ratio in the embodiment of the present invention, including the following steps:

Step 201: The local device exchanges service messages with the peer device.

Step 202: Determine whether flow control has been started. If it has been started, it proceeds to step 203; otherwise, it proceeds to step 206.

Step 203: Determine whether the message is a service message within a flow control ratio. If yes, go to step 204; otherwise, go to step 206.

Step 204: Determine whether the message is the first message of the service. If yes, proceed to step 205; otherwise, proceed to step 206.

Step 205: Discard the message and discard the current service.

Step 206: Send the message and start the congestion monitoring timer to monitor whether the message is timed out.

In order to further improve the flow control efficiency, the flow control ratio may be determined according to the congestion status of the link, and the proportion of discarded traffic messages may be determined according to the size of the flow control ratio. The larger the flow control ratio, the larger the proportion of messages that are discarded.

In a specific implementation, different flow control ratios may be set according to the congestion degree of the link. After the flow control is started, the first traffic of the local device to be sent to the remote device is discarded according to the current flow control ratio of the link. Message. In this way, after a monitoring cycle, the congestion can be basically relieved and the link can be restored to normal. In addition, an initial value of the flow control ratio can be set. After the flow control is started, the current flow control ratio of the link is adjusted in real time according to the change of the link congestion degree in the continuous monitoring period. Then, the local device determines whether the message to be sent is a message within the flow control ratio according to the current actual flow control ratio of the link, and if yes, discards the message. Of course, the greater the proportion of flow control, the greater the rate of message drop.

The invention can be applied to any network that requires business message interaction, such as in MAP (Mobile Application Part) and CAP (CAMEL Application Part) transactions. It is known to those skilled in the art that MAP is an interface protocol for operation between mobile communication systems, including signaling communication with other system entities centered on the MSC, for example with another MSC/VLR (Mobile Switching Center/Visit Location Register), HLR (Home Location Register), AC (Authentication Center), SMC (Short Message Center) connections, and the connection of system entities to each other. This protocol is used for communication between entities, such as location registration/deletion, processing of supplementary services, retrieval and switching of customer parameters during call setup, customer management, operation and maintenance, recovery after location register failure, international mobile client equipment Identification (IMEI) management and authentication, procedures for supporting short message services, etc. In these business processes, the present invention can be utilized to effectively control the traffic of the link and prevent link congestion.

For example, in the location update service of the mobile terminal, the MSC has two points to interact with the HLR, one is to take an authentication set, and the other is to send a location update request, and the congestion monitoring timing corresponding to the two interaction points needs to be started. To monitor the congestion of the link.

The application flow in the embodiment of the present invention is as shown in FIG. 3:

1. The MS (Mobile Terminal) sends a Location Update Request message to the MSC/VLR.

2. After receiving the location update request message, the MSC/VLR needs to send an authentication set or location update request message to the HLR. Before sending a message, determine whether flow control is required according to the link congestion status. If the link is not congested, that is, if flow control is not required, execute 3; if flow control is required, go to step 4.

3. Perform the normal processing flow, send the message directly to the HLR, and start the congestion monitoring timer to monitor whether the message times out.

4. The MSC determines whether to discard the message according to the current flow control ratio of the link. If you need to discard, go to step 5; otherwise, go to step 6.

5. The MSC discards the message and responds to the MS with a location update failure message.

6. The MSC sends an authentication set or location update request message to the HLR, and starts a congestion monitoring timer to monitor whether the message times out.

In order to enable entities in the network to monitor an entity's HTR (Hard to Reach) according to the transaction delay, the link traffic is effectively controlled to avoid network congestion. The present invention also provides a service network congestion control device.

Referring to Figure 4, there is shown a block diagram of the apparatus of the first embodiment of the present invention: The apparatus includes: a congestion monitoring unit 41, a flow control unit 42, and a flow control execution unit 43. among them,

The congestion monitoring unit 41 is configured to monitor link congestion status during the service message interaction between the local device and the peer device.

The flow control unit 42 is configured to enable or disable flow control according to link congestion conditions. For example, when the link congestion condition exceeds a certain threshold, flow control is started; when the link congestion status is lower than another threshold, the flow control is closed. Of course, a policy unit 44 can also be provided in the device, as shown. The policy unit 44 is configured to set and provide the flow control unit with a flow control policy for starting and/or closing flow control.

The flow control execution unit 43 performs flow control on the link according to the command of the flow control unit. For example, the service message can be discarded according to a default flow control ratio, and the first message of the service sent to the peer device is discarded.

The congestion monitoring unit can be implemented in various ways. For example, as shown in the figure, the congestion monitoring unit 41 includes: a monitoring period control unit 411 and a statistics unit 412. The monitoring period control unit 411 is configured to set a monitoring period and time the monitoring period. The statistics unit 412 collects the timeout condition and the successful situation of the message interaction, and after the monitoring unit counts the monitoring period, the monitoring period is reached. Calculating the failure ratio of the service in the monitoring period according to the timeout condition of the message interaction, and understanding the link congestion status according to the failure ratio.

The detailed process of flow control of the link is similar to that described above in the method of the present invention and will not be described herein.

