Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
The overload control method provided by the application can be applied to the application environment shown in fig. 1. Each user 102 is connected to each base station 104, and each base station 104 is connected to a core network 108 through a gateway 106. The base station may be, but is not limited to, various macro base stations, micro base stations, pico base stations, and distributed base stations; the gateways may be, but are not limited to, various application gateways, signaling gateways, relay gateways, access gateways, protocol gateways, and security gateways. The interface between the gateway 106 and the base station 104 and/or the interface between the gateway 106 and the core network 108 may be, but is not limited to, an Iuh interface, an Iu interface, an a interface, an S1 interface, and the like.
The gateway 106 stores the received message in a receiving and sending queue to wait for reading and processing; the receiving and sending queues may include a sending queue and a receiving queue, and further may include an uplink sending queue, a downlink sending queue, an uplink receiving queue and a downlink receiving queue;
the gateway 106 may forward request information about the user 102 and the base stations 104, may receive or send a handshake signal for establishing a communication link with each base station 104, and may transmit a paging message sent by the core network 108 to each base station 104.
When the gateway 106 reaches the capacity processing limit under the impact of a large number of users 102, time-consuming signaling processing and forwarding occur, if a new base station 104 continuously accesses the gateway 106 and initiates uplink and downlink services all the time, the current services of the existing base station 104 are affected, and in a serious case, the performance of the gateway 106 is affected, so that the device paralysis cannot be recovered.
The method and the device have the advantages that the congestion condition of the receiving and sending queue and the message time consumption condition are dynamically judged, the corresponding overload control strategy is executed, the congestion condition is relieved, the time interval of message processing time delay is reduced, and the purposes of stable operation and self healing of the device are achieved.
In one embodiment, as shown in fig. 2, an overload control method is provided, which is described by taking the application of the method to the gateway in fig. 1 as an example, and includes the following steps:
step 202, collecting message time consumption information and/or receiving and sending queue congestion information.
The message time consumption information may include time consumption duration of each message, or may include a message with longer time consumption. Further, the message time consumption ratio and the message serious time consumption ratio can be included; the message time consumption ratio can be the ratio of the number of time consumption message packets to the number of message messages; the number of the time-consuming message packets can be the number of message messages waiting for the time-consuming value or more or equal to the time-consuming value to be read; the message serious time consumption occupation ratio can be the ratio of the number of the message packets with serious time consumption to the number of the message messages; the number of the serious time-consuming message packets can be the number of the message messages with the waiting reading time larger than or equal to the serious time-consuming value; the severe elapsed time value is greater than the elapsed time value.
The transceiving queue congestion information may include a transmit queue congestion rate and a receive queue congestion rate; the congestion rate of the sending queue may be a ratio of the number of message packets in the sending queue to the capacity of the sending queue; the receive queue congestion rate may be a ratio of the number of message packets in the receive queue to the capacity of the receive queue.
Specifically, when a message is received, a timestamp is added to the message, and then when the message time consumption information is collected, the time consumption of the message in the gateway can be obtained according to the current time and the time in the timestamp, and the message time consumption information can be further obtained according to the time consumption of each message. Further, the time stamps of the message messages are respectively obtained from the receiving and sending queues of the gateway and time consumption statistics is carried out, and the time consumption situation of the message messages can be recorded through real-time sampling statistics of the time stamps of the message messages when the message messages are read by the gateway, so that the time consumption message packet number and the normal message packet number in a certain time range are obtained. Further, the time-consuming duration is divided into time-consuming intervals to obtain the number of message packets in each time-consuming interval, or the time-consuming duration is divided and collected according to different time-consuming grades to obtain the number of message packets in each time-consuming grade, and then the time-consuming duration and the normal number of message packets are obtained through division.
According to the number of the message messages in the receiving and sending queue and the capacity of the receiving and sending queue, the congestion information of the receiving and sending queue can be obtained.
Furthermore, the collected information can form a statistical list or an overload data structure table so as to be convenient to read and check, and further, the equipment load change and the congestion condition within a certain time range can be conveniently and effectively tracked.
And step 204, if the message time consumption information and/or the congestion information of the receiving and sending queue meets the overload control condition, starting a corresponding overload control strategy.
