CN111464574B

CN111464574B - Calling, loading, registering and managing method and route, server, node and medium

Info

Publication number: CN111464574B
Application number: CN201910053292.8A
Authority: CN
Inventors: 聂松洋; 韩陆; 李星星; 周源
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2019-01-21
Filing date: 2019-01-21
Publication date: 2022-10-21
Anticipated expiration: 2039-01-21
Also published as: CN111464574A

Abstract

The present specification provides a method and a routing, server, node and medium for invoking, loading, registering, managing. The method may include: receiving a calling request, wherein the calling request is attached with an application identifier; determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of nodes; determining a target node in a node cluster corresponding to the target business service; and sending a calling request to the target node to call the application represented by the application identification. The capability of the application to provide the service can be improved.

Description

Calling, loading, registering and managing method and route, server, node and medium

Technical Field

The present disclosure relates to the field of computer internet, and in particular, to an application invoking and loading method, a node registration and management method, a routing device, a node, a server, and a computer storage medium.

Background

In the field of human-computer interaction, one dialogue with a user may need to invoke multiple business services in the background. Specifically, for example, the user inputs "recommend a nearby food" on the man-machine interface, which may involve an understanding service and a question and answer service in the background. The user's intention is recognized by the understanding service and answered to the user by the question-and-answer service, such as "there are 1 favorite hot pot restaurant 1 m away from you". One application may have a plurality of business services, for example, the above-described understanding service and question and answer service may be provided for one application.

In the prior art, one application is often operated in one server, and the load capacity of a single server is limited, so that the capability of the application for providing services is not strong.

Disclosure of Invention

The embodiment of the specification provides an application calling and loading method, a node registration and management method, a routing device, a node, a server and a computer storage medium. The service providing capability of the application can be improved.

An embodiment of the present specification provides an application calling method, including: receiving a calling request, wherein the calling request is attached with an application identifier; determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of nodes; determining a target node in a node cluster corresponding to the target business service; and sending a calling request to the target node to call the application represented by the application identification.

An embodiment of the present specification provides a routing device, including: the receiving module is used for receiving a calling request, and the calling request is attached with an application identifier; the service determining module is used for determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of nodes; the node determining module is used for determining a target node in the node cluster corresponding to the target business service; and the request sending module is used for sending a calling request to the target node so as to call the application represented by the application identifier.

The present specification embodiments provide a computer storage medium having computer program instructions stored thereon that, when executed, implement: receiving a calling request, wherein the calling request is attached with an application identifier; determining a target business service in a plurality of business services according to the calling request, wherein the business service corresponds to a node cluster, and the node cluster comprises a plurality of nodes; determining a target node in a node cluster corresponding to the target business service; and sending a calling request to the target node to call the application represented by the application identification.

An embodiment of the present specification provides an application calling method, including: polling the working state of nodes in the node cluster; the node cluster corresponds to a business service; under the condition of receiving a node determination request, determining a fed-back node identifier according to the working state of nodes in the node cluster; the node represented by the determined node identification is in a normal state, and an application for realizing the business service is operated; and feeding back the node identification to a routing device sending the node determination request, so that the routing device sends a call request to the node represented by the node identification to call the corresponding application.

An embodiment of the present specification provides a server, including: the polling module is used for polling the working state of the nodes in the node cluster; the node cluster corresponds to a business service; the node determining module is used for determining a fed-back node identifier according to the working state of the nodes in the node cluster under the condition of receiving a node determining request; the node represented by the determined node identification is in a normal state, and an application for realizing the business service is operated; and the node identifier sending module is used for feeding back the node identifier to a routing device which sends the node determination request, so that the routing device sends a call request to the node represented by the node identifier to call the corresponding application.

The present specification embodiments provide a computer storage medium storing computer program instructions that, when executed, implement: polling the working state of nodes in a node cluster, wherein the node cluster corresponds to a service; under the condition of receiving a node determination request, determining a fed-back node identifier according to the working state of the nodes in the node cluster, wherein the node represented by the determined node identifier is in a normal state and runs an application for realizing the business service; and feeding back the node identification to a routing device sending the node determination request, so that the routing device sends a call request to the node represented by the node identification to call the corresponding application.

An embodiment of the present specification provides a node registration method, which is applied to a node, and includes: determining a node cluster corresponding to the node; determining a business service corresponding to the node cluster; wherein, the business service corresponds to an application list; wherein, the application list comprises at least one application identifier of the application providing the business service; loading the application represented by the application identifier in the application list corresponding to the business service; and sending a registration request to a management server to inform the management server that the node is in a normal state.

An embodiment of the present specification provides a node, including: the cluster determining module is used for determining a node cluster corresponding to the node; the service determining module is used for determining the service corresponding to the node cluster; wherein, the business service corresponds to an application list; wherein, the application list comprises at least one application identifier of the application providing the business service; the application loading module is used for loading the application represented by the application identifier in the application list corresponding to the business service; and the registration request sending module is used for sending a registration request to the management server so as to inform the management server that the node is in a normal state.

The present specification embodiments provide a computer storage medium having computer program instructions stored thereon that, when executed, implement: determining a node cluster corresponding to the node; determining a business service corresponding to the node cluster, wherein the business service corresponds to an application list, and the application list comprises at least one application identifier of an application providing the business service; loading the application represented by the application identifier in the application list corresponding to the business service; and sending a registration request to a management server to inform the management server that the node is in a normal state.

The embodiment of the specification provides a node management method, which is applied to a management server, wherein the management server is provided with a node list; wherein, the node list comprises at least one node identifier for representing a node; the method comprises the following steps: polling the working state of nodes in the node cluster; the node cluster corresponds to a business service; under the condition that the node is determined to be in an abnormal state, acquiring a node identifier of the node in the abnormal state; and setting a mark corresponding to the node identifier of the node in the abnormal state in the node list, or removing the node identifier of the node in the abnormal state from the node list, so that the management server does not provide the node identifier of the node in the abnormal state to a routing device sending the node determination request under the condition that the management server receives the node determination request.

