WO2023232016A1 - Cloud service system and service processing method based on cloud service system - Google Patents

Abstract

Disclosed in the present application are a cloud service system and a service processing method based on a cloud service system. Traffic passing through a load balancing node can be effectively reduced, the bandwidth consumption of the load balancing node is reduced, and a service transmission delay can be shortened, thereby improving the user experience. In the cloud service system of the present application, a load balancing node can send a first identifier to a client node; on the basis of the first identifier, the client node can convert a destination address of a first service request from the address of the load balancing node into the address of a serving-end node, so as to obtain a processed first service request, and sends the processed first service request to the serving-end node; the serving-end node can process the processed first service request, so as to obtain a first service response, and sends the first service response to the client node; and on the basis of the first identifier, the client node can further convert a source address of the first service response from the address of the serving-end node into the address of the load balancing node, so as to obtain a processed first service response.

Description

A cloud service system and a business processing method based on the cloud service system

This application requires the priority of the Chinese patent application submitted to the China Patent Office on May 31, 2022, with the application number 202210612082.X and the invention title "A cloud service system and a business processing method based on the cloud service system". The entire contents are incorporated herein by reference.

Technical field

The embodiments of the present application relate to the field of cloud technology, and in particular, to a cloud service system and a business processing method based on the cloud service system.

Background technique

In cloud scenarios such as public cloud, private cloud, and hybrid cloud, cloud service systems generally provide users with services of a certain scale through cluster deployment. In order to prevent the load from being concentrated on a certain node or certain nodes in the cluster, load balancing nodes can be added in the cluster to distribute the load of each node in the cluster, thereby optimizing the performance of the entire cloud service system.

Currently, load balancing nodes can implement load balancing in non-direct routing (DR) mode. Non-DR mode includes destination network address translation (DNAT) mode and bidirectional address translation (full network address translation, FULLNAT) mode. . For ease of explanation, the DNAT mode is introduced below. In this mode, for the service request sent by the client node, the load balancing node can perform DNAT processing on the service request and send the processed service request to the server node. For the business response returned by the server node, the load balancing node can perform source network address translation (SNAT) processing on the business response, and send the processed business request to the client node, and the client node completes the corresponding business.

It can be seen that the traffic between the client node and the server node (business requests, business responses, etc.) will continue to pass through the load balancing node. However, cloud service systems are often deployed with a large number of client nodes and server nodes, and huge traffic needs to be processed by load balancing nodes, resulting in excessive bandwidth consumption of load balancing nodes and long service transmission delays. , resulting in poor user experience.

Contents of the invention

The embodiments of this application provide a cloud service system and a business processing method based on the cloud service system, which can effectively reduce the traffic passing through the load balancing node, reduce the bandwidth consumption of the load balancing node, and can reduce the business transmission delay and improve the user experience. .

The first aspect of the embodiments of this application provides a cloud service system, which includes: a load balancing node, a client node, and a server node. After the load balancing node, the client node and the server node shake hands, the load balancing node may first send the first identifier to the client node.

When there is a certain pending service on the client node, the client node may generate a first service request indicating the pending service, where the source address of the first service request is the address of the client node and the destination address is the load balancer. The address of the node. Then, the client node can perform DNAT processing based on the first identification pair, that is, convert the destination address of the first service request from the address of the load balancing node to the address of the server node, obtain the processed first service request, and process The first service request is sent to the server node.

After receiving the processed first service request, the server node can process the processed first service request, and obtain to the first service response, and send the first service response to the client node. The source address of the first service response is the server node, and the destination address is the client node.

After receiving the first service response, the client node can perform SNAT processing on the first service response based on the first identifier, that is, convert the source address of the first service response from the address of the server node to the address of the load balancing node, and then process it. The first business response after. Thereafter, the client node can complete the corresponding service based on the processed first service response.

It can be seen from the above cloud service system that the load balancing node can send the first identification to the client node. Then, when the client node has a service to be processed, the client node can convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, and obtain the processed first service request, and directly sends the processed first service request to the server node. Then, the server node can process the processed first service request, obtain the first service response, and directly send the first service response to the client node. Finally, the client node can convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier to obtain the processed first service response. The response completes the response business. It can be seen that in the DNAT mode, the DANT processing for the first business request and the SNAT processing for the first business response, which were originally responsible for the load balancing node, are offloaded to the client node for execution by the load balancing node. In this way, the processing The first business request and the processed first business response can be directly transmitted between the client node and the server node without being processed by the load balancing node, which can effectively reduce the traffic passing through the load balancing node and reduce load balancing. Node bandwidth consumption, and can reduce business transmission delay and improve user experience.

In a possible implementation, when there is another service to be processed, the client node can send a second service request indicating the service to be processed to the load balancing node. It should be noted that the second service request refers to The included source address is the address of the client node, and the destination address is the address of the load balancing node. After obtaining the second service request, the load balancing node can perform DNAT processing on the second service request, that is, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and obtain the processed second service ask. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can perform a series of processes on the processed second service request to obtain a second service response, and send the second service response to the load balancing node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node. After obtaining the second business response, the load balancing node can perform SNAT processing on the second business response, that is, convert the source address of the second business response from the address of the server node to the address of the load balancing node, and add it to the second business response. The first identifier is used to obtain the processed second service response. Then, the load balancing node sends the processed second service response to the client node. In this way, the client node can save the first identifier based on the processed second service response and complete the corresponding service.

In a possible implementation manner, after obtaining the second service response, the load balancing node may first detect whether the second service request contains the second identifier. If the second service request contains the second identifier, it means that the load balancing node can offload the DNAT processing for the service request and the SNAT processing for the service response to the client node, so that the client node and the server node can directly complete the process. business interaction, so the load balancing node can perform SNAT processing on the second business response, that is, convert the source address of the second business response from the address of the server node to the address of the load balancing node, and add the first identifier to the second business response , and obtain the processed second service response. Then, the load balancing node sends the processed second service response to the client node. In this way, the client node can save the first identifier based on the processed second service response and complete the corresponding service.

In a possible implementation, if the second service request does not include the second identifier, it means that the load balancing node cannot offload the DNAT processing for the service request and the SNAT processing for the service response to the client node, so the load The balancing node can perform SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the service point node to the address of the load balancing node, obtain the processed second service response, and convert the processed second service response The second business response is sent to the client node. Thereafter, the client node can complete the corresponding service based on the processed second service response.

In a possible implementation, the client node is also used to send a handshake request to the load balancing node; the load balancing node is also used to select a server node from the server node resource pool based on the handshake request, and then transfer the handshake request to the server node. The destination address of the request is converted from the address of the load balancing node to the address of the server node, the processed handshake request is obtained, and the processed handshake request is sent to the server node; the server node is also used to process the processed handshake request. The handshake request is processed, a handshake response is obtained, and the handshake response is sent to the load balancing node; the load balancing node is also used to convert the source address of the handshake response from the address of the server node to the address of the load balancing node, and after processing Handshake response and send the processed handshake response to the client node.

The second aspect of the embodiment of this application provides a cloud service system, which includes: a load balancing node, a client node, and a server node. After the load balancing node, the client node and the server node shake hands, the load balancing node may first send the first identifier to the client node and the server node.

When there is a certain pending service on the client node, the client node may generate a first service request indicating the pending service, where the source address of the first service request is the address of the client node and the destination address is the load balancer. The address of the node. Then, the client node can perform DNAT processing based on the first identification pair, that is, convert the destination address of the first service request from the address of the load balancing node to the address of the server node, obtain the processed first service request, and process The first service request is sent to the server node.

After receiving the processed first service request, the server node can process the processed first service request and obtain a first service response, where the source address of the first service response is the server node and the destination address is the client. end node. Then, the server node can perform SNAT processing on the first service response based on the first identifier, that is, convert the source address of the first service response from the address of the server node to the address of the load balancing node to obtain the processed first service response. , and sends the processed first business response to the client node. Thereafter, the client node can complete the corresponding service based on the processed first service response.

It can be seen from the above cloud service system that the load balancing node can send the first identifier to the client node and the server node. Then, when the client node has a service to be processed, the client node can convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, and obtain the processed first service request, and sends the processed first service request to the server node. Then, the server node can process the processed first service request to obtain the first service response, and convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier, The processed first service response is obtained, and the processed first service response is sent to the client node to complete the response service based on the processed first service response. It can be seen that in the DNAT mode, the DANT processing for the first service request that was originally responsible for the load balancing node is offloaded to the client node for execution by the load balancing node, and the SNAT for the first service response that was originally responsible for the load balancing node The processing is offloaded by the load balancing node to the server node for execution. In this way, the processed first business request and the processed first business response can be directly transmitted between the client node and the server node without going through Load balancing node processing can effectively reduce the traffic passing through the load balancing node, reduce the bandwidth consumption of the load balancing node, and can reduce business transmission delay and improve user experience.

In a possible implementation, when there is another service to be processed, the client node can send a second service request indicating the service to be processed to the load balancing node. It should be noted that the second service request refers to The included source address is the address of the client node, and the destination address is the address of the load balancing node. After obtaining the second service request, the load balancing node can perform DNAT processing on the second service request, that is, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and add it to the second service request. The first identification is obtained by processing the second service request. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can save the first identifier, perform a series of processes on the processed second service request, thereby obtain the second service response, and send the second service response to the load. Balanced node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node. After obtaining the second business response, the load balancing node can perform SNAT processing on the second business response, that is, convert the source address of the second business response from the address of the server node to the address of the load balancing node, and add it to the second business response. The first identifier is used to obtain the processed second service response. Then, the load balancing node sends the processed second service response to the client node. In this way, the client node can save the first identifier based on the processed second service response and complete the corresponding service.

In a possible implementation, after obtaining the second service request, the load balancing node can first detect whether the second service request contains the second identifier. If the second service request contains the second identifier, it means that the load balancing node can take responsibility for itself. The DNAT processing for the business request is offloaded to the client node, and the SNAT processing for the business response is offloaded to the server node. Therefore, the load balancing node can perform DNAT processing on the second business request, that is, the destination address of the second business request is changed from The address of the load balancing node is converted into the address of the server node, and the first identifier is added to the second service request to obtain the processed second service request. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can save the first identifier, perform a series of processes on the processed second service request, thereby obtain the second service response, and send the second service response to the load. Balanced node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node. After obtaining the second business response, the load balancing node can perform SNAT processing on the second business response, that is, convert the source address of the second business response from the address of the server node to the address of the load balancing node, and add it to the second business response. The first identifier is used to obtain the processed second service response. Then, the load balancing node sends the processed second service response to the client node. Thereafter, the client node can save the first identification based on the processed second service response and complete the corresponding service.

In a possible implementation, if the second service request does not include the second identifier, it means that the load balancing node cannot offload the DNAT processing for the service request that it is responsible for to the client node, and offloads the SNAT processing for the service response to the client node. The server node, so the load balancing node can perform DNAT processing on the second service request, that is, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and obtain the processed second service request. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can perform a series of processes on the processed second service request to obtain a second service response, and send the second service response to the load balancing node. After obtaining the second service response, the load balancing node can perform SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node to obtain the processed second service response. Then, the load balancing node sends the processed second service response to the client node. Thereafter, the client node completes the corresponding service based on the processed second service response.