The overload control condition may include any one or any combination of the following conditions: the message consumed time accounts for a ratio greater than or equal to a consumed time ratio threshold, the sending queue congestion ratio is greater than or equal to a sending congestion threshold, the receiving queue congestion ratio is greater than or equal to a receiving congestion threshold, the message serious consumed time accounts for a ratio greater than or equal to a serious consumed time ratio threshold, and the consumed time message packet number is greater than or equal to a packet number limit value.
The overload control policy may include any one or any combination of the following policies: and discarding the paging message, discarding the handshake signals with the access rate greater than or equal to the allowed access rate threshold, and discarding the request messages with the access rate greater than or equal to the allowed access rate threshold.
It should be noted that, different severity levels can be further divided according to the number of conditions that can be met in the overload control conditions, and the following strategies can be started in sequence from low severity level to high severity level: the method includes discarding handshake signals having an access rate greater than or equal to an allowed access rate threshold, discarding request messages having an access rate greater than or equal to an allowed access rate threshold, and discarding paging messages.
The overload control method may be applied to the core network in fig. 1, and specifically, may be applied to an MSC (Mobile Switching Center) device, an MGW (Media Gateway) device, a PCRF (Policy and Charging Rules Function) device, an MME (Mobility Management Entity) device or an SGSN (Serving GPRS Support Node, Serving GPRS (General Packet Radio Service) Support Node) device, and may also be applied to various network side access devices, specifically, a base station device, a local controller, and a Gateway device (such as the Gateway in fig. 1). Further, when the above-mentioned overload control method is applied to various devices, the above-mentioned strategy that can be started in sequence from low severity to high severity may not be applicable to other devices except the gateway, and needs to be determined according to specific devices and services.
In the overload control method, the time-consuming information of the messages and the congestion information of the receiving and sending queues are collected, the collected information is used as the basis of overload control, a corresponding overload control strategy is dynamically made, the congestion during overload is relieved, the normal operation of the existing service is ensured, and various messages are normally received and forwarded after the congestion condition disappears.
In one embodiment, as shown in fig. 3, an overload control method is provided, which is described by taking the application of the method to the gateway in fig. 1 as an example, and includes the following steps:
step S302, collecting the message time consumption rate, the sending queue congestion rate and/or the receiving queue congestion rate.
In step S304, if any one or any combination of the following conditions is satisfied:
the message time consumption ratio is greater than or equal to the time consumption ratio threshold, the sending queue congestion rate is greater than or equal to the sending congestion threshold, and the receiving queue congestion rate is greater than or equal to the receiving congestion threshold;
then any one or any combination of the following policies are implemented: the method includes discarding paging messages, discarding handshake signals having an access rate greater than or equal to an allowed access rate threshold, and discarding request messages having an access rate greater than or equal to an allowed access rate threshold.
The handshake signals may be various signals interacted between the gateway and the base station when the communication link is established. Further, the handshake signal may be a handshake signal based on a Transmission layer Protocol such as SCTP (Stream Control Transmission Protocol), TCP (Transmission Control Protocol), UDP (User Datagram Protocol), and further, the handshake signal is a signal generated during a coupling process between the gateway and the base station based on the SCTP Protocol. According to the specification of the SCTP protocol, a coupling is a logical association or channel established by two SCTP endpoints (a base station and a gateway) through a 4-step handshake mechanism for data transmission. Further, the handshake signal may be a request primitive based on the SCTP protocol, that is, an INIT (initialization) signal generated by a 4-step handshake mechanism in the implementation of the coupling process.
It should be noted that the overload control method may also be applied to upper layer Application protocols such as RUA (User Application), S1AP (S1Application Protocol ), RANAP (Radio Access Network Application Part), BSSAP (Base Station System Application Part), and the like, and since the upper layer Application Protocol does not have a handshake signal (does not have a request primitive similar to SCTP), but an overload control policy of a transport layer Protocol may be triggered by the service severity of the upper layer Application Protocol.
The request message may include an uplink request message, a downlink request message, a base station registration request message, and a user registration request message. The uplink request message may include an initial direct transfer request message. The downlink request message comprises a macro network handover request message.
Specifically, the access rate may be obtained according to a timestamp of the message or the primitive when the message or the primitive is transmitted to the interface monitored by the gateway, and a timestamp of the message or the primitive at each service execution point.
In a specific embodiment, step S304 includes: and when the congestion rate of the sending queue is greater than or equal to the sending congestion threshold value, the paging message is discarded.