The embodiment of the specification provides a server, wherein the server is provided with a node list; wherein, the node list comprises at least one node identifier for representing a node; the server includes: the state polling module is used for polling the working states of the nodes in the node cluster; the node cluster corresponds to a business service; the identification acquisition module is used for acquiring the node identification of the node in the abnormal state under the condition that the node is determined to be in the abnormal state; and the list management module is used for setting a mark corresponding to the node identifier of the node in the abnormal state in the node list, or removing the node identifier of the node in the abnormal state from the node list, so that the management server does not provide the node identifier of the node in the abnormal state to a routing device sending the node determination request under the condition of receiving the node determination request.

The embodiment of the specification provides a computer storage medium, wherein the computer storage medium stores computer program instructions and a node list; wherein, the node list comprises at least one node identifier for representing a node; the computer program instructions when executed implement: polling the working state of nodes in the node cluster; the node cluster corresponds to a business service; under the condition that the node is determined to be in an abnormal state, acquiring a node identifier of the node in the abnormal state; and setting a mark corresponding to the node identifier of the node in the abnormal state in the node list, or removing the node identifier of the node in the abnormal state from the node list, so that the management server does not provide the node identifier of the node in the abnormal state to a routing device sending the node determination request under the condition that the management server receives the node determination request.

An embodiment of the present specification provides an application loading method, including: determining a target node cluster; wherein the target node cluster comprises a plurality of nodes; sending an application loading request to nodes in the target node cluster; wherein, the application loading request is attached with an application identifier; receiving loading state information fed back by the nodes in the target node cluster; under the condition that the loading state information indicates that the nodes in the target node cluster finish loading the application represented by the application identifier, sending an application offline request to an original node cluster running the application represented by the application identifier so as to stop running the application by the nodes in the original node cluster; wherein the target node cluster is different from the original node cluster.

An embodiment of the present specification provides a server, including: the cluster determining module is used for determining a target node cluster; wherein the target node cluster comprises a plurality of nodes; a loading request sending module, configured to send an application loading request to a node in the target node cluster; wherein, the application loading request is attached with an application identifier; the state information receiving module is used for receiving loading state information fed back by the nodes in the target node cluster; a offline request sending module, configured to send an application offline request to an original node cluster running the application indicated by the application identifier, so that a node in the original node cluster stops running the application, when the loading state information indicates that the node in the target node cluster finishes loading the application indicated by the application identifier; wherein the target node cluster is different from the original node cluster.

The present specification embodiments provide a computer storage medium storing computer program instructions that, when executed, implement: determining a target node cluster, wherein the target node cluster comprises a plurality of nodes; sending an application loading request to a node in the target node cluster, wherein the application loading request is attached with an application identifier; receiving loading state information fed back by nodes in the target node cluster; and under the condition that the loading state information indicates that the nodes in the target node cluster finish loading the application represented by the application identifier, sending an application offline request to an original node cluster running the application represented by the application identifier so as to stop running the application by the nodes in the original node cluster, wherein the target node cluster is different from the original node cluster.

In the embodiment of the present specification, the application of the service is executed by using the node cluster by aiming at the node cluster corresponding to the service. The nodes in the node cluster all run the same application. So that the business services of an application can be allocated to different node responses when invoked. Therefore, calling of the same service of the application is realized, and different nodes can respond, so that the capability of the application for providing the service is improved.

Drawings

The accompanying drawings, which are included to provide a further understanding of embodiments of the specification, are incorporated in and constitute a part of this specification, illustrate embodiments of the specification and together with the description serve to explain the principles of the specification. It is obvious that the drawings in the following description are only some embodiments of the present description, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:

FIG. 1 is a schematic view of a scene interaction provided in an embodiment of the present disclosure;

fig. 2 is a schematic diagram of a display interface of a client provided in an embodiment of the present specification;

fig. 3 is a flowchart of an application calling method provided in an embodiment of the present specification;

fig. 4 is a block diagram of a routing apparatus provided in an embodiment of the present specification;

fig. 5 is a flowchart of an application calling method provided in an embodiment of the present specification;

FIG. 6 is a block diagram of a server provided in an embodiment of the present disclosure;

fig. 7 is a flowchart of a node registration method according to an embodiment of the present disclosure;

fig. 8 is a block diagram of a node according to an embodiment of the present disclosure;

fig. 9 is a node management method according to an embodiment of the present disclosure;

FIG. 10 is a block diagram of a server provided in an embodiment of the present disclosure;

fig. 11 is a flowchart of an application loading method provided in an embodiment of the present specification;

fig. 12 is a block diagram of a server provided in an embodiment of the present specification.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present specification, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without making any creative effort shall fall within the protection scope of the present specification.

Please refer to fig. 1. In one scenario example, an application management system is provided. The application management system comprises a routing device, a management server and a node cluster.

In this scenario example, the routing device provides a routing service for the received invocation request. The routing device may send the invocation request to a node in a normal state in the node cluster. The management server may determine the operating status of the nodes in the node cluster to notify the routing device. The node cluster comprises at least one node, and the node runs an application.

In this scenario example, a user may run a client of an application through a cell phone. Specifically, for example, a user may browse goods through a client operated by a mobile phone and make a consultation for some problems.

In this scenario example, a user may use the intelligent robot service of a client. For example, using a client-provided smart assistant. The information entered by the user may be: "find a discount on the refrigerator". The client sends a call request to the routing device, where the call request may be accompanied by an application identifier of the client, for example, "shoujijingdong.

In this scenario example, after receiving the invocation request, the routing apparatus discovers that the invocation request is accompanied by user input data. That is, the call request is appended with the text data input by the user "find out the offer about the refrigerator". The routing device may determine that the target service corresponding to the invocation request may include an understanding service, so as to obtain the intention of the user through the understanding service, and further determine the target service of the next step. Specifically, the service identification to understand the service may be "lijiefuwu".

