WO2011020363A1

WO2011020363A1 - Method and system for realizing load balance and diameter client

Info

Publication number: WO2011020363A1
Application number: PCT/CN2010/073675
Authority: WO
Inventors: 甘斌; 宋大猛; 陈刚; 徐键
Original assignee: 中兴通讯股份有限公司
Priority date: 2009-08-19
Filing date: 2010-06-08
Publication date: 2011-02-24
Also published as: CN101631360B; CN101631360A

Abstract

A method and system for realizing load balance and a Diameter client are provided by the invention, therein, the method for realizing load balance includes: during the communication between the Diameter client and a first Diameter server, if it is determined that the load of the first Diameter server exceeds a pre-set first threshold, the Diameter client transmits the partial request messages of all the request messages need to be transmitted to a second Diameter server according to a predetermined rate. When the Diameter client has determined the first Diameter server is overload, it transmits the partial request messages of all the request messages to the second Diameter server, accordingly, the problem is resolved that the resource utilization ratio is not high resulted from a method for realizing load balance in the art, the load balance is realized flexibly and the resource utilization ratio is enhanced.

Description

The present invention relates to the field of communications, and in particular to a method and system for implementing load balancing and a Diameter client. BACKGROUND Since the birth of network communication technology, authentication, authentication, and accounting (AAA, Authentication, Authorization, Accounting) systems have become the cornerstone of communication network operations. At present, with the evolution of the third-generation mobile communication system to the all-IP-based network architecture, the Diameter protocol has become a new generation of AAA technology, and the Diameter protocol has attracted much attention due to its powerful scalability and security. International standards organizations such as the Alliance (ITU), the 3rd Generation Partnership Project (3GPP), and the 3rd Generation Partnership Project 2 (3GPP2) have officially adopted the Diameter protocol as the Next Generation Network (NGN). The preferred AAA protocol for future communication networks such as Wideband Code Division Multiple Access (CDMA) and Code Division Multiple Access (CDMA). More and more operators choose the Diameter protocol as the AAA preferred protocol. The Diameter protocol has better security and scalability than the previous Radius protocol. The unified Diameter interface increases the interoperability of devices between different vendors. As more and more traditional operators turn to the Diameter protocol, there is an inevitable need for a new demand for authentication, authorization, billing, etc., and load balancing needs. This load balancing is not only reflected in the client, but more on the server side. The Diameter server may need to process Diameter messages sent by multiple clients. There are many methods for implementing load balancing in the prior art. The usual practice is to achieve a certain degree of load balancing through human control or operation and maintenance configuration. For example, load balancing is implemented between different servers by configuring a certain allocation policy. These methods are ubiquitous and require manual intervention, and even if the traffic is not large at all, the load sharing is forced according to the established policy, which causes unnecessary processing load on the peer server and makes resource utilization low. In addition, the Diameter protocol clearly proposes Failover/failback compared to the Radius protocol.