It should be noted that, in the above steps, if any one or any combination of the conditions of the overload control conditions, except that the congestion rate of the transmission queue is greater than or equal to the transmission congestion threshold, may still start the corresponding overload control policy, and when the condition of the congestion rate of the transmission queue is greater than or equal to the transmission congestion threshold is met, start the overload control policy of discarding the paging message.
In a specific embodiment, the message time consumption information further includes a message serious time consumption ratio; the message serious time consumption occupation ratio is the ratio of the number of the message packets with serious time consumption to the number of the message messages; the number of the serious time-consuming message packets is the number of the message messages of which the waiting reading time is greater than or equal to the serious time-consuming value; the severe elapsed time value is greater than the elapsed time value;
the overload control condition further includes a message heavy elapsed time fraction being greater than or equal to a heavy elapsed time ratio threshold.
Wherein the severe consumption value may be set to 500 milliseconds.
In a specific embodiment, step S304 includes:
and when the message serious consumed time accounts for more than or equal to the serious consumed time ratio threshold, discarding the request messages with the access rate more than or equal to the allowed access rate threshold.
Since the message heavy time consumption ratio may be too large, but the message time consumption ratio is smaller than the time consumption ratio threshold, which may also indicate that the gateway has a congestion condition, it is not sufficient to use the message heavy time consumption ratio as an overload control condition, and the size of the message heavy time consumption ratio should be considered in the judgment of the overload control condition, so as to improve the judgment accuracy.
It should be noted that, in the foregoing steps, if any one or any combination of the overload control conditions is satisfied except that the message heavy consumed time ratio is greater than or equal to the heavy consumed time ratio threshold, the corresponding overload control policy may still be started, and when the condition that the message heavy consumed time ratio is greater than or equal to the heavy consumed time ratio threshold is satisfied, the overload control policy of discarding the request message whose access rate is greater than or equal to the allowed access rate threshold is started.
In a specific embodiment, the message consumption occupation ratio includes an uplink message consumption occupation ratio and a downlink message consumption occupation ratio; the message serious time consumption occupation ratio comprises an uplink message serious time consumption occupation ratio and a downlink message serious time consumption occupation ratio;
the sending queue comprises an uplink sending queue and a downlink sending queue; the receiving queue comprises an uplink receiving queue and a downlink receiving queue; the congestion rate of the sending queue comprises the congestion rate of an uplink sending queue and the congestion rate of a downlink sending queue; the congestion rate of the receiving queue comprises the congestion rate of an uplink receiving queue and the congestion rate of a downlink receiving queue.
Specifically, the overload control conditions are further divided into uplink and downlink, so that the accuracy of judging the overload control conditions is improved, the overload control strategy is ensured to be started when the congestion occurs, and the stability is improved.
It should be noted that uplink refers to an uplink communication link from a base station and a gateway to a core network; downlink refers to a downlink communication link from a core network and a gateway to a base station.
In a specific embodiment, step S304 includes:
and when the congestion rate of the downlink sending queue is greater than or equal to the congestion threshold value of the downlink sending queue, discarding the paging message.
It should be noted that, in the above steps, if any one or any combination of the conditions of the overload control conditions, except that the congestion rate of the downlink transmission queue is greater than or equal to the congestion threshold of the downlink transmission queue, is satisfied, the corresponding overload control policy may still be started, and when the condition that the congestion rate of the downlink transmission queue is greater than or equal to the congestion threshold of the downlink transmission queue is satisfied, the overload control policy of discarding the paging message is started.
The above mentioned features of the previous embodiment are not described herein again.
In the overload control method, the time consumption information of the uplink and downlink messages and the congestion information of the uplink and downlink transceiving queues are collected, the collected information is used as the overload control basis, a corresponding overload control strategy is dynamically made, the congestion during overload is relieved, the normal operation of the existing service is ensured, and various messages are normally received and forwarded after the congestion condition disappears. Specifically, the overload control method can more specifically initiate a corresponding overload control strategy according to a condition in the overload control conditions, for example: because the paging message is a signal sent by the core network in the downlink transmission link to the base station through the gateway, the paging message is discarded when the congestion rate of the downlink transmission queue is greater than or equal to the congestion threshold of the downlink transmission; because even if the message serious time consumption ratio is too high, when a serious congestion condition occurs, the message time consumption ratio is not necessarily larger than or equal to the time consumption ratio threshold, the judgment according to the message serious time consumption ratio can better cope with the urgent congestion condition, and therefore, when the message serious time consumption ratio is larger than or equal to the serious time consumption ratio threshold, the request message with the access rate larger than or equal to the allowed access rate threshold is discarded; because the handshake signals are signals required for establishing the communication link, discarding the handshake signals with the access rate being too fast can avoid establishing the communication link with more base stations, prevent concurrent impact of a large amount of services, and ensure normal processing of the existing services.