In this scenario example, the routing apparatus has a configuration module in which node list information is provided. And recording the corresponding cluster identifier and the node identifier in the service identifier corresponding to the service in the node list information. The service identifier may correspond to at least one cluster identifier, and the cluster identifier may correspond to at least one node identifier. The cluster identifier is used for representing a node cluster, and the node identifier corresponding to the cluster identifier is used for representing the node in the node cluster.

In this scenario example, the routing apparatus may determine that the corresponding cluster identifier is "jiqun2567" according to the service identifier "lijiefuwu", and further, the routing apparatus may select one of the node identifiers corresponding to the cluster identifier, and take the node represented by the selected node identifier as the target node. Specifically, the node identifier included in the node list information indicates that all the nodes are in a normal state. The routing device deletes the node identifier of the node in the abnormal state in the node list information. Specifically, the management server may poll the node to obtain the working state of the node, and may notify the routing apparatus when the working state of the node is found to be in an abnormal state, so that the routing apparatus may delete the node identifier of the node in the abnormal state in the node list information. Of course, when the working state of the node is restored to the normal state, the routing device may also be notified to add the node identifier to the node list information.

In the example of the scenario, the user uses an application through a mobile phone, which is provided with the target business service. Specifically, for example, the client provides an understanding service. Program instructions of the part of the client providing the understanding service can be run in each node in the node cluster corresponding to the target business service. Therefore, each node in the node cluster corresponding to the target service in the normal state can be called by the corresponding routing device.

In this scenario example, the routing apparatus may send the invocation request to the target node, so that the application in the target node invokes the corresponding application to respond according to the application identifier of the invocation request. For example, after the understanding service of the client gets the call request, the analysis and feedback result to the routing device may be performed for "checking the preference about the refrigerator". Specifically, for example, the result of understanding the service feedback is "refrigerator & offer". At this time, after the routing device receives the feedback, the recommendation service may be further invoked. And determining a target node in the node cluster corresponding to the recommendation service, and sending the application identifier and the refrigerator and the preference to the target node corresponding to the recommendation service. And the target node provides the execution result. For example, please refer to fig. 2. And recommending the refrigerator information which is discounted to the user.

Please refer to fig. 3. The embodiment of the specification provides an application calling method. The application calling method may be applied to a routing device. The routing means may be the primary device of communication between the networks. The routing means may comprise hardware units such as a network communication unit, a processor and a memory. The memory may have a configuration module therein. The application calling method may include the following steps.

Step S10: receiving a call request, wherein the call request is attached with an application identifier.

In this embodiment, a user may use a client to send a call request to call an application to implement a corresponding service function. Specifically, for example, the user uses the smart assistant function of the client. The user inputs ' looking up the logistics progress of the down jacket bought by me ' on the interactive interface '. At this time, the client of the user sends a call request to the routing device, and the call request may be accompanied by an identifier of the client, for example 1264.

In this embodiment, the application identifier may be used to uniquely identify an application. So that the corresponding application can be determined by the application identification.

Step S12: determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of service nodes.

In this embodiment, one application can implement a plurality of business services. Program instructions corresponding to different business services can be run in different node clusters. It can be realized that an application list may be set for the corresponding service, and the node in the node cluster corresponding to the service may run the application in the corresponding application list. Further, the node runs program instructions for applying the corresponding business service. Specifically, for example, the first application may provide an understanding service and a response service for a user to send voice; the second application provides an understanding service and a voice recognition service, and the third application provides an understanding service, an answering service and a voice recognition service. At this time, the program instructions of the understanding service of the three applications may be executed in the node cluster corresponding to the understanding service, the program instructions of the answering service of the first application and the third application may be executed in the node cluster corresponding to the answering service, and the program instructions of the voice recognition service of the second application and the third application may be executed in the node cluster corresponding to the answering service.

In this embodiment, the business service may have a correspondence with the node cluster. Specifically, the routing device may be configured with a configuration file, and the configuration file records a correspondence between the service and the node cluster. Further, the configuration file may also record nodes included in each node cluster. Specifically, the node cluster may be represented by a cluster identifier, and the node may be represented by a node identifier. Specifically, for example, table 1 below.

TABLE 1

In this embodiment, a target business service is determined among a plurality of business services according to the invocation request. The invocation request received by the routing device may be accompanied by a service identifier capable of representing a service. Therefore, the corresponding business service can be determined as the target business service according to the business identifier. Or, the website recently visited by the client sending the call request may be read, and the business service provided by the current interface of the client may be determined through the website. Alternatively, the dialogue data attached to the call request is analyzed, and the business service required for responding to the call request is further specified, for example, in the case that the call request is found to be attached with text data, the business service required for responding to the call request is specified to include an understanding service and a response service. The user is answered by a response service by understanding the user's intent as expressed by the service analysis of the text data. It is to be understood that the present disclosure is not limited to the above examples, and other modifications may be made by those skilled in the art within the spirit of the present disclosure.

In this embodiment, the applications run by the nodes in the same node cluster may be the same. Therefore, the node cluster corresponds to the service, and program instructions for realizing the same service of different applications can be located in each node of the same node cluster. Program instructions run by the node may indicate, via the application identification, the application to which it belongs.

Step S14: determining a target node in a node cluster corresponding to the target business service; wherein the target node is in a normal state.

In the present embodiment, a node may have a normal state and an abnormal state. The normal state may be a state in which the node executes the program instruction normally to implement a corresponding function. The abnormal state may be a state in which the node cannot normally execute the program quality to implement a corresponding function. Specifically, for example, during the operation of the node, some operation errors or hardware failures may occur, so that the node is in an abnormal state.