In a possible implementation, the client node is also used to send handshake requests to the load balancing node; load The balancing node is also used to select the server node in the server node resource pool based on the handshake request, and then convert the destination address of the handshake request from the address of the load balancing node to the address of the server node to obtain the processed handshake request. And send the processed handshake request to the server node; the server node is also used to process the processed handshake request, obtain the handshake response, and send the handshake response to the load balancing node; the load balancing node is also used to Convert the source address of the handshake response from the address of the server node to the address of the load balancing node, obtain the processed handshake response, and send the processed handshake response to the client node.

The third aspect of the embodiments of this application provides a cloud service system, which includes: a load balancing node, a client node, and a server node. After the load balancing node, the client node and the server node shake hands, the load balancing node may first send the first identifier to the client node and the server node.

When there is a certain pending service on the client node, the client node may generate a first service request indicating the pending service, where the source address of the first service request is the address of the client node and the destination address is the load balancer. The address of the node. Then, the client node can perform DNAT processing and SNAT processing based on the first identification pair, that is, convert the source address of the first service request from the address of the client node to the address of the load balancing node, and convert the destination address of the first service request to Convert the address of the load balancing node to the address of the server node, obtain the processed first business request, and send the processed first business request to the server node.

After receiving the processed first service request, the server node can process the processed first service request and obtain a first service response, where the source address of the first service response is the server node and the destination address is the client. end node. Then, the server node can perform SNAT processing on the first service response based on the first identifier, that is, convert the source address of the first service response from the address of the server node to the address of the load balancing node, and convert the corresponding destination address of the first service , convert the address of the load balancing node to the address of the client node, obtain the processed first service response, and send the processed first service response to the client node. Thereafter, the client node can complete the corresponding service based on the processed first service response.

It can be seen from the above cloud service system that the load balancing node can send the first identifier to the client node and the server node. Then, when the client node has a service to be processed, the client node can convert the source address of the first service request from the address of the client node to the address of the load balancing node based on the first identifier, and convert the source address of the first service request to the address of the load balancing node. The destination address is converted from the address of the load balancing node to the address of the server node, the processed first service request is obtained, and the processed first service request is sent to the server node. Then, the server node can process the processed first service request to obtain the first service response, and convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier, Convert the corresponding destination address of the first service from the address of the load balancing node to the address of the client node, obtain the processed first service response, and send the processed first service response to the client node to perform processing based on The first business response after the completion of the response business. It can be seen that in the FULLDNAT mode, the DANT processing and SNAT processing for the first business request that were originally responsible for the load balancing node are offloaded to the client node for execution by the load balancing node. The DANT processing and SNAT processing of the response are offloaded to the server node for execution by the load balancing node. In this way, the processed first business request and the processed first business response can be directly processed between the client node and the server node. Inter-transmission does not need to be processed by the load balancing node, which can effectively reduce the traffic passing through the load balancing node, reduce the bandwidth consumption of the load balancing node, reduce business transmission delay, and improve user experience.

In a possible implementation, when there is another service to be processed, the client node can send a second service request indicating the service to be processed to the load balancing node. It should be noted that the second service request refers to Contains the source address is the address of the client node, and the destination address is the address of the load balancing node. After obtaining the second service request, the load balancing node can perform DNAT processing and SNAT processing on the second service request, that is, convert the source address of the second service request from the address of the client node to the address of the load balancing node, and convert the second service request to the address of the load balancing node. The destination address of the request is converted from the address of the load balancing node to the address of the server node, and the first identifier is added to the second service request, thereby obtaining the processed second service request. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can save the first identifier, perform a series of processes on the processed second service request, thereby obtain the second service response, and send the second service response to the load. Balanced node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node. After obtaining the second service response, the load balancing node can perform DNAT processing and SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node, and convert the second service response to the address of the load balancing node. The destination address of the response is converted from the address of the load balancing node to the address of the client node, and the first identifier is added to the second service response to obtain a processed second service response. Then, the load balancing node sends the processed second service response to the client node. In this way, the client node can save the first identifier based on the processed second service response and complete the corresponding service.

In a possible implementation, after obtaining the second service request, the load balancing node can first detect whether the second service request contains the second identifier. If the second service request contains the second identifier, it means that the load balancing node can take responsibility for itself. The DNAT processing and SNAT processing for the business request are offloaded to the client node, and the DNAT processing and SNAT processing for the business response are offloaded to the server node. Therefore, the load balancing node can perform DNAT processing and SNAT processing on the second business request, that is, The source address of the second service request is converted from the address of the client node to the address of the load balancing node, and the destination address of the second service request is converted from the address of the load balancing node to the address of the server node, and is forwarded to the second service node. The request is to add the first identifier, thereby obtaining the processed second service request. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can save the first identifier, perform a series of processes on the processed second service request, thereby obtain the second service response, and send the second service response to the load. Balanced node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node. After obtaining the second service response, the load balancing node can perform DNAT processing and SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node, and convert the second service response to the address of the load balancing node. The destination address of the response is converted from the address of the load balancing node to the address of the client node, and the first identifier is added to the second service response to obtain a processed second service response. Then, the load balancing node sends the processed second service response to the client node. In this way, the client node can save the first identifier based on the processed second service response and complete the corresponding service.

In a possible implementation, if the second service request does not include the second identifier, it means that the load balancing node cannot offload the DNAT processing and SNAT processing for the service request that it is responsible for to the client node, and will transfer the DNAT processing and SNAT processing for the service response to the client node. DNAT processing and SNAT processing are offloaded to the server node, so the load balancing node can perform DNAT processing and SNAT processing on the second service request, that is, convert the source address of the second service request from the address of the client node to the address of the load balancing node , convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and add the first identifier to the second service request, thereby obtaining the processed second service request. Then, the load balancing node may send the processed second service request to the server node. After obtaining the processed second service request, the server node can perform a series of processes on the processed second service request to obtain the second service response, and send the second service response. to the load balancing node. After obtaining the second service response, the load balancing node can perform DNAT processing and SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node, and convert the second service response to the address of the load balancing node. The destination address of the response is converted from the address of the load balancing node to the address of the client node, and the first identifier is added to the second service response to obtain a processed second service response. Then, the load balancing node sends the processed second service response to the client node. Thereafter, the client node completes the corresponding service based on the processed second service response.

In a possible implementation, the client node is also used to send the handshake request to the load balancing node; the load balancing node is also used to select the server node in the server node resource pool based on the handshake request, and send the handshake request to the server node. The source address is converted from the address of the client node to the address of the load balancing node, the destination address of the handshake request is converted from the address of the load balancing node to the address of the server node, the processed handshake request is obtained, and the processed The handshake request is sent to the server node; the server node is also used to process the processed handshake request, obtain the handshake response, and send the handshake response to the load balancing node; the load balancing node is also used to process the handshake response The source address is converted from the address of the server node to the address of the load balancing node, the destination address of the handshake response is converted from the address of the load balancing node to the address of the client node, the processed handshake response is obtained, and the processed The handshake response is sent to the client node.

The fourth aspect of the embodiment of the present application provides a business processing method based on a cloud service system. The system includes a load balancing node, a client node, and a server node. The method includes: the load balancing node sends the first identification to the client. End node; the client node converts the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, obtains the processed first service request, and converts the processed first service request The request is sent to the server node; the server node processes the processed first service request, obtains the first service response, and sends the first service response to the client node; the client node sends the first service request based on the first identifier. The source address of the response is converted from the address of the server node to the address of the load balancing node, and the processed first business response is obtained.

In a possible implementation, the load balancing node sending the first identification to the client node includes: the load balancing node converts the destination address of the second service request from the client node from the address of the load balancing node to the server node. The address of the node, obtains the processed second business request, and sends the processed second business request to the server node; the load balancing node transfers the source address of the second business response from the server node to the service point node. The address is converted into the address of the load balancing node, the first identifier is added to the second service response, the processed second service response is obtained, and the processed second service response is sent to the client node, and the second service response is the server The node processes the processed second service request.

In a possible implementation, the method further includes: the load balancing node detects whether the second service request contains the second identifier; then the load balancing node sends the first identifier to the client node including: if the second service request includes the second identifier; Second identifier, the load balancing node converts the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, adds the first identifier to the second service response, and obtains the processed third Second business response, and send the processed second business response to the client node.

In a possible implementation, the method further includes: if the second service request does not contain the second identifier, the load balancing node converts the source address of the second service response from the server node from the address of the service point node. is the address of the load balancing node, obtains the processed second service response, and sends the processed second service response to the client node.

The fifth aspect of the embodiment of the present application provides a business processing method based on a cloud service system. The system includes a load balancing node, a client node, and a server node. The method includes: the load balancing node sends the first identification to the client. The end node and the server node; the client node transfers the destination address of the first service request from the load balancing based on the first identifier. The address of the node is converted into the address of the server node, the processed first business request is obtained, and the processed first business request is sent to the server node; the server node processes the processed first business request, and obtains The first business response is to convert the source address of the first business response from the address of the server node to the address of the load balancing node based on the first identifier, obtain the processed first business response, and convert the processed first business response Sent to client node.

In a possible implementation, the load balancing node sending the first identification to the client node and the server node includes: the load balancing node sends the destination address of the second service request from the client node to the address of the load balancing node. Convert to the address of the server node, add the first identifier to the second business request, obtain the processed second business request, and send the processed second business request to the server node; the load balancing node will come from the server node The source address of the second business response is converted from the address of the service point node to the address of the load balancing node, the first identifier is added to the second business response, the processed second business response is obtained, and the processed second business response is The response is sent to the client node, and the second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the method further includes: the load balancing node detects whether the second service request contains the second identifier; the load balancing node sends the first identifier to the client node and the server node includes: if the second service request The request contains the second identifier, and the load balancing node converts the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, adds the first identifier to the second service request, and is processed The second service request after processing, and sends the processed second service request to the server node; the load balancing node converts the source address of the second service response from the server node from the address of the service point node to the load balancing node address, add the first identifier to the second service response, obtain the processed second service response, and send the processed second service response to the client node.

In a possible implementation, the method further includes: if the second service request does not contain the second identifier, the load balancing node converts the destination address of the second service request from the client node from the address of the load balancing node. is the address of the server node, obtains the processed second business request, and sends the processed second business request to the server node; the load balancing node transfers the source address of the second business response from the server node to the service node. The address of the point node is converted into the address of the load balancing node, a processed second service response is obtained, and the processed second service response is sent to the client node.

The sixth aspect of the embodiment of the present application provides a business processing method based on a cloud service system. The system includes a load balancing node, a client node, and a server node. The method includes: the load balancing node sends the first identification to the client. The end node and the server node; the client node converts the source address of the first service request from the address of the client node to the address of the load balancing node based on the first identifier, and converts the destination address of the first service request from the address of the load balancing node The address is converted into the address of the server node, the processed first business request is obtained, and the processed first business request is sent to the server node; the server node processes the processed first business request and obtains the A business response, converting the source address of the first business response from the address of the server node to the address of the load balancing node based on the first identifier, and converting the corresponding destination address of the first business from the address of the load balancing node to the client The address of the node, obtains the processed first service response, and sends the processed first service response to the client node.