(failover/failback) mechanism. The failover mechanism described in the protocol means that when the node detects a transmission failure, if possible, all request messages that do not receive a response are resent to the alternate agent. When the node detects the original transmission recovery, subsequent messages will be sent to the peer. This process is called fault backtracking. 1 is a flow chart of a Failover/failback process according to the prior art. As shown in FIG. 1 , the process of sending a failover when a sending node sends a request message includes the following steps: Step S102: When a sending node needs to send a message, perform routing analysis according to the destination domain name in the request message, and obtain a corresponding next hop node. Link S104: The sending node sends the request message to the next hop, and then caches the request message to the local request message queue. Step S106: The sending node periodically detects the link status, if the link status is normal, for example, does not occur. If the fault is interrupted, the process goes to step S112. Otherwise, the process goes to step 108 to start the failover. Step S108: The sending node scans the message queue, and performs route analysis again through the destination domain name to obtain the link corresponding to the candidate peer. Step S110: The sending node sends a message to the candidate peer. Step S112: The sending node receives the response message corresponding to the request message, and removes the request message from the locally cached message queue. After the failover is initiated, once the primary peer returns to normal, subsequent messages are sent to the primary peer. In the above prior art, the timing of initiating the failover must be a failure of the primary server, such as a link interruption, that is, the failover operation is performed only when the failure occurs. This kind of load balancing implementation often causes one link and server to be overloaded, while the other link and server are not fully utilized, resulting in insufficient resource utilization. In view of the problem that the resource utilization caused by the load balancing implementation method in the prior art is not high, an effective solution has not been proposed yet. SUMMARY OF THE INVENTION The present invention has been made in view of the problem of low resource utilization caused by the load balancing implementation method in the prior art. To this end, the main object of the present invention is to provide a method and system for implementing load balancing and a Diameter client. Solve the above problem. In order to achieve the above object, according to an aspect of the present invention, a load balancing implementation method is provided, including: in a process of communicating between a Diameter client and a first Diameter server, if it is determined that a load of the first Diameter server exceeds The preset first threshold, the Diameter client sends a partial request message in all the request messages to be sent to the second Diameter server according to a predetermined ratio. According to another aspect of the present invention, a Diameter client is provided, including: a receiving module, configured to receive a response message from a first Diameter server; and a detecting module, configured to detect a Diameter client and a first Diameter client Whether the communication traffic between the devices exceeds a preset threshold; the equalization module is configured to: when the response message carries information indicating that the server is busy, or when the detecting module detects that the communication traffic between the Diameter client and the first Diameter server exceeds When the threshold is preset, part of the request messages in all the request messages to be sent are sent to the second Diameter server according to a predetermined ratio. According to still another aspect of the present invention, a load balancing implementation system is provided, including a Diameter client, a first Diameter server, and a second Diameter server. The above-mentioned Diameter client includes: a first receiving module, configured to receive a response message from the first Diameter server; and a first detecting module, configured to detect whether the communication traffic between the Diameter client and the first Diameter server exceeds a preset number a thresholding module, configured to: when the response message carries information indicating that the server is busy, or when the first detecting module detects that the communication traffic between the Diameter client and the first Diameter server exceeds a preset first threshold, A part of the request messages in all the request messages to be sent are sent to the second Diameter server according to a predetermined ratio. The first Diameter server includes: a second receiving module, configured to receive a request message from a Diameter client; a second detecting module, configured to detect whether an overhead of the device exceeds a preset second threshold; and a sending module, configured to be in the second When the detection module detects that its own overhead exceeds the preset second threshold, the detection module sends a response message carrying the information indicating that the server is busy to the Diameter client. According to the present invention, the Diameter client determines that the load of the first Diameter server is too large A part of the request messages in all the request messages to be sent are sent to the second Diameter server, thereby flexibly implementing load sharing and improving resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are set to illustrate,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In the drawings: FIG. 1 is a flowchart of a Failover/failback process according to the prior art; FIG. 2 is a flowchart of a method for implementing load balancing according to an embodiment of the present invention; FIG. 3 is a flowchart according to an embodiment of the present invention. FIG. 4 is a preferred flowchart of a method for implementing load balancing according to an embodiment of the present invention; FIG. 5 is a Diameter client according to an embodiment of the present invention; FIG. 6 is a load balancing diagram shown in FIG. Implement the structural diagram of the system. BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict. Considering the problem of low resource utilization caused by the load balancing implementation method in the prior art, according to the present invention, the Diameter client determines a partial request message in all request messages to be sent when the load of the first Diameter server is excessively large. It is sent to the second Diameter server, which flexibly realizes load sharing and improves resource utilization. Method Embodiments According to the present invention, a method of implementing load balancing is provided. 2 is a flow chart of a method of implementing load balancing in accordance with the present invention. As shown in FIG. 2, the method includes the following steps S202 to S206: Step S202, the Diameter client communicates with the first Diameter server. For example, the Diameter client sends all request messages to the first Diameter server. Step S204: Determine a load condition of the first Diameter server. If the load of the first Diameter server exceeds the preset threshold, then go to step S206; otherwise, go to step S202, where the threshold may be preset in the actual situation. There are two ways to judge the load of the first Diameter server: 1)