In this embodiment, in order to ensure the implementation of the service function, when the routing device specifies a target node, the routing device may first determine the operating state of the node in the node cluster corresponding to the target service, and then determine the node in the normal state as the target node. Specifically, for example, the routing device may interact with the management server, and the management server provides the node identifier of the node in the normal state in the node cluster corresponding to the target service. So that the routing device can determine the target node according to the node identification. Or, the routing device may also maintain a configuration file, where the configuration file records a node currently in a normal state, and thus, a node cluster corresponding to the target service is obtained through matching in the configuration file, and the node in the normal state is further designated as the target node in the node cluster.

Step S16: and sending a calling request to the target node to call the application represented by the application identification.

In this embodiment, after the target node is determined, an application run by the target node may be invoked. Specifically, the application represented by the application identifier attached to the call request may be obtained by matching the application identifier attached to the call request with the application identifiers of the multiple applications run by the target node.

In this embodiment, the application indicated by the application identifier is called, and the call request may be responded to by using the application. Specifically, for example, in response to a user entering "check the logistics progress of my purchased down jacket" on the interactive interface, the result of the application response may be a reply "you buy down jacket already in dispatch, please note keep the phone clear".

In this embodiment, by forming a node cluster from nodes and corresponding the node cluster to a service, when the routing device receives a call request, the service is determined first, and then a target node in a normal state is determined in the node cluster corresponding to the service. Therefore, the system has better stability in operation. Moreover, the application is operated in a node cluster mode, so that the data processing method has strong data processing capacity. Third, in this embodiment, program instructions of different applications are divided according to service services, and the program instructions for implementing the same service are set in a node cluster, where each node in the node cluster may have a service capability of responding to a call request. Therefore, a plurality of call requests aiming at the same application and the same service can be processed by different nodes in the same node cluster, and the implementation mode can have very strong data processing capacity.

In one embodiment, the call request is further accompanied by a service identifier; determining a target business service among a plurality of business services according to the invocation request may include: determining a target business service in the plurality of business services according to the business identifier; wherein the service identifier corresponds to the target service.

In this embodiment, the service identifier is used to indicate a service. The service represented by different service identities may be different. Therefore, the service identifier is attached to the call request, so that the service to be called can be specified by the client initiating the call request, and the service required by the user is more definite.

In this embodiment, the service identifier attached to the call request may be matched with a plurality of stored service identifiers. Specifically, for example, a service indicated by a service identifier that is the same as the service identifier of the invocation request in the stored service identifiers may be used as the target service.

In one embodiment, determining a target business service among a plurality of business services according to the invocation request may include: determining a client identification of a client sending the call request; inquiring the access record of the client identification; wherein, the access record records the access data of the client; and determining the target business service corresponding to the calling request according to the access record.

In this embodiment, the user can continuously perform network access by operating the client. Specifically, the user may use a browser to browse a web page, may use a shopping application to make a shopping, may use an intelligent dialog function of the application, may have a dialog with a robot, and the like. The routing device may record an access record for a network access of a user. The access record may have access data of the client recorded therein corresponding to the client identifier. The client identification may be used to uniquely represent a client. Therefore, the current displayed page or the running function of the client can be determined by analyzing the latest access data of the client represented by the client identifier, and the target business service corresponding to the call request is further determined. Specifically, for example, the access record may correspond to the client identifier, where the website accessed by the client is recorded. Therefore, the corresponding page can be determined according to the website corresponding to the client identifier, and the target business service can be determined according to the function provided by the page. For example, the function of the page is a function of translating Chinese into English, and the target business service can be a function of translating Chinese into English.

In this embodiment, the client identifier may be a physical address of the client, or a user account name of the user.

In one embodiment, the invocation request is accompanied by user-entered dialog data; determining a target business service among a plurality of business services according to the invocation request may include: and determining the target business service responding to the calling request according to the dialogue data.

In this embodiment, the routing device may determine, according to data accompanying the invocation request, a plurality of target business services that respond to the invocation request. And may be invoked in a preset order. Specifically, for example, the dialog data input by the user may include "tomorrow weather conditions", and after the routing device receives the call request and obtains the dialog data, the routing device may call the understanding service, that is, send the dialog data to the nodes of the node cluster for understanding the service. And receiving information fed back by the understanding service, further calling a weather service to obtain weather data, and finally calling a question-answering service, packaging the weather data into a communication message mode, and feeding back the communication message mode to a client of the user.

In one embodiment, determining a target node in a node cluster corresponding to the target business service may include: sending a node determination request to a management server, wherein the node determination request is attached with a cluster identifier, and the cluster identifier is used for representing a node cluster corresponding to a target business service; and receiving the node identification of the node determined by the management server in the node cluster represented by the cluster identification, wherein the node represented by the node identification is used as the target node.

In this embodiment, the management server may maintain a node list of nodes. The cluster identifier, the node identifier, the record and the corresponding working state can be correspondingly recorded in the node list. As such, the operating state may include a normal state and an abnormal state. Of course, the management server may also only maintain a node list in a normal state, that is, all nodes in the list are in a normal state, and nodes not in the node list are in an abnormal state.

In this embodiment, the routing device sends a node determination request to the management server, so that the management server can determine a corresponding node cluster according to the cluster identifier. And one of the nodes in the normal state in the node cluster can be selected through the node list and fed back to the corresponding node identifier of the routing device. So that the routing device can access the node according to the node identification.

In one embodiment, determining a target node in a node cluster corresponding to the target business service may include: reading node list information corresponding to the target service in a configuration module, wherein the node list information comprises at least one node identifier, and the node identifier is used for representing a node; and determining a target node in the nodes represented by the node list information.

In this embodiment, the routing device may maintain a configuration module. The node list information in the configuration module may be recorded with a service identifier, a cluster identifier, and a node identifier. In this way, when the routing device determines the target node, after determining the target service, one node identifier may be selected from the node identifiers corresponding to the corresponding cluster identifiers, and the node represented by the node identifier may be used as the target node.