In a possible implementation, the load balancing node sending the first identification to the client node and the server node includes: the load balancing node sends the source address of the second service request from the client node to the address of the client node. Convert to the address of the load balancing node, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, add the first identifier to the second service request, and obtain the processed second service request, And send the processed second business request to the server node; the load balancing node will use the source address of the second business response from the server node, Convert the address of the service point node to the address of the load balancing node, convert the corresponding destination address of the second service from the address of the load balancing node to the address of the client node, add the first identifier to the second service response, and obtain the processed the second service response, and sends the processed second service response to the client node. The second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the method further includes: the load balancing node detects whether the second service request contains the second identifier; the load balancing node sends the first identifier to the client node and the server node includes: if the second service request The request contains the second identifier, and the load balancing node converts the source address of the second service request from the client node from the address of the client node to the address of the load balancing node, and converts the destination address of the second service request from the load balancing node. The address of the node is converted into the address of the server node, the first identifier is added to the second business request, the processed second business request is obtained, and the processed second business request is sent to the server node; the load balancing node will come from The source address of the second service response of the server node is converted from the address of the service point node to the address of the load balancing node, and the corresponding destination address of the second service is converted from the address of the load balancing node to the address of the client node. The first identifier is added to the second service response, a processed second service response is obtained, and the processed second service response is sent to the client node.

In a possible implementation, the method further includes: if the second service request does not contain the second identifier, the load balancing node converts the source address of the second service request from the client node to the address of the client node. is the address of the load balancing node, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, obtain the processed second service request, and send the processed second service request to Server node; the load balancing node converts the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, and converts the corresponding destination address of the second service from the address of the load balancing node Convert to the address of the client node, obtain the processed second service response, and send the processed second service response to the client node.

The seventh aspect of the embodiment of the present application also provides a cloud service system. The load balancing node, client node and server node of the system all include a sending module and a processing module, wherein the sending module of the load balancing node is used to Send the first identification to the client node; the processing module of the client node is used to convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identification, and obtain the processed The first business request; the sending module of the client node is used to send the processed first business request to the server node; the processing module of the server node is used to process the processed first business request to obtain the first service request. A business response; the sending module of the server node is used to send the first business response to the client node; the processing module of the client node is also used to send the source address of the first business response from the server based on the first identifier. The address of the node is converted into the address of the load balancing node, and the processed first service response is obtained.

In a possible implementation, the processing module of the load balancing node is specifically used to convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node. After processing, the second business request; the sending module of the load balancing node, specifically used to send the processed second business request to the server node; the processing module of the load balancing module, specifically used to respond to the second business from the server node The source address is converted from the address of the service point node to the address of the load balancing node, and the first identifier is added to the second business response to obtain the processed second business response; the sending module of the load balancing node is specifically used to send the processed The second service response is sent to the client node, and the second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the processing module of the load balancing node is also used to detect whether the second service request contains the second identifier; the processing module of the load balancing node is specifically configured to detect if the second service request contains the second identifier. will come from The source address of the second service response of the server node is converted from the address of the service point node to the address of the load balancing node, and the first identifier is added to the second service response to obtain the processed second service response; sending by the load balancing node Module, specifically configured to send the processed second service response to the client node.

In a possible implementation, the processing module of the load balancing node is also configured to, if the second service request does not contain the second identifier, change the source address of the second service response from the server node to the service point node. The address is converted into the address of the load balancing node to obtain the processed second service response; the sending module of the load balancing node is also used to send the processed second service response to the client node.

The eighth aspect of the embodiment of the present application also provides a cloud service system. The load balancing node, client node and server node of the system all include a sending module and a processing module, wherein the sending module of the load balancing node is used to Send the first identifier to the client node and the server node; the processing module of the client node is used to convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, Obtain the processed first business request; the sending module of the client node is used to send the processed first business request to the server node; the processing module of the server node is used to process the processed first business request Process, obtain the first business response, convert the source address of the first business response from the address of the server node to the address of the load balancing node based on the first identifier, and obtain the processed first business response; the sending module of the server node , used to send the processed first service response to the client node.

In a possible implementation, the processing module of the load balancing node is specifically configured to convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, to the second The business request adds a first identifier to obtain the processed second business request; the sending module of the load balancing node is specifically used to send the processed second business request to the server node; the processing module of the load balancing node is specifically used to Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, add the first identifier to the second service response, and obtain the processed second service response; load balancing node The sending module is specifically used to send the processed second service response to the client node, and the second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the processing module of the load balancing node is also used to detect whether the second service request contains the second identifier; the processing module of the load balancing node is specifically configured to detect if the second service request contains the second identifier. Then convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, add the first identifier to the second service request, and obtain the processed second service request; load balancing The sending module of the node is specifically used to send the processed second service request to the server node; the processing module of the load balancing node is specifically used to send the source address of the second service response from the server node to the service point node. The address is converted into the address of the load balancing node, the first identifier is added to the second business response, and the processed second business response is obtained; the sending module of the load balancing node is specifically used to send the processed second business response to the client. end node.

In a possible implementation, the processing module of the load balancing node is also configured to, if the second service request does not contain the second identifier, change the destination address of the second service request from the client node to the load balancing node. The address is converted into the address of the server node to obtain the processed second business request; the sending module of the load balancing node is also used to send the processed second business request to the server node; the processing module of the load balancing node is also used to Used to convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node to obtain the processed second service response; the sending module of the load balancing node is also used to convert The processed second service response is sent to the client node.

The ninth aspect of the embodiment of the present application also provides a cloud service system. The load balancing node, client node and server node of the system all include a sending module and a processing module, wherein the sending module of the load balancing node is used to Send the first identification to the client node and the server node; the processing module of the client node is used to convert the source address of the first service request from the address of the client node to the address of the load balancing node based on the first identification, Convert the destination address of the first business request from the address of the load balancing node to the address of the server node to obtain the processed first business request; the sending module of the client node is used to send the processed first business request to the server node; the processing module of the server node is used to process the processed first service request, obtain the first service response, and change the source address of the first service response from the address of the server node based on the first identifier. Convert to the address of the load balancing node, convert the corresponding destination address of the first service from the address of the load balancing node to the address of the client node, and obtain the processed first service response; the sending module of the server node is used to The processed first service response is sent to the client node.

In one possible implementation, the processing module of the load balancing node is specifically configured to convert the source address of the second service request from the client node from the address of the client node to the address of the load balancing node, and convert the second service request from the client node to the address of the load balancing node. The destination address of the service request is converted from the address of the load balancing node to the address of the server node, and the first identifier is added to the second service request to obtain the processed second service request; the sending module of the load balancing node is specifically used to The processed second service request is sent to the server node; the processing module of the load balancing node is specifically used to convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node , convert the corresponding destination address of the second service from the address of the load balancing node to the address of the client node, add the first identifier to the second service response, and obtain the processed second service response; the sending module of the load balancing node, Specifically, it is used to send the processed second service response to the client node, and the second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the processing module of the load balancing node is also used to detect whether the second service request contains the second identifier; the processing module of the load balancing node is specifically configured to detect if the second service request contains the second identifier. Then the source address of the second service request from the client node is converted from the address of the client node to the address of the load balancing node, and the destination address of the second service request is converted from the address of the load balancing node to the address of the server node. Address, add the first identifier to the second service request to obtain the processed second service request; the sending module of the load balancing node, specifically used to send the processed second service request to the server node; processing of the load balancing node The module is specifically used to convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, and convert the corresponding destination address of the second service from the address of the load balancing node to The address of the client node adds the first identifier to the second service response to obtain the processed second service response; the sending module of the load balancing node is specifically used to send the processed second service response to the client node.

In a possible implementation, the processing module of the load balancing node is also configured to, if the second service request does not contain the second identifier, change the source address of the second service request from the client node to The address is converted into the address of the load balancing node, and the destination address of the second service request is converted from the address of the load balancing node to the address of the server node to obtain the processed second service request; the sending module of the load balancing node also uses The processed second service request is sent to the server node; the processing module of the load balancing node is also used to convert the source address of the second service response from the server node from the address of the service point node to the load balancing node address, convert the corresponding destination address of the second service from the address of the load balancing node to the address of the client node, and obtain the processed second service response; the sending module of the load balancing node is also used to convert the processed second service response. The second business response is sent to the client node.

A tenth aspect of the embodiment of the present application provides a load balancing node, which includes a memory and a processing The memory stores code, and the processor is configured to execute the code. When the code is executed, the load balancing node executes any one of the possible implementation methods of the fourth aspect, the fifth aspect, and the fifth aspect. Steps for implementing the load balancing node in any possible implementation manner, the sixth aspect, or any possible implementation manner of the sixth aspect.

An eleventh aspect of the embodiment of the present application provides a client node. The client node includes a memory and a processor; the memory stores code, and the processor is configured to execute the code. When the code is executed, the client node executes For example, in the fourth aspect, any one of the possible implementation methods of the fourth aspect, the fifth aspect, any one of the possible implementation methods of the fifth aspect, the sixth aspect, or any one of the possible implementation methods of the sixth aspect, the customer Steps for end node implementation.

A twelfth aspect of the embodiment of the present application provides a server node. The server node includes a memory and a processor; the memory stores code, and the processor is configured to execute the code. When the code is executed, the server node executes Such as the fourth aspect, any possible implementation method of the fourth aspect, the fifth aspect, any possible implementation method of the fifth aspect, the sixth aspect, or any possible implementation method of the sixth aspect. Steps for end node implementation.

A thirteenth aspect of the embodiments of the present application provides a computer storage medium. The computer storage medium stores one or more instructions. When executed by one or more computers, the instructions cause one or more computers to implement the fourth aspect, Any possible implementation manner in the fourth aspect, the fifth aspect, any possible implementation manner in the fifth aspect, the sixth aspect, or the method described in any possible implementation manner in the sixth aspect.

A fourteenth aspect of the embodiments of the present application provides a computer program product. The computer program product stores instructions. When the instructions are executed by a computer, the computer implements any one of the possible implementation methods of the fourth aspect and the fourth aspect. , the fifth aspect, any possible implementation manner of the fifth aspect, the sixth aspect, or the method described in any possible implementation manner of the sixth aspect.

In the cloud service system provided by the embodiment of the present application, the load balancing node can send the first identifier to the client node. Then, when the client node has a service to be processed, the client node can convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, and obtain the processed first service request, and directly sends the processed first service request to the server node. Then, the server node can process the processed first service request, obtain the first service response, and directly send the first service response to the client node. Finally, the client node can convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier to obtain the processed first service response. The response completes the response business. It can be seen that in the DNAT mode, the DANT processing for the first business request and the SNAT processing for the first business response, which were originally responsible for the load balancing node, are offloaded to the client node for execution by the load balancing node. In this way, the processing The first business request and the processed first business response can be directly transmitted between the client node and the server node without being processed by the load balancing node, which can effectively reduce the traffic passing through the load balancing node and reduce load balancing. Node bandwidth consumption, and can reduce business transmission delay and improve user experience.