Referring to Figure 5, Figure 5 is a block diagram of the apparatus of the second embodiment of the present invention:

Different from the first embodiment of the apparatus of the present invention shown in Fig. 4, in this embodiment, a flow control ratio setting unit 45, a monitoring unit 46, and a flow control ratio adjustment unit 47 are also provided. among them,

The flow control ratio setting unit 45 is connected to the flow control execution unit 43 and sets a flow control ratio for determining the number of the first message of the service to be discarded, so that the flow control execution unit 43 determines the need according to the flow ratio. The first message of the discarded service.

In order to further improve the flow control efficiency, the monitoring unit 46 connected to the statistical unit 412 monitors the value of the failure ratio in one of the monitoring periods and the variation in a plurality of consecutive monitoring periods, and the flow control ratio The adjusting unit 47 adjusts the flow control according to the monitoring result of the monitoring unit The current flow control ratio of the link set by the ratio setting unit. If the timeout ratio exceeds the predetermined flow control ratio adjustment threshold for two consecutive monitoring periods, increase the current flow control ratio of the link; The timeout ratio in the monitoring period is lower than the flow control proportional adjustment threshold, thereby reducing the current flow control ratio of the link. Of course, other adjustment strategies can also be used to adjust the flow control ratio of the link.

In this way, the flow control execution unit 43 can determine the proportion of discarded traffic messages according to the size of the flow control ratio. The larger the flow control ratio, the larger the proportion of messages that are discarded. Therefore, the flow control efficiency can be improved, and after a monitoring period, the congestion can be basically cancelled, and the link can be restored to normal.

The device of the present invention can be applied to any network device that needs service interaction to control message traffic on the link between the network device and other network devices, so as to avoid accidents such as downtime caused by large traffic impacts. The safe operation of the equipment.

It will be apparent to those skilled in the art that the various units or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device, such that they may be stored in the storage device by the computing device, or they may be separately fabricated into individual integrated circuit modules, or they may be Multiple units or steps are made in a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software.

While the invention has been described by the embodiments of the present invention, it will be understood that

Claims

Rights request

A service network congestion control method, the method includes: monitoring a link congestion status during a service message exchange between a local device and a peer device; and starting or stopping flow control according to a link congestion condition. ;

2. The method according to claim 1, wherein the step of monitoring a link congestion condition comprises:

Set the monitoring period;

Counting the proportion of failures of the business during the monitoring period;

A link congestion condition is determined according to the failure ratio.

The method according to claim 1, wherein the step of starting or closing the flow control according to the link congestion condition comprises:

During the set monitoring period, if the failure ratio exceeds a predetermined flow control threshold, it is activated;

During the set monitoring period, if the failure ratio is lower than the predetermined flow control closing threshold, the flow control is turned off.

The method according to claim 1, wherein the step of discarding the service message sent to the peer device comprises:

The service message sent to the peer device is discarded according to the current flow control ratio of the link.

The method according to any one of claims 2 to 4, further comprising: adjusting a current flow control ratio of the link in real time according to the failure ratio.

The method according to claim 5, wherein the step of adjusting the flow control ratio in real time according to the failure ratio comprises:

If the failure ratio exceeds a predetermined flow control ratio adjustment threshold in two consecutive monitoring periods, the flow control ratio is increased;

If the failure ratio in the last monitoring period is lower than the flow control ratio adjustment threshold, the flow control ratio is lowered.

The method according to any one of claims 1 to 4, further comprising: classifying the service, and setting a priority corresponding to each type of service; When multiple services are coexisting on the local device, the service messages with lower priority are preferentially discarded.

The method according to any one of claims 1 to 4, further comprising: the mobile terminal performing the location update service, starting the congestion monitoring timing when the authentication set is taken and/or the location update request is sent To monitor the congestion of the link.

A service network congestion control device, comprising:

The congestion monitoring unit is configured to monitor a link congestion condition during a service message exchange between the local device and the peer device.

10. The apparatus according to claim 9, wherein the congestion monitoring unit comprises:

a monitoring period control unit, configured to set a monitoring period, and time the monitoring period; a statistical unit, configured to count the timeout condition of the message interaction, and after the monitoring unit counts the monitoring period to the monitoring period, according to the The timeout condition of the message interaction calculates the failure ratio of the service in the monitoring period, and determines the link congestion status according to the failure ratio.

11. The apparatus according to claim 9 or 10, further comprising: a flow control ratio setting unit configured to provide and provide the flow control execution unit with the first item for determining a service to be discarded The flow control ratio of the number of messages.

The device according to claim 9 or 10, further comprising: a monitoring unit, connected to the statistical unit, configured to monitor the value of the failure ratio in one of the monitoring periods and continuously Changes in the monitoring period;

a flow control ratio adjusting unit, configured to adjust a current flow control ratio of the link set by the flow control ratio setting unit according to the monitoring result of the monitoring unit;

If the timeout ratio exceeds a predetermined flow control ratio adjustment threshold in two consecutive monitoring periods, the flow control ratio adjustment unit increases the current flow control ratio of the link;

If the timeout ratio is lower than the flow control ratio adjustment threshold in the previous monitoring period, the flow control ratio adjustment unit reduces the current flow control ratio of the link.

The device according to claim 9 or 10, further comprising: The policy unit is configured to set and provide the flow control unit with a flow control policy for starting and/or closing the flow control.