Further, the node list information may also record a working state corresponding to the node identifier. The routing device may communicate with the management server at intervals, and the management server may provide the operating status of the node. The routing device can update the working state stored in the corresponding node identifier in the configuration module according to the working state provided by the management server. In this way, when the target node is determined, a node in a normal state may be selected as the target node from the data stored in the configuration module in the node cluster corresponding to the target service. Of course, the node list information may include only the node identifiers of the nodes in the normal state. In this manner, the node identification of the node in the abnormal state may be removed from the node list. And when the nodes in the normal state are newly added, adding the node identification of the newly added nodes into the node list information.

In one embodiment, determining a target business service among a plurality of business services according to the invocation request may include: determining a first target business service and a second target business service in a plurality of business services according to the calling request; correspondingly, determining a target node in the node cluster corresponding to the target service includes: determining a first target node in a node cluster corresponding to the first target business service; determining a second target node in the node cluster corresponding to the second target business service; correspondingly, sending a call request to the target node includes: sending a first calling request to the first target node to call the application represented by the application identification; receiving first result data fed back by the first target node; and sending a second calling request to the second target node, wherein the second calling request is accompanied by the first result data, so that the application represented by the application identifier in the second target node generates feedback data based on the first result data.

In this embodiment, the call request may correspond to a plurality of business services. The routing device may determine the first target business service and the second target business service according to the invocation request. Specifically, for example, the invocation request may be accompanied by a plurality of service identifiers. In this way, the routing device can determine a plurality of service services corresponding to the call request according to a plurality of service identifiers attached to the call request. Alternatively, in the case where the invocation request is accompanied by specified data, the routing means may invoke a plurality of business services according to a preset invocation rule. Specifically, for example, the invoking request is accompanied by text data input by the user, and the invoking rule of the routing device includes first requesting to invoke the understanding service, and then determining the next business service according to a result obtained by the understanding service. For example, the result of understanding the service indicates that the user needs to know the weather forecast of the next day, at this time, the routing device further invokes the weather service, and feeds back the weather condition data of the next day fed back by the weather service to the client.

In this embodiment, the routing device may determine the first destination node and the second destination node in the manner described in the other embodiments. Further, the routing device may first invoke an application running in the first target node, and may take the first result data output by the first target node as an input to an application of the second target node. Therefore, the appointed service function can be realized through the interaction of data among the applications in the called nodes. Specifically, for example, the first target business service may be a recommendation service, and the second target business service may be a presentation service. And the display service packages the commodity information fed back by the recommendation service into data which can be displayed on the current interface of the user.

Please refer to fig. 4. The embodiment of the specification provides a routing device. The routing device may include: the receiving module is used for receiving a calling request, and the calling request is attached with an application identifier; the service determining module is used for determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of nodes; the node determining module is used for determining a target node in the node cluster corresponding to the target business service; wherein the target node is in a normal state; and the request sending module is used for sending a calling request to the target node so as to call the application represented by the application identifier.

In this embodiment, the functions and effects achieved by the routing device may be explained in comparison with other embodiments, and are not described in detail.

The embodiment of the specification provides a computer storage medium. The computer storage medium stores computer program instructions that, when executed, implement: receiving a calling request, wherein the calling request is attached with an application identifier; determining a target business service in a plurality of business services according to the calling request, wherein the business service corresponds to a node cluster, and the node cluster comprises a plurality of nodes; determining a target node in a node cluster corresponding to the target business service, wherein the target node is in a normal state; and sending a calling request to the target node to call the application represented by the application identification.

In this embodiment, the computer storage medium includes, but is not limited to, a Random Access Memory (RAM), a Read-Only Memory (ROM), a Cache (Cache), a Hard Disk Drive (HDD), or a Memory Card (Memory Card).

In this embodiment, specific functions implemented by the computer program instructions in the computer storage medium can be explained with reference to other embodiments.

Please refer to fig. 5. The embodiment of the specification provides an application calling method. The application calling method may be applied to a management server. The management server may be an electronic device with a strong computing power. The management server may include a network communication unit, a processor, a memory, and a storage. Of course, the management server may also be software running in the electronic device. The application calling method may include the following steps.

Step S18: polling the working state of nodes in the node cluster; the node cluster corresponds to a business service.

In this embodiment, the management server may send a status query instruction to each node at a specified time interval, so that each node feeds back the operating status to the management server after receiving the status query instruction. Specifically, the operating state may be represented by a specified flag. Specifically, for example, the Normal state may be "Normal", and the Abnormal state may be "Abnormal".

In this embodiment, after the management server sends a status query instruction to a node, if no feedback of the corresponding node is received within a specified time period, the node may be considered to be in an abnormal status. Or the management server receives state information which is fed back by the node and represents an abnormal state. So that the management server can know the operating state of the node.

Step S20: under the condition of receiving a node determination request, determining a fed-back node identifier according to the working state of nodes in the node cluster; and the node represented by the determined node identifier is in a normal state, and an application for realizing the business service is operated.

In this embodiment, the management server may manage the operating state of at least one node cluster. In this embodiment, the management server may maintain a node list of nodes. The cluster identifier, the node identifier and the corresponding working state can be correspondingly recorded in the node list. Of course, the management server may also maintain only one node list in a normal state, that is, all the nodes in the list are in a normal state, and the nodes not in the node list are in an abnormal state. After receiving the node determination request, the management server may randomly send the node identifier of the node in the node cluster to the routing device.

In this embodiment, in some cases, the node determination request transmitted by the routing apparatus may be accompanied by a cluster identifier. As such, the request may be determined by the node to specify a cluster of nodes. In this way, the management server may determine a node in the node cluster represented by the cluster identifier and feed back the node identifier.

Step S22: and feeding back the node identification to a routing device sending the node determination request, so that the routing device sends a call request to the node represented by the node identification to call the corresponding application.