Description of the drawings

Figure 1 is a schematic structural diagram of the cloud service system provided by the embodiment of the present application;

Figure 2 is a schematic structural diagram of a client node provided by an embodiment of the present application;

Figure 3 is a schematic structural diagram of a server node provided by an embodiment of the present application;

Figure 4 is a schematic flow chart of the business processing method based on the cloud service system provided by the embodiment of the present application;

Figure 5 is another structural schematic diagram of the cloud service system provided by the embodiment of the present application;

Figure 6 is another schematic flowchart of a business processing method based on a cloud service system provided by an embodiment of the present application;

Figure 7 is another structural schematic diagram of the cloud service system provided by the embodiment of the present application;

Figure 8 is another schematic flowchart of a business processing method based on a cloud service system provided by an embodiment of the present application;

Figure 9 is another structural schematic diagram of the cloud service system provided by the embodiment of the present application;

Figure 10 is a schematic structural diagram of a load balancing node provided by an embodiment of the present application;

Figure 11 is another schematic structural diagram of a client node provided by an embodiment of the present application;

Figure 12 is another structural schematic diagram of a server node provided by an embodiment of the present application;

Figure 13 is another structural schematic diagram of a load balancing node provided by an embodiment of the present application;

Figure 14 is another schematic structural diagram of a client node provided by an embodiment of the present application;

Figure 15 is another schematic structural diagram of a server node provided by an embodiment of the present application.

Detailed ways

The embodiments of this application provide a cloud service system and a business processing method based on the cloud service system, which can effectively reduce the traffic passing through the load balancing node, reduce the bandwidth consumption of the load balancing node, and can reduce the business transmission delay and improve the user experience. .

The terms "first", "second", etc. in the description and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the terms so used are interchangeable under appropriate circumstances, and are merely a way of distinguishing objects with the same attributes in describing the embodiments of the present application. Furthermore, the terms "include" and "having" and any variations thereof, are intended to cover non-exclusive inclusions, such that a process, method, system, product or apparatus comprising a series of elements need not be limited to those elements, but may include not explicitly other elements specifically listed or inherent to such processes, methods, products or equipment.

In cloud scenarios such as public cloud, private cloud, and hybrid cloud, cloud service systems generally provide users with services of a certain scale through cluster deployment. In order to prevent the load from being concentrated on a certain node or certain nodes in the cluster, load balancing nodes can be added in the cluster to distribute the load of each node in the cluster, thereby optimizing the performance of the entire cloud service system.

Currently, load balancing nodes can implement load balancing in non-DR mode. Non-DR mode includes DNAT mode and FULLNAT mode. In the DNAT mode, the load balancing node can perform DNAT processing on the service request sent by the client node, and send the processed service request to the server node. For the business response returned by the server node, the load balancing node can perform SNAT processing on the business response, and send the processed business request to the client node, so that the client node completes the corresponding business. In FULLNAT mode, the load balancing node can perform SNAT+DNAT processing on the service request sent by the client node, and send the processed service request to the server node. For the business response returned by the server node, the load balancing node can perform DNAT+SNAT processing on the business response, and send the processed business request to the client node, so that the client node completes the corresponding business.

It can be seen that, whether in DNAT mode or FULLNAT mode, the traffic between client nodes and server nodes (business requests, business responses, etc.) will continue to pass through the load balancing node. However, cloud service systems are often deployed with a large number of client nodes and server nodes, and huge traffic needs to be processed by load balancing nodes, resulting in excessive bandwidth consumption of load balancing nodes and long service transmission delays. , resulting in poor user experience.

Furthermore, due to the limited hardware resources of the load balancing node, it is unable to handle excessively large traffic, resulting in a limit on the business volume that the cloud service system can bear. In the face of the increasing business needs of users, it is often difficult to quickly expand the capacity. It is unable to provide high-quality cloud services to more and more users.

In order to solve the above problems, embodiments of the present application provide a business processing method, which can be implemented through a cloud service system, as shown in Figure 1 (Figure 1 is a schematic structural diagram of the cloud service system provided by the embodiment of the present application), The cloud service system can include load balancing nodes, multiple client nodes, and multiple server nodes. These three types of physical nodes will be introduced below:

Load balancing node, this node can be a physical machine (also called a server), or it can be built from multiple physical machines, or it can be a cloud instance (for example, a virtual machine or a container, etc.) on a certain physical machine. This node can also be presented as a cloud platform, which can uniformly manage multiple client nodes and multiple server nodes. For example, for a client node, the load balancing node can allocate a server node to the client node so that the server node provides business processing for the client node and implements the client node and load balancing node. And the handshake between the server nodes. For another example, the load balancing node can indirectly implement business interaction between the client node and the server node based on the DNAT mode or the FULLNAT mode. For another example, the load balancing node can offload the business interaction between the client node and the server node to the client node and the server node, and the two nodes directly implement the business interaction.

Client node, which is usually a physical machine on which one or more virtual clients (which can also be understood as virtual instances such as virtual machines or containers) are deployed. As shown in Figure 2 (Figure 2 is a schematic structural diagram of a client node provided by an embodiment of the present application), based on virtualization technology, the client node can be divided into three layers. The bottom layer is the hardware resources of the node, and the middle layer is the virtual machine. Monitor (hypervisor), with multiple virtual clients at the top. Among them, the virtual client can initiate various requests, such as handshake requests, business requests, etc., and the virtual machine monitor can process the requests from the virtual client and send the processed requests to the outside world (of course, it can also not The request is processed and sent directly to the outside world). The virtual machine monitor can also process the response from the outside world and send the processed response to the virtual client (of course, the response can also be sent directly to the virtual client without processing the request. end).

Server node is usually a physical machine deployed with one or more virtual servers (which can also be understood as virtual instances such as virtual machines or containers). As shown in Figure 3 (Figure 3 is a schematic structural diagram of a server node provided by an embodiment of the present application), based on virtualization technology, the server node can be divided into three layers. The bottom layer is the hardware resources of the node, and the middle layer is the virtual machine. Monitor (hypervisor), with multiple virtual servers at the top. Among them, the virtual server can receive and process various requests from the outside world, such as handshake requests and business requests, etc. The virtual machine monitor can process requests from the outside world and send the processed requests to the virtual server ( Of course, you can also send the request directly to the virtual server without processing the request). The virtual machine monitor can also process the response from the virtual server and send the processed response to the outside world (of course, you can also not process the response. sent directly to the outside world).

In order to further understand the workflow of the above-mentioned cloud service system, the following will introduce the workflow of the cloud service system in three situations. First, the workflow of the cloud service system in the first case is introduced. Figure 4 is a schematic flow chart of the business processing method based on the cloud service system provided by the embodiment of the present application. As shown in Figure 4, the method includes :

401. The client node sends the handshake request to the load balancing node.

In this embodiment, when a client node has business requirements, the client node can initiate a handshake to the load balancing node, that is, the client node sends a handshake request to the load balancing node. It should be noted that the handshake request includes The source address is the address of the client node, and the destination address is the address of the load balancing node.

For example, as shown in Figure 5 (Figure 5 is another structural schematic diagram of the cloud service system provided by this embodiment of the present application), the customer Virtual client 1 of end node 1 can first generate a handshake request. The source address of the handshake request is the address of virtual client 1 of client node 1. The destination address of the handshake request is the front-end address of the load balancing node. Then, virtual client 1 of client node 1 sends a handshake request to the virtual machine monitor of client node 1, so that the virtual machine monitor of client node 1 then sends a handshake request to the load balancing node.

402. The load balancing node selects the server node from the server node resource pool based on the handshake request, then performs DNAT processing on the handshake request, obtains the processed handshake request, and sends the processed handshake request to the server node.

After receiving the handshake request from the client node, the load balancing node can determine that the client node has business needs based on the handshake request, and then according to certain rules (for example, preset load balancing algorithm, etc.) in the server node resource pool (including multiple server nodes), select a server node to perform business processing for the client node. Then, the load balancing node can also perform DNAT processing on the handshake request, that is, convert the destination address of the handshake request from the address of the load balancing node to the address of the server node, and obtain the processed handshake request. The load balancing node can then send the processed handshake request to the server node.

Still as in the above example, after receiving the handshake request, the load balancing node can select the virtual server 1 of the server node 1 to serve the virtual client 1 of the client node 1. Then, the load balancing node can perform DNAT processing on the handshake request to obtain the processed handshake request. At this time, the source address of the processed handshake request is still the address of virtual client 1 of client node 1. The processed handshake request The destination address is replaced with the address of virtual server 1 of server node 1. Then, the load balancing node can send the processed handshake request to the virtual machine monitor of server node 1.

403. The server node processes the processed handshake request, obtains a handshake response, and sends the handshake response to the load balancing node.

After obtaining the processed handshake request from the load balancing node, the server node can perform a series of processes on the processed handshake request to obtain a handshake response, and send the handshake response to the load balancing node. It should be noted that the source address of the handshake response is the address of the server node, and the destination address of the handshake response is the address of the client node.

Still as in the above example, after receiving the processed handshake request, the virtual machine monitor of the server node 1 can send the processed handshake request to the virtual server 1 of the server node 1 and record the same origin and same destination logic. Then, the virtual server 1 of the server node 1 can process the processed handshake request and obtain a handshake response. The source address of the handshake response is the address of the virtual server 1 of the server node 1, and the destination address of the handshake response is the client. The address of virtual client 1 for end node 1. Then, the virtual server 1 of the server node 1 may return the handshake response to the virtual machine monitor of the server node 1 . Due to the existence of the same origin and same destination logic, when the virtual machine monitor of server node 1 sends a handshake response to the outside world, the handshake response can be received by the load balancing node first.

404. The load balancing node performs SNAT processing on the handshake response, obtains the processed handshake response, and sends the processed handshake response to the client node.

After obtaining the handshake response from the server node, the load balancing node can perform SNAT processing on the handshake response, that is, convert the source address of the handshake response from the address of the server node to the address of the load balancing node to obtain the processed handshake response. The load balancing node then sends the processed handshake response to the client node. At this point, the client node and the server node complete the handshake.

Still as in the above example, after the load balancing node receives the handshake response, it can perform SNAT processing on the handshake response to obtain the processed handshake response. At this time, the source address of the processed handshake response is replaced by the front-end address of the load balancing node, and the processing The destination address of the subsequent handshake response is still the address of virtual client 1 of client node 1. Then, load balancing The node may send the processed handshake response to the virtual machine monitor of client node 1. Then, the virtual machine monitor of client node 1 can send the processed handshake response to virtual client 1 of client node 1. At this point, the handshake is completed.

405. The client node sends the second service request to the load balancing node.

When there is a certain pending service, the client node can send a second service request indicating the pending service to the load balancing node. It should be noted that the source address included in the second service request is the client. The address of the end node, and the destination address is the address of the load balancing node.

Still as in the above example, virtual client 1 of client node 1 can first generate a second service request. The source address of the second service request is the address of virtual client 1 of client node 1, and the destination address of the second service request is The front-end address of the load balancing node. Then, the virtual client 1 of the client node 1 sends the second service request to the virtual machine monitor of the client node 1, so that the virtual machine monitor of the client node 1 then sends the second service request to the load balancing node. .

406. The load balancing node performs DNAT processing on the second service request, obtains the processed second service request, and sends the processed second service request to the server node.

After obtaining the second service request from the client node, the load balancing node can perform DNAT processing on the second service request, that is, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, Obtain the processed second service request. Then, the load balancing node may send the processed second service request to the server node.

Still as in the above example, after receiving the second service request, the load balancing node can perform DNAT processing on the second service request to obtain the processed second service request. At this time, the source address of the processed second service request is still The address of the virtual client 1 of the client node 1, and the destination address of the processed second service request is replaced by the address of the virtual server 1 of the server node 1. Then, the load balancing node may send the processed second service request to the virtual machine monitor of the server node 1 .