The present embodiment will be described with reference to a specific example.
An overload control method, which is described by taking the application of the method to the gateway in fig. 1 as an example, includes:
step 1, as shown in fig. 4, when receiving an uplink message packet and a downlink message packet, performing uplink message time consumption statistics and downlink message time consumption statistics: according to the time interval (waiting for reading time) from writing in a receiving queue to reading a packet by a service thread, performing division acquisition according to different time-consuming time grades, and counting the number of normal message packets and the number of time-consuming message packets;
step 2, on the premise that overload control is started, establishing a message consumption counting item in the period to regularly acquire counted consumption data (normal message packet number and consumption message packet number) and obtain an overload packet number ratio (message consumption ratio) and a severe overload packet number ratio (message severe consumption ratio) which consumes more than 500 milliseconds; meanwhile, as shown in fig. 5, the current utilization rates (the congestion rates of the transmit queues and the receive queues) of the receive queues and the transmit queues of all the current interfaces are obtained. As shown in fig. 6, if any one or more of the conditions (overload control conditions) are satisfied by the above various conditions, signaling overload control is enabled (request messages with an access rate greater than or equal to an allowed access rate threshold are discarded);
as shown in fig. 6, if the ratio of the number of packets consumed by the messages (the ratio of consumed time of the messages) of the base station side interface and the core network side interface exceeds the preset ratio parameter (the time consumption ratio threshold), or when the signaling overload control is enabled, the SCTP barring access state is enabled (the request primitive based on the SCTP protocol with the access rate greater than or equal to the allowed access rate threshold is discarded), and the SCTP link establishment rate is set;
as shown in fig. 6, if the congestion of the transmission queues of the base station side interface and the core network side interface (transmission queue congestion rate) exceeds or approaches to a set expected value (transmission congestion threshold), the downlink paging broadcast service control is enabled (paging message is discarded);
step 3, when the corresponding overload control state is enabled, executing a rate control strategy according to different service types, wherein,
when the SCTP restricted access state is enabled:
if the request primitive based on the SCTP exceeds or approaches the SCTP link establishment rate, discarding the request primitive;
as shown in fig. 7 and 8, when the signaling overload control is enabled:
if the access rate of the base station registration request message exceeds or approaches the allowable access rate, refusing to access the base station registration request message;
if the access rate of the user registration request message exceeds or approaches the allowable access rate, refusing to access the user registration request message;
it should be noted that the base station/user access request message in fig. 7 is a base station registration request message and a user registration request message.
If the access rate of the uplink request message (initial direct transmission request message) exceeds or approaches the allowable access rate, filtering and not processing the uplink request message;
if the access rate of the downlink request message (the macro network switching request message) exceeds or approaches the allowable access rate, filtering and not processing the downlink request message;
when the downlink paging broadcast service control state is started:
the downlink paging broadcast message (paging message) directly filters the message and discards when the downlink signaling control state is activated;
step 4, timing detection is carried out to monitor the processing state of the gateway in real time, when the pressure of the gateway is recovered, the overload control state is closed after the recovery of various collected data to a normal level is detected, and at the moment, various services are processed normally;
and 5, outputting the overload control condition of the current gateway in real time, wherein the overload control condition comprises indexes such as time-consuming packet number statistics (time-consuming message packet number statistics), overload frequency (frequency for executing an overload control strategy), discarded packet number, queue utilization rate (sending queue congestion rate and receiving queue congestion rate) and the like, and the health state of the current gateway is reflected.
Any programming language may be utilized in order to implement the above-described methods. The overload control implementation method for network equipment performance congestion can solve the problem that the equipment cannot recover work due to overload due to the fact that gateway processing capacity under the condition of large gateway traffic impact can be solved; the gateway can be applied to a gateway in a Global System for Mobile Communication (GSM) network System, and can effectively solve the problem of untimely gateway processing under the impact of traffic; the provided performance data acquisition mechanism is not limited to the gateway, and other devices based on different network standards in practical application, including a local controller, a core network device, and the like, can be provided.