In this embodiment, the management server may determine the corresponding node cluster according to the cluster identifier. And one of the nodes in the normal state in the node cluster can be selected through the node list and fed back to the corresponding node identifier of the routing device. So that the routing device can access the node according to the node identification.

In one embodiment, the number of the node clusters is multiple, and the node determines that the request is accompanied by a cluster identifier; the cluster identifier is used for representing a node cluster; correspondingly, in the case of receiving a node determination request, determining a fed-back node identifier according to the working state of the nodes in the node cluster, including: and determining the fed back node identification according to the working state of the node in the node cluster represented by the cluster identification.

In this embodiment, each node cluster may correspond to one service. A business service may correspond to at least one cluster of nodes. Therefore, the method can provide stronger data processing capacity for each business service.

In this embodiment, the node identification is attached to the node specifying request, so that the routing apparatus specifies the node cluster where the node to be called is located. The management server selects a node among the node cluster indicated by the cluster identifier specified by the routing device.

In this embodiment, the management server determines the node identifier according to the operating state of the node. So that the management server can feed back the node identification of the node in the normal state to the routing device. In this way, the routing device sends the call request of the client to the node represented by the node identification. The implementation can ensure that the calling request of the client can be effectively responded. The stability of system operation has been promoted.

In an embodiment, in a case that a node determination request is received, determining a fed-back node identifier according to an operating state of a node in the node cluster may include: querying a plurality of nodes in a normal state in the node cluster; determining a workload of the plurality of nodes; and taking the node identifier of the node with smaller workload in the plurality of nodes as the fed back node identifier.

In this embodiment, the management server may further select the node identifier of the node with a smaller workload on the basis of determining the node identifier according to the operating state. Therefore, the node represented by the node identifier fed back to the routing device can relatively quickly respond to the calling request of the routing device, and the response speed for user access is improved.

In this embodiment, multiple nodes in a node cluster may run the same application. So that each node in the normal state can respond to a request to invoke a running application. In this way, the number of requests being responded to by different nodes is made different, and the computation amount of the function executed by different applications may also be different. So that the workload of different nodes may not be the same. The management server may query the workload of each node in a polling manner, or calculate the workload of the node by counting the time difference between the time when each node receives the call request and the time when each node feeds back the result, or the node may feed back the workload to the management server at a certain interval. So that the management server can determine the node identifier fed back to the routing device according to the workload of the node.

Please refer to fig. 6. The embodiment of the specification provides a server. The server may include the following modules.

The polling module is used for polling the working state of the nodes in the node cluster; the node cluster corresponds to a business service.

The node determining module is used for determining a fed-back node identifier according to the working state of the nodes in the node cluster under the condition of receiving a node determining request; and the node represented by the determined node identification is in a normal state, and an application for realizing the business service is operated.

And the node identifier sending module is used for feeding back the node identifier to a routing device which sends the node determination request, so that the routing device sends a call request to the node represented by the node identifier to call the corresponding application.

The functions and effects achieved by the modules included in the server provided by this embodiment may be explained in contrast to other embodiments, and are not described again.

The embodiment of the specification provides a computer storage medium. The computer storage medium stores computer program instructions that, when executed, implement: polling the working state of nodes in a node cluster, wherein the node cluster corresponds to a service; under the condition of receiving a node determination request, determining a fed-back node identifier according to the working state of the nodes in the node cluster, wherein the node represented by the determined node identifier is in a normal state and runs an application for realizing the business service; and feeding back the node identification to a routing device sending the node determination request, so that the routing device sends a call request to the node represented by the node identification to call the corresponding application.

Please refer to fig. 7. The implementation mode of the specification also provides a node registration method. The node registration method may be applied to a node. The node may be an electronic device with a strong computing power. In particular, a node may include a processor, a network communication unit, and a memory. Of course, the node may also include a memory storage, i.e. the data of the storage will not disappear after power is cut off. Of course, the node may also be a virtual server implemented by cloud technology. The node registration method may include the following steps.

Step S24: and determining a node cluster corresponding to the node.

In this embodiment, after the node is restarted or when a node is newly added, the node cluster corresponding to the node itself needs to be determined. Specifically, the node may actively send a cluster determination request to a designated server to request the designated server to feed back a node cluster corresponding to the node; or, the node receives a cluster identifier input by a background management end, and takes a node cluster represented by the cluster identifier as a node cluster corresponding to the node; alternatively, the designated server may store a node list, and the node list may record a node cluster corresponding to each node. The designated server may poll for the node to learn the operational status of the node. Furthermore, the designated server may, after discovering the node for newly establishing communication, query the node list for the cluster identifier corresponding to the node identifier of the node. The cluster identification may be sent to the node to inform the node of the corresponding cluster of nodes.

Step S26: acquiring a business service corresponding to the node cluster; the business service corresponds to an application list; wherein the application list includes an application identifier of at least one application providing the business service.

In this embodiment, the node may further determine, according to the obtained cluster identifier, a service identifier corresponding to the cluster identifier. Therefore, the node is made to know the corresponding service.

In this embodiment, the node may send a service determination request to the specified server, and the service determination request may be accompanied by the cluster identifier. The designated server may record a service identifier corresponding to the cluster identifier. The service is represented by a service identity. In this way, the designated server sends the service identifier corresponding to the cluster identifier to the node, so that the node can load the corresponding application according to the service identifier. Of course, the information list of the cluster identifier and the service identifier may be stored locally in the node. Therefore, after acquiring the cluster identifier corresponding to the node, the node can determine the corresponding service through table lookup operation. Alternatively, the node may send a service determination request to the routing apparatus, and the routing apparatus may determine the service identifier corresponding to the node according to the stored configuration file.

In an embodiment, the application list corresponding to the business service may provide the business service for the applications represented by the application identifiers in the application list. Specifically, for example, the service identifier "tianqi" indicates a weather service, and the application identifiers included in the corresponding application list include "moji" and "tenxunqq". Wherein, the application represented by 'moji' can be ink weather, and the application represented by 'tengxun QQ' can be Tencent QQ, and both can provide weather service. In the node cluster corresponding to the weather service, part of the program instructions for providing the weather service by the application can be run.