The Diameter client detects that the communication traffic between the first Diameter server and the first Diameter server exceeds a preset traffic threshold, and determines that the load of the first Diameter server exceeds a preset threshold. Here, the traffic threshold may be preset according to actual conditions. The load of the first Diameter server exceeds a preset threshold, for example, the CPU usage exceeds a preset threshold, for example, the CPU usage exceeds 80%, or the occupied memory resource exceeds a preset threshold. For example, the occupied memory resource exceeds 80% of the total amount; 2) the Diameter client receives the response message from the first Diameter server, and determines that the load of the first Diameter server exceeds a preset threshold, where the first Diameter When the server exceeds the preset threshold, the server sends the response message, where the response message carries the information that the tag server is busy. Here, the overhead exceeds a preset threshold, for example, the CPU usage exceeds a preset threshold, for example, , CPU usage exceeds 80%, or the memory resources occupied exceed The threshold value is set, for example, memory resources occupied by more than 80% of the total and the like. Step S206: The Diameter client sends a part of the request messages in all the request messages to be sent to the second Diameter server according to a predetermined ratio. The predetermined ratio may be preset in the actual situation, for example, one-half, that is,

The Diameter client sends one-half of all the request messages to be sent to the second Diameter server. Then, the Diameter client discovers that the load condition of the first Diameter server is restored to a normal level through traffic monitoring, and starts the fault backtracking process, and sends all the request messages to be sent to the first Diameter server for processing. According to the present invention, the Diameter client sends a partial request message in all the request messages to be sent to the second Diameter server when determining that the load of the first Diameter server is too large, thereby flexibly implementing load sharing and improving resource utilization. In addition, the failover is avoided only in the event of a failure of the first Diameter server, which reduces the possibility of message transmission failure. In addition, the load balancing implementation method according to the embodiment of the present invention does not affect the original The function of Failover/failback, that is, failover and failback operations can still be performed in the event of a failure of the first Diameter server. The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. 3 is a schematic diagram of an implementation system of load balancing according to an embodiment of the present invention. As shown in FIG. 3, the system includes: a Diameter Client 302, a Diameter Server 304, and a Diameter Server 306. The Diameter client 302 sends a Diameter authentication ί Authenticated Charging Request message to the Diameter server 304 or the Diameter server 306 through the Diameter ten server, or receives the Diameter server 304 or the Diameter calendar. The Diameter authentication · ί authorized charging response message returned by the server 306. Here, the Diameter client 302 may be an S-CSCF (Serving-Call Session Control Function) or a GGSN (Gateway GPRS Supporting Node). For the Diameter client 302, the Diameter client 302 detects the load condition of the Diameter server 304, and when determining that the load of the Diameter server 304 is normal, sends the relevant authentication and payment request message to the Diameter service. The device 304, or receives the returned Diameter authentication ί authorized charging response message from the Diameter server 304. When it is determined that the load condition of the Diameter server 304 is abnormal, that is, the Diameter client 302 detects that the communication traffic between the Diameter client 302 and the Diameter server 304 exceeds a preset traffic threshold (the traffic threshold may be preset according to actual conditions, For example, the traffic threshold is 80 Mbps, or the Diameter server 304 returns a response message carrying the DIAMETER TOO BUSY information (where the Diameter server 304 detects that its own overhead exceeds a preset threshold, and sends the response) The message, the response message indicates that the Diameter server 304 is busy;), then the Diameter client 302 sends a partial request message in all Diameter request messages to be sent to the Diameter server 306 according to a predetermined ratio, and from the Diameter server 