407. The server node processes the processed second service request, obtains the second service response, and sends the second service response to the load balancing node.

After obtaining the processed second service request from the load balancing node, the server node can perform a series of processes on the processed second service request to obtain a second service response, and send the second service response to the load Balanced node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node.

Still as in the above example, after receiving the processed second service request, the virtual machine monitor of the server node 1 can send the processed second service request to the virtual server 1 of the server node 1 . Then, the virtual server 1 of the server node 1 can process the processed second service request to obtain a second service response. The source address of the second service response is the address of the virtual server 1 of the server node 1. The destination address of the second service response is the address of virtual client 1 of client node 1. Then, the virtual server 1 of the server node 1 may return the second service response to the virtual machine monitor of the server node 1 . Due to the existence of same-origin and same-destination logic, when the virtual machine monitor of server node 1 sends a second service response to the outside, the second service response can be received by the load balancing node with priority.

408. The load balancing node detects whether the second service request contains the second identifier.

409. If the second service request contains the second identifier, the load balancing node performs SNAT processing on the second service response, adds the first identifier to the second service response, obtains the processed second service response, and adds the processed second service response to the second service response. Business response sent to the client node.

After obtaining the second service response from the server node, the load balancing node can first detect whether the second service request contains the second identifier to determine whether to offload subsequent business interactions to the client node and the server node. The two nodes themselves directly complete the business interaction. Then, there are two situations:

(1) If the second business request contains the second identifier, it means that the load balancing node can offload subsequent business interactions to the client node and the server node, and the two nodes themselves directly complete the business interaction, so the load The balancing node can perform SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node, and add a first identifier (also known as Metadata) is processed and the second service response is obtained. Then, the load balancing node sends the processed second service response to the client node. Thereafter, the client node can save the first identification based on the processed second service response and complete the corresponding service.

Still as in the above example, after receiving the second service response, the load balancing node can first determine whether the second service request contains the second identifier. If it is determined that the second service request contains the second identifier, then perform SNAT processing on the second service response. , and add the first identifier to the second service response, thereby obtaining the processed second service response. At this time, the source address of the processed second service response is replaced by the front-end address of the load balancing node, and the processed second service response The destination address of the second service response is still the address of virtual client 1 of client node 1. Then, the load balancing node may send the processed second service response to the virtual machine monitor of the client node 1 . Then, the virtual machine monitor of client node 1 can parse the processed second service response to save the first identification, and send the processed second service response to virtual client 1 of client node 1, so that the client The virtual client 1 of the end node 1 completes the corresponding service based on the processed second service response.

(2) If the second service request does not include the second identifier, it means that the load balancing node cannot offload subsequent business interactions to the client node and the server node, so the load balancing node can perform SNAT processing on the second service response. , that is, convert the source address of the second service response from the address of the service point node to the address of the load balancing node, obtain the processed second service response, and send the processed second service response to the client node. Thereafter, the client node can complete the corresponding service based on the processed second service response.

It should be understood that in this embodiment, only the first identifier is included in the second service response and is delivered to the client node by the load balancing node (as the service proceeds) as an example for schematic introduction, and does not describe the load balancing node. The method of delivering the first identifier to the client node constitutes a restriction. For example, after completing the handshake, the load balancing node can directly send the first identifier to the client node (without accompanying the business), etc.

It should also be understood that in this embodiment, only the load balancing node determines whether the second service request contains the second identifier and thereby determines whether to issue the first identifier is used as an example for schematic introduction. In actual applications, the load balancing node also The first identifier and so on can be directly delivered to the client node without any judgment.

It should also be understood that in this embodiment, only the second identifier is carried in the second service request for schematic introduction. In actual applications, the second identifier can also be carried in the handshake request and other service requests, and so on.

410. The client node performs DNAT processing on the first service request based on the first identifier, obtains the processed first service request, and sends the processed first service request to the server node.

When there is another service to be processed, the client node can first obtain the first service request indicating the service to be processed. It should be noted that the source address included in the first service request is the address of the client node. , the destination address is the address of the load balancing node. Due to the existence of the first identifier, the client node can perform DNAT processing on the first service request based on the first identifier, that is, convert the destination address of the first service request from the address of the load balancing node to the server node. The address after which the first business request is processed. Then, the client node can directly send the processed first service request to the server node.

Still as in the above example, virtual client 1 of client node 1 can first generate a first service request. The source address of the first service request is the address of virtual client 1 of client node 1, and the destination address of the first service request is The front-end address of the load balancing node. Then, the virtual client 1 of the client node 1 sends the first service request to the virtual machine monitor of the client node 1 . Then, the virtual machine monitor of client node 1 can perform DNAT processing on the first service request based on the first identifier to obtain the processed first service request. At this time, the source address of the processed first service request is still The address of the virtual client 1 of the client node 1, and the destination address of the processed first service request is replaced by the address of the virtual server 1 of the server node 1. Subsequently, the virtual machine monitor of the client node 1 can directly send the processed first service request to the virtual machine monitor of the server node 1 .

411. The server node processes the processed first service request, obtains the first service response, and sends the first service response to the client node.

After obtaining the processed first service request from the client node, the server node can perform a series of processes on the processed first service request to obtain the first service response and directly send the first service response. to the client node. It should be noted that the source address of the first service response is the address of the server node, and the destination address of the first service response is the address of the client node.

Still as in the above example, after receiving the processed first service request, the virtual machine monitor of the server node 1 can clear the same origin and same destination logic, and send the processed first service request to the virtual service of the server node 1 Terminal 1. Then, the virtual server 1 of the server node 1 can process the processed first service request to obtain a first service response. The source address of the first service response is the address of the virtual server 1 of the server node 1. The destination address of a service response is the address of virtual client 1 of client node 1. Then, the virtual server 1 of the server node 1 may return the first service response to the virtual machine monitor of the server node 1 . Since the same origin and same destination logic no longer exists, when the virtual machine monitor of the server node 1 sends the first service response to the outside, the first service response is directly received by the virtual machine monitor of the client node 1.

412. The client node performs SNAT processing on the first service response based on the first identifier, and obtains the processed first service response.

After obtaining the first service response from the server node, due to the existence of the first identifier, the client node can perform SNAT processing on the first service response based on the first identifier, that is, the source address of the first service response, from the service The address of the point node is converted into the address of the load balancing node, and the processed first service response is obtained. Thereafter, the client node can complete the corresponding service based on the processed first service response.

Still as in the above example, after receiving the first service response, the virtual machine monitor of client node 1 can perform SNAT processing on the first service response based on the first identifier, thereby obtaining the processed first service response. At this time, The source address of the processed first service response is changed to the front-end address of the load balancing node, and the destination address of the processed first service response is still the address of virtual client 1 of client node 1. Then, the virtual machine monitor of the client node 1 may send the processed first service response to the virtual client 1 of the client node 1 , so that the virtual client 1 of the client node 1 is based on the processed first service Respond to complete the corresponding business.

It should be understood that the address mentioned in this embodiment can be understood as an Internet Protocol (IP) address.

In the cloud service system provided by the embodiment of the present application, the load balancing node can send the first identifier to the client node. That Well, when the client node has a service to be processed, the client node can convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, and obtain the processed first service request, and directly sends the processed first service request to the server node. Then, the server node can process the processed first service request, obtain the first service response, and directly send the first service response to the client node. Finally, the client node can convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier to obtain the processed first service response. The response completes the response business. It can be seen that in the DNAT mode, the DANT processing for the first business request and the SNAT processing for the first business response, which were originally responsible for the load balancing node, are offloaded to the client node for execution by the load balancing node. In this way, the processing The first business request and the processed first business response can be directly transmitted between the client node and the server node without being processed by the load balancing node, which can effectively reduce the traffic passing through the load balancing node and reduce load balancing. Node bandwidth consumption, and can reduce business transmission delay and improve user experience.

Furthermore, in the embodiments of this application, since the load balancing nodes no longer need to process huge traffic, the upper limit of the business volume that the cloud service system can bear can be increased. In the face of the increasing business needs of users, it can be added to the existing Rapidly expand the system capacity based on the system to provide high-quality cloud services to more and more users.

The above is a detailed description of the workflow of the cloud service system in the first case. The workflow of the cloud service system in the second case will be introduced below. Figure 6 is another schematic flowchart of a business processing method based on a cloud service system provided by an embodiment of the present application. As shown in Figure 6, the method includes:

601. The client node sends the handshake request to the load balancing node.

602. The load balancing node selects the server node from the server node resource pool based on the handshake request, then performs DNAT processing on the handshake request, obtains the processed handshake request, and sends the processed handshake request to the server node.

603. The server node processes the processed handshake request, obtains a handshake response, and sends the handshake response to the load balancing node.

604. The load balancing node performs SNAT processing on the handshake response, obtains the processed handshake response, and sends the processed handshake response to the client node.

605. The client node sends the second service request to the load balancing node.

For an introduction to steps 601 to 605, reference may be made to the relevant descriptions of steps 401 to 405 in the embodiment shown in FIG. 4 , which will not be described again here.

606. The load balancing node detects whether the second service request contains the second identifier.

607. If the second service request contains the second identifier, the load balancing node performs DNAT processing on the second service request, adds the first identifier to the second service request, obtains the processed second service request, and adds the processed second service request to the second service request. Business requests are sent to the server node.

608. The server node processes the processed second service request, obtains the second service response, and sends the second service response to the load balancing node.

609. The load balancing node performs SNAT processing on the second service response, adds the first identifier to the second service response, obtains the processed second service response, and sends the processed second service response to the client node.

After obtaining the second service request from the client node, the load balancing node can first detect whether the second service request contains the second identifier to determine whether to offload subsequent business interactions to the client node and the server node. The two nodes themselves directly complete the business interaction. Then, there are two situations:

(1) If the second business request contains the second identifier, it means that the load balancing node can offload subsequent business interactions to the client node and the server node, and the two nodes themselves directly complete the business interaction, so the load The balancing node can perform DNAT processing on the second service request, that is, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and add the first identifier to the second service request. After processing second business request. Then, the load balancing node may send the processed second service request to the server node.

After obtaining the processed second service request from the load balancing node, the server node can save the first identifier, and perform a series of processes on the processed second service request, thereby obtaining the second service response, and converting the second service request to the second service request. The second service response is sent to the load balancing node. It should be noted that the source address of the second service response is the address of the server node, and the destination address of the second service response is the address of the client node.

After obtaining the second service response from the server node, the load balancing node can perform SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node, and Add the first identifier to the second service response to obtain the processed second service response. Then, the load balancing node sends the processed second service response to the client node. Thereafter, the client node can save the first identification based on the processed second service response and complete the corresponding service.

For example, as shown in Figure 7 (Figure 7 is another structural schematic diagram of the cloud service system provided by the embodiment of the present application), after receiving the second service request, the load balancing node may first determine whether the second service request contains the second service request. identification, if it is determined that the second service request contains the second identification, DNAT processing is performed on the second service request, and the first identification is added to the second service request to obtain the processed second service request. At this time, after processing The source address of the second service request is still the address of the virtual client 1 of the client node 1, and the destination address of the processed second service request is changed to the address of the virtual server 1 of the server node 1. Then, the load balancing node may send the processed second service request to the virtual machine monitor of the server node 1 .