Step S28: and loading the application represented by the application identifier in the application list corresponding to the business service.

In this embodiment, the node may load the application represented by the application identifier. So that the nodes can respond to the corresponding call requests to provide the specified business services. The node may send an identification of the application that needs to be loaded to the specified server. In this way, after receiving the program instruction of the application fed back by the designated server, the application is loaded.

Step S30: and sending a registration request to a management server to inform the management server that the node is in a normal state.

In this embodiment, after the node loads the application indicated by the application identifier, it may be considered that the node has completed preparation for responding to the call request, so that the node may be put into normal operation. The node may send a registration request to the management server. The registration request may be accompanied by a node identification of the node. Thus, the management server is notified of the node identification that needs to be registered. The registration request may also be accompanied by a status identification. The state identifier is used for representing the working state of the node. Specifically, a state flag indicating that the node is in a normal state may be attached. Of course, the management server may also consider the corresponding node to be in a normal state after receiving the registration request, without attaching the state identifier to the registration request.

Please refer to fig. 8. The embodiment of the specification provides a node. The node may comprise: the cluster determining module is used for determining a node cluster corresponding to the node; the service determining module is used for determining the service corresponding to the node cluster; wherein, the business service corresponds to an application list; wherein, the application list comprises at least one application identifier of the application providing the business service; the application loading module is used for loading the application represented by the application identifier in the application list corresponding to the business service; and the registration request sending module is used for sending a registration request to the management server so as to inform the management server that the node is in a normal state.

The functions and effects of the node implementation provided in this embodiment may be explained in comparison with other embodiments, and are not described in detail.

The embodiment of the specification provides a computer storage medium. The computer storage medium stores computer program instructions that, when executed, implement: determining a node cluster corresponding to the node; determining a business service corresponding to the node cluster, wherein the business service corresponds to an application list, and the application list comprises at least one application identifier of an application providing the business service; loading the application represented by the application identifier in the application list corresponding to the business service; and sending a registration request to a management server to inform the management server that the node is in a normal state.

Please refer to fig. 9. The embodiment of the specification also provides a node management method. The node management method can be applied to a management server. The management server has a node list in a normal state; wherein, the node list comprises at least one node identifier for representing a node; the method of managing the node may include the following steps.

Step S32: polling the working state of nodes in the node cluster; the node cluster corresponds to a business service.

Step S34: and under the condition that the node is determined to be in the abnormal state, acquiring the node identification of the node in the abnormal state.

Step S36: and setting a mark corresponding to the node identifier of the node in the abnormal state in the node list, or removing the node identifier of the node in the abnormal state from the node list, so that the management server does not provide the node identifier of the node in the abnormal state to a routing device sending the node determination request under the condition that the management server receives the node determination request.

In this embodiment, the determining that the node is in the abnormal state may include: within a specified feedback duration after a state request is sent to a node, no feedback of the node is received; or after sending the state request to the node, receiving the state information which is fed back by the node and indicates that the node is in an abnormal state; or monitoring the task processing progress of the node, wherein the task processing progress of the node does not progress in the specified processing duration. So that the management server can know the abnormal state of the node. Specifically, when the response time of the node is out, that is, the specified feedback duration has no feedback, the node is considered to be in an abnormal state. And when the node actively sends out the state information representing the abnormal state, the node can know that the node is in the abnormal state. The node actively informs the management server, and the management server is prevented from finding that the node is in an abnormal working state and is not timely enough. Furthermore, in some cases, the node may respond normally to the access of the management server, but the application inside the node may already have a running error exception and cannot respond to the call request. By means of the mode that the node informs the management server, the management server can obtain the working state of the node relatively truly. Moreover, the management server can also determine the working state of the node by monitoring the task processing progress of the node.

In the present embodiment, the node list maintained by the management server may have only the node identifier of the node in the normal state. And when the node is no longer in the normal state, removing the node identification of the node from the node list. Therefore, the data size of the node list can be reduced to a certain extent, and the management server is convenient to improve the response speed of the node determination request. Of course, the management server may also set a flag in the node list corresponding to the node identifier of the node in the abnormal state. The flag may be used to indicate that the node represented by the corresponding node identification is in an abnormal state. The tag itself may be a predefined string. Therefore, the node list of the management server can record the state information of the nodes more comprehensively.

Please refer to fig. 10. The embodiment of the specification provides a server. The server has a normal node list in a normal state; wherein, the normal node list includes at least one node identifier for representing a node. The server also includes the following modules.

The state polling module is used for polling the working state of the nodes in the node cluster; the node cluster corresponds to a business service.

And the identifier acquisition module is used for acquiring the node identifier of the node in the abnormal state under the condition that the node is determined to be in the abnormal state.

A list management module, configured to set a flag in the node list corresponding to the node identifier of the node in the abnormal state, or remove the node identifier of the node in the abnormal state from the normal node list, so that the management server does not provide the node identifier of the node in the abnormal state to the routing device that sent the node determination request when receiving the node determination request.

In this embodiment, the functions and effects implemented by the server may be explained in comparison with other embodiments, and are not described in detail.

Embodiments of the present description also provide a computer storage medium having stored thereon computer program instructions and a node list; wherein, the node list comprises at least one node identifier for representing a node; the computer program instructions when executed implement: polling the working state of nodes in the node cluster; the node cluster corresponds to a business service; under the condition that the node is determined to be in an abnormal state, acquiring a node identifier of the node in the abnormal state; and setting a mark corresponding to the node identifier of the node in the abnormal state in the node list, or removing the node identifier of the node in the abnormal state from the node list, so that the management server does not provide the node identifier of the node in the abnormal state to a routing device sending the node determination request under the condition that the management server receives the node determination request.