306) The Diameter response message returned. Upon determining that the load condition of the Diameter server 304 is back to normal, the Diameter client 302 may again send all Diameter request messages to be sent to the Diameter server 304. Here, the Diameter client 302 monitors the communication traffic with the Diameter server 304 through a preset traffic threshold. If the traffic exceeds a preset traffic threshold, the load of the Diameter server 304 is considered to exceed the pre-load. Threshold, need to fail over; or The Diameter client 302 receives the error response returned by the Diameter server 304, and the response carries the information of the DIAMETER TOO BUSY. The load of the Diameter server 304 is also considered to exceed the threshold, and a failover is required. After determining that the failover needs to be performed, the Diameter client 302 may send a partial request message of all the request messages to be sent to the Diameter server 306 according to a predetermined ratio, wherein the predetermined ratio may be preset according to actual conditions. For example, set the scale to one-half. For the Diameter server 304, the Diameter server 304 receives the Diameter authentication authorization charging request message sent by the Diameter client 302, and processes the request message according to the local policy to reply the response message. Further, when the Diameter server 304 receives too many messages locally and the overhead exceeds the threshold (here, the threshold can be preset according to the actual situation), the Diameter client can be notified by replying an error response to the request message from the Diameter client 302. The load control is performed by the 302, wherein the error response carries the error code "DIAMETER_TOO_BUSY," and the foregoing cost exceeds the threshold, which may be: the CPU usage exceeds a preset threshold, for example, the CPU usage exceeds 80%; or The occupied memory resource exceeds a preset threshold, for example, the occupied memory resource exceeds 80% of the total amount. For the Diameter server 306, the Diameter server 306 receives the Diameter authentication sent by the Diameter client 302. Authorizing the charging request message, and processing the request message according to the local policy, replying the response message. Further, the Diameter server 306 needs to perform data synchronization with the Diameter server 304 so that the request sent by the client can be processed in cooperation with the Diameter server 304. Message. Diameter month in the embodiment of the present invention The server 304 and the Diameter server 306 may be an HSS (Home Subscriber Server) or an OCS (Online Charging System). Based on the system described above, an embodiment according to the present invention FIG. 4 is a flowchart of a method for implementing load balancing according to an embodiment of the present invention. As shown in FIG. The method mainly includes the following steps: Step S402: The Diameter client 302 determines that the communication with the Diameter server 304 as the primary server is normal. The method for determining that the communication is normal may be that the message is detected through the Diameter link, or may be passed. Business interaction messages are coming. Step 4: S304: The Diameter client 302 sends all the request messages to the Diameter server 304. Step S406: The Diameter client 302 detects whether the communication traffic between the Diameter client 302 and the Diameter server 304 exceeds a preset traffic threshold during communication with the Diameter server 304. If the communication traffic exceeds the preset traffic threshold, the Diameter server 304 is overloaded, that is, the load of the Diameter server 304 exceeds the threshold, and the process goes to step S408; otherwise, the load of the Diameter server 304 is normal, and the process goes to step S412. Step S408: Since the load of the Diameter server 304 is too large, load balancing control is required. At this time, the Failover mechanism is activated to send a part of all the request messages to be sent to the standby server, that is, the Diameter server 306, according to a preset ratio. Step S410: The Diameter client 302 receives the response message returned from the Diameter server 306 and continues processing. Step S412: The Diameter client 302 determines that the traffic flow between the Diameter client and the Diameter server 304 is at the traffic threshold, continues normal processing, and receives the primary server, that is, the Diameter server 304 returns. Response message. Step S414: The Diameter client 302 determines whether the response message carries