It should be understood that although the steps in the flowcharts of fig. 2 and 3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2 and 3 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 9, there is provided an overload control apparatus including:
an acquisition module 910, configured to acquire message time consumption information and/or queue congestion information for transceiving;
and an overload control policy module 920, configured to start a corresponding overload control policy when the message time consumption information and/or the congestion information of the transceiving queue meet an overload control condition.
In a specific example, the message consumption time information includes a message consumption time ratio; the message consumption ratio is the ratio of the number of consumed message packets to the number of message messages; the number of the time-consuming message packets is the number of the message messages waiting for reading the time-consuming value or more;
the receiving and sending queue congestion information comprises a sending queue congestion rate and a receiving queue congestion rate; the congestion rate of the sending queue is the ratio of the number of message messages in the sending queue to the capacity of the sending queue; the congestion rate of the receiving queue is the ratio of the number of the message messages in the receiving queue to the capacity of the receiving queue;
the overload control condition includes any one or any combination of the following conditions: the message consumption ratio is greater than or equal to the consumption ratio threshold, the transmit queue congestion rate is greater than or equal to the transmit congestion threshold, and the receive queue congestion rate is greater than or equal to the receive congestion threshold.
In one specific example, the overload control policy includes any one or any combination of the following policies: the method includes discarding paging messages, discarding handshake signals having an access rate greater than or equal to an allowed access rate threshold, and discarding request messages having an access rate greater than or equal to an allowed access rate threshold.
In a specific example, the overload control policy module 920 is configured to discard the paging message when the transmit queue congestion rate is greater than or equal to the transmit congestion threshold.
In a specific example, the message time consumption information further includes a message heavy time consumption ratio; the message serious time consumption occupation ratio is the ratio of the number of the message packets with serious time consumption to the number of the message messages; the number of the serious time-consuming message packets is the number of the message messages waiting for reading the time-consuming value greater than or equal to the serious time-consuming value; the severe elapsed time value is greater than the elapsed time value;
the overload control condition further includes a message heavy elapsed time fraction being greater than or equal to a heavy elapsed time ratio threshold.
In a specific example, the overload control policy module 920 is configured to discard request messages with an access rate greater than or equal to the allowed access rate threshold when the message heavy consumption ratio is greater than or equal to the heavy consumption ratio threshold.
In a specific example, the message time consumption occupation ratio includes an uplink message time consumption occupation ratio and a downlink message time consumption occupation ratio; the message serious time consumption occupation ratio comprises an uplink message serious time consumption occupation ratio and a downlink message serious time consumption occupation ratio;
the sending queue comprises an uplink sending queue and a downlink sending queue; the receiving queue comprises an uplink receiving queue and a downlink receiving queue; the congestion rate of the sending queue comprises the congestion rate of an uplink sending queue and the congestion rate of a downlink sending queue; the congestion rate of the receiving queue comprises the congestion rate of an uplink receiving queue and the congestion rate of a downlink receiving queue.
In a specific example, the overload control policy module 920 is configured to discard the paging message when the congestion rate of the downlink transmission queue is greater than or equal to a downlink transmission congestion threshold.
In one specific example, the handshake signals include a base station registration request message and a user registration request message.
In a specific example, the handshake signal is a request primitive based on the SCTP protocol.
In one particular example, the request message includes an initial direct transfer request message and a macro handover request message.
The present embodiment will be described in detail with reference to a specific example.