After receiving the processed second service request, the virtual machine monitor of the server node 1 can parse the processed second service request to save the first identification, and send the processed second service request to the server node 1 Virtual server 1. Then, the virtual server 1 of the server node 1 can process the processed second service request to obtain a second service response. The source address of the second service response is the address of the virtual server 1 of the server node 1. The destination address of the second service response is the address of virtual client 1 of client node 1. Then, the virtual server 1 of the server node 1 may return the second service response to the virtual machine monitor of the server node 1 . Due to the existence of same-origin and same-destination logic, when the virtual machine monitor of server node 1 sends a second service response to the outside, the second service response can be received by the load balancing node with priority.

After receiving the second service response, the load balancing node can perform SNAT processing on the second service response, and add the first identifier to the second service response, thereby obtaining the processed second service response. At this time, the processed The source address of the second service response is changed to the front-end address of the load balancing node, and the destination address of the processed second service response is still the address of virtual client 1 of client node 1. Then, the load balancing node may send the processed second service response to the virtual machine monitor of the client node 1 . Then, the virtual machine monitor of client node 1 can parse the processed second service response to save the first identification, and send the processed second service response to virtual client 1 of client node 1, so that the client The virtual client 1 of the end node 1 completes the corresponding service based on the processed second service response.

(2) If the second service request does not include the second identifier, it means that the load balancing node cannot offload subsequent business interactions to the client node and the server node, so the load balancing node can perform DNAT processing on the second service request. , that is, the destination address of the second business request is converted from the address of the load balancing node to the address of the server node, and after processing second business request. Then, the load balancing node may send the processed second service request to the server node.

After obtaining the processed second service request from the load balancing node, the server node can perform a series of processes on the processed second service request to obtain a second service response, and send the second service response to the load Balanced node.

After obtaining the second service response from the server node, the load balancing node can perform SNAT processing on the second service response, that is, convert the source address of the second service response from the address of the server node to the address of the load balancing node, obtaining The processed second business response. Then, the load balancing node sends the processed second service response to the client node. Thereafter, the client node completes the corresponding service based on the processed second service response.

It should be understood that in this embodiment, it is only taken as an example that the first identifier is included in the second service request and the second service response and is delivered by the load balancing node to the client node and the server node respectively (as the service proceeds). A schematic introduction does not limit the way in which the load balancing node issues the first identifier to the client node. For example, after completing the handshake, the load balancing node can directly send the first identifier to the client node and the server node ( It does not need to be accompanied by business) etc.

It should also be understood that in this embodiment, only the load balancing node determines whether the second service request contains the second identifier and thereby determines whether to issue the first identifier is used as an example for schematic introduction. In actual applications, the load balancing node also The first identifier and so on can be directly issued to the client node and server node without any judgment.

It should also be understood that in this embodiment, only the second identifier is carried in the second service request for schematic introduction. In actual applications, the second identifier can also be carried in the handshake request and other service requests, and so on.

610. The client node performs DNAT processing on the first service request based on the first identifier, obtains the processed first service request, and sends the processed first service request to the server node.

For the introduction of step 610, please refer to the relevant description of step 410 in the embodiment shown in FIG. 4, which will not be described again here.

611. The server node processes the processed first service request, obtains the first service response, performs SNAT processing on the first service response based on the first identifier, obtains the processed first service response, and sends the processed first service response. A business response is sent to the client node.

After obtaining the processed first service request from the client node, the server node can perform a series of processes on the processed first service request, thereby obtaining a first service response. It should be noted that the source address of the first service response is the address of the server node, and the destination address of the first service response is the address of the client node.

Then, due to the existence of the first identifier, the server node can perform SNAT processing on the first service response based on the first identifier, that is, the source address of the first service response is converted from the address of the service point node to the address of the load balancing node. , get the processed first business response. Then, the server node can directly send the processed first service response to the client node. Thereafter, the client node can complete the corresponding service based on the processed first service response.

Still as in the above example, after receiving the processed first service request, the virtual machine monitor of the server node 1 can clear the same origin and same destination logic, and send the processed first service request to the virtual service of the server node 1 Terminal 1. Then, the virtual server 1 of the server node 1 can process the processed first service request to obtain a first service response. The source address of the first service response is the address of the virtual server 1 of the server node 1. The destination address of a service response is the address of virtual client 1 of client node 1. Subsequently, the virtual server 1 of the server node 1 may return the first service response to the virtual machine monitor of the server node 1 .

Then, the virtual machine monitor of the server node 1 can perform SNAT processing on the first service response based on the first identifier, thereby obtaining the processed first service response. At this time, the source address of the processed first service response is replaced. for load balancing section The front-end address of the node, the destination address of the processed first service response is still the address of virtual client 1 of client node 1. Since the same origin and same destination logic no longer exists, when the virtual machine monitor of the server node 1 sends the processed first service response to the outside, the processed first service response is directly received by the virtual machine monitor of the client node 1.

Then, the virtual machine monitor of the client node 1 sends the processed first service response to the virtual client 1 of the client node 1, so that the virtual client 1 of the client node 1 is based on the processed first service response. Complete the corresponding business.

In the cloud service system provided by the embodiment of the present application, the load balancing node can send the first identifier to the client node and the server node. Then, when the client node has a service to be processed, the client node can convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, and obtain the processed first service request, and sends the processed first service request to the server node. Then, the server node can process the processed first service request to obtain the first service response, and convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier, The processed first service response is obtained, and the processed first service response is sent to the client node to complete the response service based on the processed first service response. It can be seen that in the DNAT mode, the DANT processing for the first service request that was originally responsible for the load balancing node is offloaded to the client node for execution by the load balancing node, and the SNAT for the first service response that was originally responsible for the load balancing node The processing is offloaded by the load balancing node to the server node for execution. In this way, the processed first business request and the processed first business response can be directly transmitted between the client node and the server node without going through Load balancing node processing can effectively reduce the traffic passing through the load balancing node, reduce the bandwidth consumption of the load balancing node, reduce business transmission delay, and improve user experience.

Furthermore, in the embodiments of this application, since the load balancing nodes no longer need to process huge traffic, the upper limit of the business volume that the cloud service system can bear can be increased. In the face of the increasing business needs of users, it can be added to the existing Rapidly expand the system capacity based on the system to provide high-quality cloud services to more and more users.

The above is a detailed description of the workflow of the cloud service system in the second case. The workflow of the cloud service system in the third case will be introduced below. Figure 8 is another schematic flowchart of a business processing method based on a cloud service system provided by an embodiment of the present application. As shown in Figure 8, the method includes:

801. The client node sends the handshake request to the load balancing node.

802. The load balancing node selects a server node from the server node resource pool based on the handshake request, performs SNAT processing and DNAT processing on the handshake request, obtains the processed handshake request, and sends the processed handshake request to the server node.

803. The server node processes the processed handshake request, obtains a handshake response, and sends the handshake response to the load balancing node.

804. The load balancing node performs SANT processing and DNAT processing on the handshake response, obtains the processed handshake response, and sends the processed handshake response to the client node.

For the introduction of steps 801 to 804, reference may be made to the relevant descriptions of steps 601 to 604 in the embodiment shown in FIG. 6 , which will not be described again here.

It should be noted that the difference between step 602 and step 802 is that in step 602, the load balancing node only performs DNAT processing on the handshake request to obtain the processed handshake request, while in step 802, the load balancing node performs DNAT processing on the handshake request and SNAT processing, that is, converting the source address of the handshake request from the address of the client node to the load balancer The address of the node converts the destination address of the handshake request from the address of the load balancing node to the address of the server node to obtain the processed handshake request.

The difference between step 604 and step 804 is that in step 604, the load balancing node only performs SNAT processing on the handshake response to obtain the processed handshake response, while in step 804, the load balancing node performs DNAT processing and SNAT processing on the handshake request, that is, the handshake The source address of the response is converted from the address of the server node to the address of the load balancing node, and the destination address of the handshake response is converted from the address of the load balancing node to the address of the client node to obtain the processed handshake response.

805. The client node sends the second service request to the load balancing node.

806. The load balancing node detects whether the second service request includes the second identifier.

807. If the second service request contains the second identifier, the load balancing node performs DNAT processing and SNAT processing on the second service request, adds the first identifier to the second service request, obtains the processed second service request, and will The second service request is sent to the server node.

808. The server node processes the processed second service request, obtains the second service response, and sends the second service response to the load balancing node.

809. The load balancing node performs DNAT processing and SNAT processing on the second service response, adds the first identifier to the second service response, obtains the processed second service response, and sends the processed second service response to the client node. .

For an introduction to steps 805 to 809, reference may be made to the relevant descriptions of steps 605 to 609 in the embodiment shown in FIG. 6 , which will not be described again here.

It should be noted that if the second service request contains the second identifier, the difference between step 607 and step 807 is that in step 607, the load balancing node only performs DNAT processing on the second service request and adds the first identifier to the second service request. , the processed second service request can be obtained, and in step 807, the load balancing node performs DNAT processing and SNAT processing on the second service request, that is, converting the source address of the second service request from the address of the client node to the load balancing node address, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and add the first identifier to the second service request, thereby obtaining the processed second service request.

If the second service request contains the second identifier, the difference between step 609 and step 809 is that in step 609, the load balancing node only performs SNAT processing on the second service response and adds the first identifier to the second service response, and the processed The second service response, and in step 809, the load balancing node performs DNAT processing and SNAT processing on the second service request, that is, converts the source address of the second service response from the address of the server node to the address of the load balancing node, and converts the second service request to the address of the load balancing node. The destination address of the second service response is converted from the address of the load balancing node to the address of the client node, and the first identifier is added to the second service response to obtain a processed second service response.

Further, if the second service request does not contain the second identifier, in the embodiment shown in Figure 6, the load balancing node only performs DNAT processing on the second service request, and the processed second service request can be obtained, while the implementation shown in Figure 8 In the example, the load balancing node performs DNAT processing and SNAT processing on the second service request, that is, converts the source address of the second service request from the address of the client node to the address of the load balancing node, and converts the destination address of the second service request from The address of the load balancing node is converted into the address of the server node, and the processed second service request is obtained.

Similarly, if the second service request does not contain the second identifier, the load balancing node in the embodiment shown in Figure 6 only performs DNAT processing on the second service response to obtain the processed second service response, while the implementation shown in Figure 8 In the example, the load balancing node performs DNAT processing and SNAT processing on the second service request, that is, converts the source address of the second service response from the address of the server node to the address of the load balancing node, and converts the destination address of the second service response from The address of the load balancing node is translated to The address of the client node and the processed second service response.

810. The client node performs DNAT processing and SNAT processing on the first service request based on the first identifier, obtains the processed first service request, and sends the processed first service request to the server node.

811. The server node processes the processed first service request, obtains the first service response, performs DNAT processing and SNAT processing on the first service response based on the first identifier, obtains the processed first service response, and processes the first service request. The first service response is sent to the client node.

For the introduction of steps 810 to 811, please refer to the relevant descriptions of steps 610 to 611 in the embodiment shown in FIG. 6, which will not be described again here.