Information about "DIAMETER TOO BUSY". If so, it is determined that the load of the Diameter server 304 exceeds a predetermined threshold, and the process goes to step S408; otherwise, the process continues. Subsequently, if the Diameter client 302 detects that the traffic between it and the Diameter server 304 as the primary server falls below a preset threshold, or does not receive the piggyback from the Diameter server 304 within a predetermined time. If there is a response message of "DIAMETER_TOO_BUSY,", the failback mechanism is started, and all the request messages to be sent are sent to the Diameter server 304. In summary, for the Diameter client and the load balancing implementation system of the present invention, the Diameter client All the request messages to be sent when the terminal determines that the load of the first Diameter server is too large. A part of the request message is sent to the second Diameter server, thereby flexibly implementing load sharing and improving resource utilization. In addition, the failover is avoided only in the event of a failure of the first Diameter server, which reduces the possibility of message transmission failure. In addition, the load balancing implementation method according to the embodiment of the present invention does not affect the original failover/failback function, that is, the failover and fault backtracking operations can still be performed when the first Diameter server fails. Apparatus Embodiment According to an embodiment of the present invention, a Diameter client is provided. Figure 5 is a Diameter client in accordance with an embodiment of the present invention. As shown in FIG. 5, the Diameter client includes: a receiving module 502, configured to receive a response message from a first Diameter server; and a detecting module 504, configured to detect a communication between a Diameter client and a first Diameter server Whether the traffic exceeds a preset threshold, where the threshold may be preset according to an actual situation, for example, the traffic threshold is 80 Mbps; the equalization module 506 is configured to: when the response message carries information indicating that the server is busy, or in the detection module 504: When detecting that the communication traffic between the Diameter client and the first Diameter server exceeds the foregoing preset threshold, send a partial request message in all request messages to be sent to the second Diameter server according to a predetermined ratio, where the predetermined The ratio can be set according to the actual situation, for example, one-half. The equalization module 506 is further configured to: when the receiving module 502 does not receive the response message carrying the information indicating that the server is busy from the first Diameter server within a predetermined time, or the detection module 504 detects the Diameter client and the first Diameter. When the communication traffic between the servers is lower than the preset threshold, all the request messages to be sent are sent to the first Diameter server. When the first Diameter server exceeds the preset threshold, the response message is sent, and the response message carries the information that the tag server is busy. Here, the overhead exceeds a preset threshold, which may be: CPU usage The rate exceeds the preset threshold. For example, the CPU usage exceeds 80%. Or, the occupied memory resources exceed the preset threshold. For example, the occupied memory resources exceed 80% of the total. System Embodiment According to an embodiment of the present invention, a load balancing implementation system is provided. FIG. 6 is a structural diagram of the load balancing implementation system shown in FIG. 2. As shown in FIG. 6, the implementation system includes a Diameter client 302, a first Diameter server 304, and a second Diameter server. 306. The above-mentioned Diameter client 302 includes: a first receiving module 602, configured to receive a response message from the first Diameter server 304; a first detecting module 604, configured to detect between the Diameter client 302 and the first Diameter server 304 Whether the communication traffic exceeds a preset threshold, wherein the threshold may be preset according to an actual situation, for example, the traffic threshold is 80 Mbps; and the equalization module 606 is configured to: when the response message carries information indicating that the server is busy Or when the first detecting module 604 detects that the communication traffic between the Diameter client 302 and the first Diameter server 304 exceeds the preset threshold, part of the request messages to be sent according to a predetermined ratio. The message is sent to the second Diameter server 306, where the predetermined ratio can be set, for example, one-half. The equalization module 606 is further configured to: when the first receiving module 602 does not receive the response message from the first Diameter server 304 carrying the information indicating that the server is busy, or the first detecting module 604 detects the Diameter client. When the communication traffic between the terminal 302 and the first Diameter server 304 is lower than the preset threshold, all the request messages to be sent are sent to the first Diameter server 304. The first Diameter server 304 includes: a second receiving module 612, configured to receive a request message from the Diameter client 302, and a second detecting module 614, configured to detect whether the overhead of the first Diameter server 304 exceeds a preset threshold, where The threshold may be preset according to the actual situation. The first sending module 616 is configured to send, to the Diameter client 302, when the second detecting module 614 detects that the overhead of the first Diameter server 304 exceeds the preset threshold. A response message indicating that the server is busy. Here, the overhead exceeds a preset threshold, for example, the CPU usage exceeds a preset threshold, for example, the CPU usage exceeds 80%, or the occupied memory resource exceeds a preset threshold, for example, the occupied memory resource exceeds The second Diameter server 306, which includes 80% of the total amount, includes: a third receiving module 622, configured to receive a request message from the Diameter client 302, and a third detecting module 624, configured to detect whether the overhead of the second Diameter server 306 exceeds a preset threshold, wherein the threshold may be preset according to an actual situation; the second sending module 626 is configured to: when the third detecting module 624 detects that the overhead of the second Diameter server 306 exceeds the preset threshold, The Diameter client 302 sends a response message carrying information indicating that the server is busy. In summary, for the Diameter client and the load balancing implementation system of the present invention, the Diameter client determines all the request messages to be sent when the load of the first Diameter server is too large. A part of the request message is sent to the second Diameter server, thereby improving resource utilization. In addition, it is avoided that failover is performed only when the first Diameter server fails, which reduces the possibility of message transmission failure. In addition, the Diameter client and the load balancing implementation system according to the embodiment of the present invention do not affect the original failover/failback function, that is, the failover and fault backtracking operations can still be performed when the first Diameter server fails. It should be noted that the steps shown in the flowchart of the accompanying drawings may be performed in a computer system such as a set of computer executable instructions, and, although the logical order is shown in the flowchart, in some cases, The steps shown or described may be performed in an order different than that herein. Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device, which can be concentrated on a single computing device or distributed over a network composed 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 modules or steps are made into a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the scope of the present invention are intended to be included within the scope of the present invention.