As shown in fig. 10, an overload control apparatus is provided, which is applied to the gateway in fig. 1 as an example, and includes:
the overload control parameter configuration unit is used for providing a parameter preset value reflecting the current gateway load upper limit; the preset parameter values comprise an overload control function enabling identifier (controlling whether to forbid the starting of an overload control strategy), a message time consumption reference preset value (a time consumption value), an interface overload packet number ratio preset value (a time consumption ratio threshold value), a normal packet number ratio preset value, a packet number ratio preset value (a serious time consumption ratio threshold value) with the message time consumption of more than 500 milliseconds, the utilization rate configuration of a sending queue and a receiving queue and the like;
the message time-consuming packet number acquisition unit is used for adding a timestamp to a data message or a primitive forwarded by the SCTP kernel, forwarding the data message or the primitive to a receiving queue, sampling and counting the message time-consuming situation in real time according to the timestamp after the message is read by an application layer thread, simultaneously acquiring uplink and downlink message packets from the receiving queue of a gateway interface for time-consuming statistics, and calculating the serious time-consuming message packet number, the time-consuming message packet number and the normal message packet number in a certain time range;
a receiving and sending queue congestion calculation unit, configured to calculate, according to the number of message packets backlogged in the current queue (the number of message packets in the sending queue and the receiving queue) and a preset total size of the queue (the capacity of the sending queue and the receiving queue), respectively, to obtain a current usage rate of the sending queue (a congestion rate of the sending queue) and a current usage rate of the receiving queue (a congestion rate of the receiving queue), and to reflect the message backlogged degrees of the sending queue and the receiving queue;
a load overload detection unit, which is used for processing the data obtained by the message time-consuming packet number acquisition unit and the transceiving queue congestion calculation unit to obtain the congestion rate of the transmitting queue, the congestion rate of the receiving queue, the message time-consuming occupation ratio (the sum of the time-consuming message packet number divided by the time-consuming message packet number and the normal message packet number) and the message serious time-consuming occupation ratio (the sum of the serious time-consuming message packet number divided by the time-consuming message packet number and the normal message packet number), and forming an overload data structure table; and executing gateway performance overload judgment according to the overload data structure table:
if one or more conditions (overload control conditions) of the congestion rate of a receiving queue, the congestion rate of a sending queue and the time consumption rate of messages of a gateway base station side or a core network side interface meet overload judgment, starting overload control of uplink and downlink request type messages of a gateway; otherwise, closing the control state and requesting the normal processing and forwarding of the class message;
if the ratio of the number of delay packets of the actual uplink and downlink signaling (the ratio of the consumed time of the message) exceeds the preset ratio of the consumed time of the message (the threshold value of the consumed time ratio), or one or more conditions for starting the overload control of the uplink and downlink request messages of the gateway meet the overload judgment, starting the overload control strategy of the SCTP link establishment rate limitation of the gateway; otherwise, closing the control state and recovering the access rate;
if the congestion rate of the sending queue of the gateway base station side interface exceeds a preset value, indicating that the downlink sending queue is congested, starting the gateway to control the paging message; otherwise, the paging message is broadcast and forwarded normally;
an access rate control unit, configured to obtain an access rate of a message or a primitive according to a timestamp of a monitored interface or a service execution point, and if the access rate meets a preset value (access rate allowed) under a control policy, the base station normally accesses and normally processes and forwards the message, otherwise, the base station terminates the processing; the access rate judgment and control strategy is carried out on various request messages again and respectively in the next time interval, which specifically comprises the following steps:
when the overload control strategy of SCTP link establishment rate limitation is enabled, discarding INIT data blocks (INIT signals) with access rate exceeding or approaching the allowable access rate;
when the overload control of the uplink and downlink request messages of the gateway is enabled, rejecting the access of a base station access request message (base station registration request message) and a user access request message (base station registration request message) with the access rate exceeding or approaching the allowable access rate, and terminating the forwarding of the uplink and downlink request message with the access rate exceeding or approaching the allowable access rate; otherwise, continuing to process and forward the request message and perform subsequent service flow;
when the gateway enables the control of the paging message, the paging message is discarded; otherwise, the paging broadcast (paging message) is forwarded to the base station.
Specifically, the message time-consuming packet number acquisition unit divides the time-consuming duration into time-consuming intervals to obtain the message packet number in each time-consuming interval, or performs division acquisition according to different time-consuming time grades to obtain the message packet number in each time-consuming time grade, and then divides the message packet number according to the message packet number in each time-consuming time grade to obtain the time-consuming message packet number and the normal message packet number; and meanwhile, the message time-consuming packet number collecting unit respectively acquires the uplink and downlink message packets from the receiving queue of the gateway interface and carries out time-consuming statistics, and calculates the time-consuming message packet number and the normal message packet number within a certain time range.
It should be noted that the performance upper limit index of the gateway (which reflects the parameter preset value of the current gateway load upper limit) may be obtained or configured by the overload control parameter configuration unit, and the obtaining manner may be that communication is performed through a system interface in the gateway or an application module is configured by a third party.
Specifically, the load overload detection unit acquires data obtained by the message time-consuming packet number acquisition unit and the transceiving queue congestion calculation unit at regular time, calculates and obtains a sending queue congestion rate, a receiving queue congestion rate, a message time-consuming occupation ratio (the sum of the time-consuming message packet number divided by the time-consuming message packet number and the normal message packet number) and a message severe time-consuming occupation ratio (the sum of the severe time-consuming message packet number divided by the time-consuming message packet number and the normal message packet number) according to the data, and further forms an overload data structure table in a cycle time according to the calculated data, wherein the overload data structure table can embody the health state of the gateway in the cycle time and can be used as a basis for executing various overload control strategies.