It should be noted that the difference between step 610 and step 810 is that in step 610, the client node only performs DNAT processing on the first service request based on the first identifier, and can obtain the processed first service request, while in step 810, the client node The node performs DNAT processing and SNAT processing on the first service request based on the first identifier, that is, based on the first identifier, the source address of the first service request is converted from the address of the client node to the address of the load balancing node, and the first service request is The destination address is converted from the address of the load balancing node to the address of the server node, and the processed first service request is obtained.

The difference between step 611 and step 811 is that in step 611, the server node only performs SNAT processing on the first service response based on the first identifier to obtain the processed first service response, while in step 811, the server node only performs SNAT processing on the first service response based on the first identifier. Perform DNAT processing and SNAT processing on the first service response, that is, convert the source address of the first service response from the address of the server node to the address of the load balancing node, and convert the corresponding destination address of the first service from the address of the load balancing node Convert it to the address of the client node and get the processed first service response.

For example, as shown in Figure 9 (Figure 9 is a schematic structural diagram of the cloud service system provided by the embodiment of the present application), the virtual client 1 of the client node 1 can first generate a first service request, the source address of the first service request is the address of virtual client 1 of client node 1, and the destination address of the first service request is the front-end address of the load balancing node. Then, the virtual client 1 of the client node 1 sends the first service request to the virtual machine monitor of the client node 1 . Then, the virtual machine monitor of the client node 1 can perform DNAT processing and SNAT processing on the first service request based on the first identifier to obtain the processed first service request. At this time, the source of the processed first service request The address is replaced with the backend address of the load balancing node, and the destination address of the processed first service request is replaced with the address of virtual server 1 of server node 1. Subsequently, the virtual machine monitor of the client node 1 can directly send the processed first service request to the virtual machine monitor of the server node 1 .

After receiving the processed first service request, the virtual machine monitor of the server node 1 may send the processed first service request to the virtual server 1 of the server node 1 . Then, the virtual server 1 of the server node 1 can process the processed first service request to obtain a first service response. The source address of the first service response is the address of the virtual server 1 of the server node 1. The destination address of a service response is the backend address of the load balancing node. Subsequently, the virtual server 1 of the server node 1 may return the first service response to the virtual machine monitor of the server node 1 .

Then, the virtual machine monitor of the server node 1 can perform DANT processing and SNAT processing on the first service response based on the first identifier, thereby obtaining the processed first service response. At this time, the processed first service response The source address is replaced with the front-end address of the load balancing node, and the destination address of the processed first service response is replaced with the address of virtual client 1 of client node 1. When the virtual machine monitor of the server node 1 sends the processed first service response to the outside, the processed first service response is directly received by the virtual machine monitor of the client node 1 .

Then, the virtual machine monitor of client node 1 sends the processed first service response to the virtual machine monitor of client node 1. The virtual client 1 of the client node 1 completes the corresponding service based on the processed first service response.

In the cloud service system provided by the embodiment of the present application, the load balancing node can send the first identifier to the client node and the server node. Then, when the client node has a service to be processed, the client node can convert the source address of the first service request from the address of the client node to the address of the load balancing node based on the first identifier, and convert the source address of the first service request to the address of the load balancing node. The destination address is converted from the address of the load balancing node to the address of the server node, the processed first service request is obtained, and the processed first service request is sent to the server node. Then, the server node can process the processed first service request to obtain the first service response, and convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier, Convert the corresponding destination address of the first service from the address of the load balancing node to the address of the client node, obtain the processed first service response, and send the processed first service response to the client node to perform processing based on The first business response after the completion of the response business. It can be seen that in the FULLDNAT mode, the DANT processing and SNAT processing for the first business request that were originally responsible for the load balancing node are offloaded to the client node for execution by the load balancing node. The DANT processing and SNAT processing of the response are offloaded to the server node for execution by the load balancing node. In this way, the processed first business request and the processed first business response can be directly processed between the client node and the server node. Inter-transmission does not need to be processed by the load balancing node, which can effectively reduce the traffic passing through the load balancing node, reduce the bandwidth consumption of the load balancing node, reduce business transmission delay, and improve user experience.

Furthermore, in the embodiments of this application, since the load balancing nodes no longer need to process huge traffic, the upper limit of the business volume that the cloud service system can bear can be increased. In the face of the increasing business needs of users, it can be added to the existing Rapidly expand the system capacity based on the system to provide high-quality cloud services to more and more users.

The above is a detailed description of the business processing method based on the cloud service system provided by the embodiment of the present application. Each node in the cloud service system provided by the embodiment of the present application will be introduced below. As shown in Figures 10 to 12 (Figure 10 is a schematic structural diagram of a load balancing node provided by an embodiment of the present application, Figure 11 is another schematic structural diagram of a client node provided by an embodiment of the present application, and Figure 12 is a schematic diagram of the implementation of the present application. Another structural diagram of the server node provided by the example), the following will introduce the nodes in the cloud service system in three situations.

In the first case: the sending module 1002 of the load balancing node is used to send the first identification to the client node; the processing module 1101 of the client node is used to send the destination address of the first service request based on the first identification, Convert the address of the load balancing node to the address of the server node to obtain the processed first service request; the sending module 1102 of the client node is used to send the processed first service request to the server node; the server node The processing module 1201 is used to process the processed first business request to obtain the first business response; the sending module 1202 of the server node is used to send the first business response to the client node; the processing of the client node Module 1101 is also configured to convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier to obtain the processed first service response.

In a possible implementation, the processing module 1001 of the load balancing node is specifically configured to convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, and then process it. The second service request after The source address of the business response is converted from the address of the service point node to the address of the load balancing node, and the first identifier is added to the second business response to obtain The processed second service response; the sending module 1002 of the load balancing node is specifically used to send the processed second service response to the client node. The second service response is that the server node performs the processing on the processed second service request. processed.

In a possible implementation, the processing module 1001 of the load balancing node is also used to detect whether the second service request contains the second identifier; the processing module 1001 of the load balancing node is specifically used to detect if the second service request contains the second identifier. identification, then convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, add the first identification to the second service response, and obtain the processed second service response; The sending module 1002 of the load balancing node is specifically configured to send the processed second service response to the client node.

In a possible implementation, the processing module 1001 of the load balancing node is also configured to, if the second service request does not contain the second identifier, transfer the source address of the second service response from the server node to the service point node. The address is converted into the address of the load balancing node to obtain the processed second service response; the sending module 1002 of the load balancing node is also used to send the processed second service response to the client node.

In the second case, the sending module 1002 of the load balancing node is used to send the first identification to the client node and the server node; the processing module 1101 of the client node is used to send the first service request based on the first identification. The destination address is converted from the address of the load balancing node to the address of the server node to obtain the processed first service request; the sending module 1102 of the client node is used to send the processed first service request to the server node ; The processing module 1201 of the server node is used to process the processed first business request, obtain the first business response, and convert the source address of the first business response from the address of the server node into a load based on the first identifier The address of the balancing node is used to obtain the processed first service response; the sending module 1202 of the server node is used to send the processed first service response to the client node.

In a possible implementation, the processing module 1001 of the load balancing node is specifically configured to convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, to the address of the server node. Add a first identifier to the second service request to obtain a processed second service request; the sending module 1002 of the load balancing node is specifically used to send the processed second service request to the server node; the processing module 1001 of the load balancing node, Specifically used to convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, add the first identifier to the second service response, and obtain the processed second service response; The sending module 1002 of the load balancing node is specifically configured to send the processed second service response to the client node. The second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the processing module 1001 of the load balancing node is also used to detect whether the second service request contains the second identifier; the processing module 1001 of the load balancing node is specifically used to detect if the second service request contains the second identifier. identification, then convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, add the first identification to the second service request, and obtain the processed second service request; The sending module 1002 of the load balancing node is specifically used to send the processed second service request to the server node; the processing module 1001 of the load balancing node is specifically used to send the source address of the second service response from the server node, Convert the address of the service point node to the address of the load balancing node, add the first identifier to the second service response, and obtain the processed second service response; the sending module 1002 of the load balancing node is specifically used to send the processed second service response. The business response is sent to the client node.

In a possible implementation, the processing module 1001 of the load balancing node is also configured to, if the second service request does not contain the second identifier, change the destination address of the second service request from the client node to the load balancing node. The address is converted into the address of the server node to obtain the processed second service request; the sending module 1002 of the load balancing node is also used to send the processed second service request to the server node; the processing module of the load balancing node 1001, also used to convert data from The source address of the second service response of the server node is converted from the address of the service point node to the address of the load balancing node to obtain the processed second service response; the sending module 1002 of the load balancing node is also used to convert the processed The second service response is sent to the client node.

In the third case: the sending module 1002 of the load balancing node is used to send the first identification to the client node and the server node; the processing module 1101 of the client node is used to send the first service request based on the first identification. The source address is converted from the address of the client node to the address of the load balancing node, and the destination address of the first business request is converted from the address of the load balancing node to the address of the server node to obtain the processed first business request; The sending module 1102 of the client node is used to send the processed first business request to the server node; the processing module 1201 of the server node is used to process the processed first business request and obtain the first business response. , based on the first identifier, convert the source address of the first service response from the address of the server node to the address of the load balancing node, and convert the corresponding destination address of the first service from the address of the load balancing node to the address of the client node , obtain the processed first service response; the sending module 1202 of the server node is used to send the processed first service response to the client node.

In a possible implementation, the processing module 1001 of the load balancing node is specifically configured to convert the source address of the second service request from the client node from the address of the client node to the address of the load balancing node, and convert the source address of the second service request from the client node to the address of the load balancing node. The destination address of the second service request is converted from the address of the load balancing node to the address of the server node, and the first identifier is added to the second service request to obtain the processed second service request; the sending module 1002 of the load balancing node is specifically used For sending the processed second service request to the server node; the processing module 1001 of the load balancing node is specifically used to convert the source address of the second service response from the server node from the address of the service point node to the load balancing The address of the node, convert the corresponding destination address of the second service from the address of the load balancing node to the address of the client node, add the first identifier to the second service response, and obtain the processed second service response; The sending module 1002 is specifically configured to send the processed second service response to the client node. The second service response is obtained by the server node processing the processed second service request.

In a possible implementation, the processing module 1001 of the load balancing node is also used to detect whether the second service request contains the second identifier; the processing module 1001 of the load balancing node is specifically used to detect if the second service request contains the second identifier. identification, the source address of the second service request from the client node is converted from the address of the client node to the address of the load balancing node, and the destination address of the second service request is converted from the address of the load balancing node to the server The address of the node, add the first identifier to the second service request, and obtain the processed second service request; the sending module 1002 of the load balancing node is specifically used to send the processed second service request to the server node; load balancing The processing module 1001 of the node is specifically used to convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, and convert the corresponding destination address of the second service from the address of the load balancing node. The address is converted into the address of the client node, the first identifier is added to the second service response, and the processed second service response is obtained; the sending module 1002 of the load balancing node is specifically used to send the processed second service response to client node.

In a possible implementation, the processing module 1001 of the load balancing node is also configured to, if the second service request does not contain the second identifier, transfer the source address of the second service request from the client node to The address of the load balancing node is converted into the address of the load balancing node, and the destination address of the second service request is converted from the address of the load balancing node to the address of the server node to obtain the processed second service request; the sending module 1002 of the load balancing node, It is also used to send the processed second service request to the server node; the processing module 1001 of the load balancing node is also used to convert the source address of the second service response from the server node from the address of the service point node to The address of the load balancing node will correspond to the second business The destination address is converted from the address of the load balancing node to the address of the client node to obtain the processed second service response; the sending module 1002 of the load balancing node is also used to send the processed second service response to the client node.