Claims

A method for implementing load balancing, characterized in that it comprises:

During the communication between the Diameter client and the first Diameter server, if it is determined that the load of the first Diameter server exceeds a preset first threshold, the Diameter client is to send all the request messages according to a predetermined ratio. A partial request message is sent to the second Diameter server. The method according to claim 1, wherein the step of determining that the load of the first Diameter server exceeds a preset first threshold comprises:

If the Diameter client detects that the communication traffic between the first Diameter server and the first Diameter server exceeds a preset second threshold, determining that the load of the first Diameter server exceeds the preset first Threshold. The method according to claim 1, wherein the step of determining that the load of the first Diameter server exceeds a preset first threshold comprises:

The first Diameter server sends a response message to the Diameter client when detecting that the overhead of the device exceeds a preset third threshold, where the response message carries information that the tag server is busy;

The Diameter client receives the response message from the first Diameter server and determines that the load of the first Diameter server exceeds the preset first threshold. The method according to any one of claims 1 to 3, wherein after the Diameter client sends the partial request message to the second Diameter server, the method further comprises:

And when the Diameter client detects that the communication traffic between the first Diameter server and the first Diameter server is lower than a preset fourth threshold, all the request messages sent by the Diameter client are sent to the first Diameter month. Server. The method according to any one of claims 1 to 3, wherein after the Diameter client sends the partial request message to the second Diameter server, the method further comprises: And all requests that the Diameter client will send when the Diameter client does not receive a response message from the first Diameter server carrying the information that the tag server is busy within a predetermined time. A message is sent to the first Diameter server.

The method according to any one of claims 1 to 3, wherein the request message includes a Diameter Authentication ί Authenticated Charging Request message.

7. A Diameter client, characterized in that it comprises:

a receiving module, configured to receive a response message from the first Diameter server; and a detecting module, configured to detect whether a communication traffic between the Diameter client and the first Diameter server exceeds a preset threshold;

An equalization module, configured to: when the response message carries information indicating that the server is busy, or when the detecting module detects that the communication traffic between the Diameter client and the first Diameter server exceeds the When the threshold is preset, part of the request messages in all the request messages to be sent are sent to the second Diameter server according to a predetermined ratio.

The Diameter client according to claim 7, wherein the equalization module is further configured to: when the receiving module does not receive the indication server busy from the first Diameter server within a predetermined time When the response message of the information is sent, or when the detecting module detects that the communication traffic between the Diameter client and the first Diameter server is lower than the preset threshold, all the request messages to be sent are sent to The first Diameter server.

A load balancing implementation system, wherein the implementation system includes a Diameter client, a first Diameter server, and a second Diameter server, where

The Diameter client includes:

a first receiving module, configured to receive a response message from the first Diameter server;

a first detecting module, configured to detect whether the communication traffic between the Diameter client and the first Diameter server exceeds a preset first threshold; and the equalization module is configured to: when the response message carries, indicating that the server is busy When the information is detected, or when the first detecting module detects that the communication traffic between the Diameter client and the first Diameter server exceeds the preset first threshold, Sending, in a predetermined proportion, a part of the request messages of all the request messages to be sent to the second Diameter server;

The first Diameter server includes:

a second receiving module, configured to receive a request message from the Diameter client, a second detecting module, configured to detect whether an overhead of the device exceeds a preset second threshold, and a sending module, configured to detect, by the second detecting module When the cost of the self exceeds the preset second threshold, the response message carrying the information indicating that the server is busy is sent to the Diameter client.

The system according to claim 9, wherein the equalization module is further configured to: when the first receiving module does not receive the carried from the first Diameter server within a predetermined time, When the response message indicating the busy information of the server is used, or when the first detecting module detects that the communication traffic between the Diameter client and the first Diameter server is lower than the preset first threshold Sending all request messages to be sent to the first Diameter server.