The overload control device solves the problem that equipment is paralyzed due to traffic impact when the gateway performance reaches the overload load; the method provides a structural body data management mode (forming an overload data structure table) of message time consumption and cache queue statistical analysis, can effectively track the equipment load data change within a certain time range, and each unit is not limited to one equipment and can also be realized by an interface negotiation organization among a plurality of equipments.
For the specific definition of the overload control device, reference may be made to the above definition of the overload control method, which is not described herein again. The modules in the overload control apparatus can be implemented in whole or in part by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a network device is provided, which includes a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the overload control method when executing the computer program.
In a specific embodiment, the network device is a base station device, a local controller, a gateway device, an MME device, an HSS device, an MSC device, an MGW device, a PCRF device, or an SGSN device.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
collecting message time consumption information and/or receiving and transmitting queue congestion information;
and if the message time consumption information and/or the receiving and sending queue congestion information meet the overload control conditions, starting a corresponding overload control strategy.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the message time consumption information comprises the message time consumption ratio; the message consumption ratio is the ratio of the number of consumed message packets to the number of message messages; the number of the time-consuming message packets is the number of the message messages waiting for reading the time-consuming value or more;
the receiving and sending queue congestion information comprises a sending queue congestion rate and a receiving queue congestion rate; the congestion rate of the sending queue is the ratio of the number of message messages in the sending queue to the capacity of the sending queue; the congestion rate of the receiving queue is the ratio of the number of the message messages in the receiving queue to the capacity of the receiving queue;
the overload control condition includes any one or any combination of the following conditions: the message consumption ratio is greater than or equal to the consumption ratio threshold, the transmit queue congestion rate is greater than or equal to the transmit congestion threshold, and the receive queue congestion rate is greater than or equal to the receive congestion threshold.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the overload control strategy comprises any one or any combination of the following strategies: the method includes discarding paging messages, discarding handshake signals having an access rate greater than or equal to an allowed access rate threshold, and discarding request messages having an access rate greater than or equal to an allowed access rate threshold.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
if the message time consumption information and/or the receiving and sending queue congestion information meet the overload control condition, the step of starting the corresponding overload control strategy comprises the following steps:
and when the congestion rate of the sending queue is greater than or equal to the sending congestion threshold value, the paging message is discarded.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the message time consumption information also comprises the rate of the serious message time consumption; the message serious time consumption occupation ratio is the ratio of the number of the message packets with serious time consumption to the number of the message messages; the number of the serious time-consuming message packets is the number of the message messages of which the waiting reading time is greater than or equal to the serious time-consuming value; the severe elapsed time value is greater than the elapsed time value;
the overload control condition further includes a message heavy elapsed time fraction being greater than or equal to a heavy elapsed time ratio threshold.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
if the message time consumption information and/or the receiving and sending queue congestion information meet the overload control condition, the step of starting the corresponding overload control strategy comprises the following steps:
and when the message serious consumed time accounts for more than or equal to the serious consumed time ratio threshold, discarding the request messages with the access rate more than or equal to the allowed access rate threshold.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the message time consumption occupation ratio comprises an uplink message time consumption occupation ratio and a downlink message time consumption occupation ratio; the message serious consumed time accounts for the ratio, including the uplink message serious consumed time accounts for the ratio and the downlink message serious consumed time accounts for the ratio;
the sending queue comprises an uplink sending queue and a downlink sending queue; the receiving queue comprises an uplink receiving queue and a downlink receiving queue; the congestion rate of the sending queue comprises the congestion rate of an uplink sending queue and the congestion rate of a downlink sending queue; the congestion rates of the receiving queues comprise the congestion rate of the uplink receiving queue and the congestion rate of the downlink receiving queue.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
if the message time consumption information and/or the receiving and sending queue congestion information meet the overload control condition, the step of starting the corresponding overload control strategy comprises the following steps:
and when the congestion rate of the downlink sending queue is greater than or equal to the congestion threshold value of the downlink sending queue, discarding the paging message.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the request message comprises a base station registration request message and a user registration request message.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
the handshake signals are request primitives based on the SCTP protocol.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.