It should be noted that the information interaction, implementation process, etc. between the modules/units of the above-mentioned device are based on the same concept as the method embodiments of the present application, and the technical effects they bring are the same as those of the method embodiments of the present application. The specific content can be Refer to the description in the method embodiments shown above in the embodiments of the present application, which will not be described again here.

Figure 13 is another schematic structural diagram of a load balancing node provided by an embodiment of the present application. As shown in Figure 13, an embodiment of a load balancing node may include one or more central processors 1301, memory 1302, input and output interfaces 1303, wired or wireless network interfaces 1304, and power supplies 1305.

Memory 1302 may be ephemeral storage or persistent storage. Furthermore, the central processing unit 1301 may be configured to communicate with the memory 1302 and execute a series of instruction operations in the memory 1302 on the load balancing node.

In this embodiment, the central processor 1301 can execute the method steps executed by the load balancing node in the embodiment shown in FIG. 4, FIG. 6 or FIG. 8, which will not be described again here.

In this embodiment, the specific functional module division in the central processor 1301 may be similar to the division of the processing module and the sending module described in FIG. 10 , and will not be described again here.

Figure 14 is another schematic structural diagram of a client node provided by an embodiment of the present application. As shown in Figure 14, one embodiment of the client node may include one or more central processors 1401, memory 1402, input and output interfaces 1403, wired or wireless network interfaces 1404, and power supplies 1405.

Memory 1402 may be ephemeral storage or persistent storage. Furthermore, the central processing unit 1401 may be configured to communicate with the memory 1402 and execute a series of instruction operations in the memory 1402 on the client node.

In this embodiment, the central processor 1401 can execute the method steps executed by the client node in the embodiments shown in FIG. 4, FIG. 6 or FIG. 8, which will not be described again here.

In this embodiment, the specific functional module division in the central processor 1401 may be similar to the division of the processing module and the sending module described in FIG. 11 , and will not be described again here.

Figure 15 is another schematic structural diagram of a server node provided by an embodiment of the present application. As shown in Figure 15, an embodiment of the server may include one or more central processors 1501, memory 1502, input and output interfaces 1503, wired or wireless network interfaces 1504, and power supplies 1505.

Memory 1502 may be ephemeral storage or persistent storage. Furthermore, the central processing unit 1501 may be configured to communicate with the memory 1502 and execute a series of instruction operations in the memory 1502 on the server.

In this embodiment, the central processing unit 1501 can perform the method steps performed by the server in the embodiments shown in Figure 4, Figure 6 or Figure 8, which will not be described again here.

In this embodiment, the specific functional module division in the central processor 1501 may be similar to the division of the processing module and the sending module described in FIG. 10 , and will not be described again here.

Embodiments of the present application also relate to a computer storage medium. The computer-readable storage medium stores a program for signal processing. When the program is run on a computer, it causes the computer to execute as shown in Figure 4, Figure 6 or Figure 8. Steps in Examples.

Embodiments of the present application also relate to a computer program product that stores instructions that, when executed by a computer, cause the computer to perform the steps in the embodiments shown in FIG. 4, FIG. 6, or FIG. 8.

Those skilled in the art can clearly understand that for the convenience and simplicity of description, the system and device described above For the specific working process of the unit, please refer to the corresponding process in the foregoing method embodiment, which will not be described again here.

In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program code. .

Claims (20)

A cloud service system, characterized in that the system includes: The load balancing node is used to send the first identification to the client node; The client node is configured to convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, obtain the processed first service request, and convert The processed first service request is sent to the server node; The server node is configured to process the processed first service request, obtain a first service response, and send the first service response to the client node; The client node is further configured to convert the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier to obtain the processed first Business response. The system according to claim 1, characterized in that the load balancing node is specifically used for: Convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node to obtain the processed second service request, and convert the processed second service request Sent to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, add a first identifier to the second service response, and obtain the processed second service response, and sends the processed second service response to the client node, where the second service response is obtained by processing the processed second service request by the server node . The system according to claim 2, characterized in that the load balancing node is also used to detect whether the second service request contains a second identifier; The load balancing node is specifically configured to, if the second service request contains a second identifier, convert the source address of the second service response from the server node from the address of the service point node to the The address of the load balancing node, adding a first identifier to the second service response, obtaining a processed second service response, and sending the processed second service response to the client node. The system according to claim 3, characterized in that the load balancing node is further configured to, if the second service request does not contain a second identifier, send the source of the second service response from the server node to The address is converted from the address of the service point node to the address of the load balancing node, a processed second service response is obtained, and the processed second service response is sent to the client node. A cloud service system, characterized in that the system includes: The load balancing node is used to send the first identification to the client node and the server node; The client node is configured to convert the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, obtain the processed first service request, and convert The processed first service request is sent to the server node; The server node is configured to process the processed first service request to obtain a first service response, and obtain the source address of the first service response from the server node based on the first identifier. The address is converted into the address of the load balancing node, a processed first service response is obtained, and the processed first service response is sent to the client node. The system according to claim 5, characterized in that the load balancing node is specifically used for: Convert the destination address of the second service request from the client node from the address of the load balancing node to the address of the server node, add a first identifier to the second service request, and obtain the processed second service request , and send the processed second service request to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, add a first identifier to the second service response, and obtain the processed second service response, and sends the processed second service response to the client node, where the second service response is obtained by processing the processed second service request by the server node . The system according to claim 6, wherein the load balancing node is further configured to detect whether the second service request contains a second identifier; The load balancing node is specifically configured to: if the second service request contains a second identifier, convert the destination address of the second service request from the client node from the address of the load balancing node to the server node. address, add a first identifier to the second service request, obtain a processed second service request, and send the processed second service request to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, add a first identifier to the second service response, and obtain the processed respond to the second service, and send the processed second service response to the client node. The system according to claim 7, characterized in that the load balancing node is further configured to: if the second service request does not contain the second identifier, change the destination address of the second service request from the client node to , convert the address of the load balancing node to the address of the server node, obtain the processed second service request, and send the processed second service request to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node to obtain a processed second service response, and convert the processed The second service response is sent to the client node. A cloud service system, characterized in that the system includes: The load balancing node is used to send the first identification to the client node and the server node; The client node is configured to convert the source address of the first service request from the address of the client node to the address of the load balancing node based on the first identifier, and convert the destination of the first service request. Address, convert the address of the load balancing node to the address of the server node, obtain the processed first service request, and send the processed first service request to the server node; The server node is configured to process the processed first service request to obtain a first service response, and obtain the source address of the first service response from the server node based on the first identifier. Convert the address to the address of the load balancing node, convert the corresponding destination address of the first service from the address of the load balancing node to the address of the client node, and obtain the processed first service response, and send the processed first service response to the client node. The system according to claim 9, characterized in that the load balancing node is specifically used for: Convert the source address of the second service request from the client node from the address of the client node to the address of the load balancing node, and convert the destination address of the second service request from the address of the load balancing node. The address is converted into the address of the server node, a first identifier is added to the second service request, the processed second service request is obtained, and the The processed second service request is sent to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, and convert the corresponding destination address of the second service from the address of the load balancing node Convert the address of the node to the address of the client node, add a first identifier to the second service response, obtain a processed second service response, and send the processed second service response to the client The end node, the second service response is obtained by processing the processed second service request by the service end node. The system according to claim 10, characterized in that the load balancing node is also used to detect whether the second service request contains a second identifier; The load balancing node is specifically configured to: if the second service request contains a second identifier, convert the source address of the second service request from the client node from the address of the client node to the load The address of the balancing node, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, add a first identifier to the second service request, and obtain the processed second service request, and send the processed second service request to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, and convert the corresponding destination address of the second service from the address of the load balancing node Convert the address of the node to the address of the client node, add a first identifier to the second service response, obtain a processed second service response, and send the processed second service response to the client end node. The system according to claim 11, characterized in that the load balancing node is further configured to: if the second service request does not contain the second identifier, change the source address of the second service request from the client node to , convert the address of the client node to the address of the load balancing node, convert the destination address of the second service request from the address of the load balancing node to the address of the server node, and obtain the processed request a second service, and send the processed second service request to the server node; Convert the source address of the second service response from the server node from the address of the service point node to the address of the load balancing node, and convert the corresponding destination address of the second service from the address of the load balancing node The address of the node is converted into the address of the client node, a processed second service response is obtained, and the processed second service response is sent to the client node. A business processing method based on a cloud service system, characterized in that the system includes a load balancing node, a client node and a server node, and the method includes: The load balancing node sends the first identification to the client node; The client node converts the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, obtains the processed first service request, and processes the first service request. The subsequent first service request is sent to the server node; The server node processes the processed first service request, obtains a first service response, and sends the first service response to the client node; The client node converts the source address of the first service response from the address of the server node to the address of the load balancing node based on the first identifier to obtain a processed first service response. A business processing method based on a cloud service system, characterized in that the system includes a load balancing node, a client node and a server node, and the method includes: The load balancing node sends the first identification to the client node and the server node; The client node converts the destination address of the first service request from the address of the load balancing node to the address of the server node based on the first identifier, obtains the processed first service request, and processes the first service request. The subsequent first service request is sent to the server node; The server node processes the processed first service request to obtain a first service response, and converts the source address of the first service response from the address of the server node based on the first identifier. For the address of the load balancing node, obtain a processed first service response, and send the processed first service response to the client node. A business processing method based on a cloud service system, characterized in that the system includes a load balancing node, a client node and a server node, and the method includes: The load balancing node sends the first identification to the client node and the server node; The client node converts the source address of the first service request from the address of the client node to the address of the load balancing node based on the first identifier, and converts the destination address of the first service request from the address of the client node to the address of the load balancing node. Convert the address of the load balancing node to the address of the server node, obtain the processed first service request, and send the processed first service request to the server node; The server node processes the processed first service request to obtain a first service response, and converts the source address of the first service response from the address of the server node based on the first identifier. As the address of the load balancing node, convert the corresponding destination address of the first service from the address of the load balancing node to the address of the client node, obtain the processed first service response, and convert the corresponding destination address of the first service to the address of the client node. The processed first service response is sent to the client node. A load balancing node, characterized in that the load balancing node includes a memory and a processor; the memory stores code, the processor is configured to execute the code, and when the code is executed, the The load balancing node performs the steps implemented by the load balancing node in the method according to any one of claims 13 to 15. A client node, characterized in that the client node includes a memory and a processor; the memory stores code, the processor is configured to execute the code, and when the code is executed, the The client node performs the steps implemented by the client node in the method according to any one of claims 13 to 15. A server node, characterized in that the server node includes a memory and a processor; the memory stores code, the processor is configured to execute the code, and when the code is executed, the The server node performs the steps implemented by the server node in the method described in any one of claims 13 to 15. A computer storage medium, characterized in that the computer storage medium stores one or more instructions, which when executed by one or more computers cause the one or more computers to implement any of claims 13 to 15. The method described in 1. A computer program product, characterized in that the computer program product stores instructions, which when executed by a computer, cause the computer to implement the method described in any one of claims 13 to 15.