Please refer to fig. 11. The embodiment of the specification provides an application loading method, which can be applied to a management end. The management terminal may be an electronic device with certain data processing capability. It has administrative rights for the nodes running applications in the system. Therefore, the working personnel can issue corresponding instructions through the management terminal, and the application is transferred among different node clusters. The application loading method may include the following steps.

Step S38: determining a target node cluster; wherein the target node cluster includes a plurality of nodes.

In this embodiment, in some cases, the application access amount corresponding to one node cluster is very large, which results in that the nodes in the node cluster have very large workload. While the application access volume of another cluster of nodes is relatively small. At this time, the application may be transferred to the node cluster with a smaller workload to run.

In this embodiment, the determining the target node cluster may include: a target node cluster is appointed in a mode of inputting a cluster identifier through a management terminal; or, according to the workload of each node cluster, selecting one node cluster with smaller workload. It is to be understood that other modifications may occur to those skilled in the art without departing from the spirit and scope of the embodiments described herein, and that all such modifications and variations are intended to be included herein within the scope of the present invention as defined in the appended claims.

Step S40: sending an application loading request to nodes in the target node cluster; wherein, the application loading request is attached with an application identifier.

In this embodiment, the management terminal may cause the node in the target node cluster to load the application represented by the application identifier by sending the application loading request. And in the process that the target node cluster loads the application represented by the application identifier, the original node cluster is still in a state of running the application represented by the application identifier. Therefore, the calling of the application by the end user is not influenced in the process of transferring the application among different node clusters. Therefore, the whole system runs stably.

In this embodiment, after the target node cluster receives the application loading request, the corresponding application may be loaded according to the application identifier. Specifically, for example, a node in the target node cluster may access the storage server to download the program instructions of the application according to the application identification. Alternatively, the node may also load the application for finding the application from the data stored in the node. In some cases, a node loading an application may be placing the application's program instructions into the node's cache memory so that quick calls may be made to the application.

Step S42: and receiving loading state information fed back by the nodes in the target node cluster.

In this embodiment, after the nodes in the target node cluster finish loading the application, the management server may be notified. Therefore, the management server can know the loading condition of the application in the target node cluster. Specifically, after each node in the target node cluster finishes loading the application, a piece of loading state information may be sent to the management server. Of course, each target node cluster may also have a node serving as a management function, and the nodes having the management function may uniformly feed back the loading state information. The node with the management function may send the loading state information only after all the nodes of the target node cluster are loaded with the application.

In this embodiment, the node is not limited to sending the loading state information when the loading is completed, and may send the loading state information during the loading process. As such, the load status information may be used to represent the progress of loading of the node for the application. Specifically, for example, the loading status information may be "Onloading".

Step S44: under the condition that the loading state information indicates that the nodes in the target node cluster finish loading the application represented by the application identifier, sending an application offline request to an original node cluster running the application represented by the application identifier so as to stop running the application by the nodes in the original node cluster; wherein the target node cluster is different from the original node cluster.

In this embodiment, corresponding to the foregoing description, the management side may consider that the nodes in the target node cluster complete the loading of the application when receiving the loading status information. The content of the loading state information may also indicate that the node in the target node cluster completes the loading of the application. Specifically, for example, the load status information is "Done". At this point, the nodes in the target node cluster are considered to complete the loading of the application.

In this embodiment, the original node cluster may be a node cluster that runs the application before the management end designates the target node cluster. In some cases, a node in the original node cluster may have multiple applications running. Because the frequency of applications in the original node cluster is very high, the workload of the original node cluster is very large. And the application in the target node cluster is called with low frequency, so that the node workload in the target node cluster is lower. Thus, the management side can transfer at least one application in the original node cluster to the target node cluster by the method of the embodiment.

In this embodiment, the original node cluster and the target node cluster may correspond to the same service. Therefore, the application transfer is carried out between the node clusters corresponding to the business services, and one node cluster cannot run the applications of different business services. Therefore, the transfer of the application is realized, and the disorder of the whole system is not caused.

In this embodiment, after the node of the target node cluster finishes loading the application, the management end instructs the node of the original node cluster to stop running the application. Thus, the application is transferred from the original node cluster to the target node cluster. Further, the workload of the original node cluster is reduced.

In one embodiment, the application loading method may further include: sending a change request to a routing device to change the cluster identifier corresponding to the application identifier in the configuration information of the routing device; the configuration information is recorded with an application identifier and a cluster identifier correspondingly, and the node in the node cluster represented by the cluster identifier runs the application represented by the application identifier.

In this embodiment, the routing device may have configuration information, and is configured to send, according to content of the configuration information, a call request to a node cluster represented by a cluster identifier corresponding to the application identifier when the routing device receives the call request from the client. Specifically, for example, amazon's application identification is "yamaxun" and the corresponding cluster is identified as "2345". The cluster of nodes represented by cluster identification "2345" includes three nodes, i.e., the three nodes each run amazon application.

In the present embodiment, the management side transmits a change request to the router apparatus, and causes the router apparatus to change the configuration information. Therefore, the cluster identifier corresponding to the application identifier is changed from the cluster identifier of the original node cluster to the cluster identifier of the target node cluster. Specifically, for example, the application identifies "yamaxun" as the original node's cluster identification "2345" and the target node's cluster identification "4567". As such, when the routing device receives a call request for amazon, a node in the node cluster represented by cluster identity "4567" may be called.

Please refer to fig. 12. The embodiment of the specification also provides a server. The server may include the following modules.

The cluster determining module is used for determining a target node cluster; wherein the target node cluster includes a plurality of nodes.

A loading request sending module, configured to send an application loading request to a node in the target node cluster; wherein, the application loading request is attached with an application identifier.

And the state information receiving module is used for receiving the loading state information fed back by the nodes in the target node cluster.

A offline request sending module, configured to send an application offline request to an original node cluster running the application indicated by the application identifier, so that a node in the original node cluster stops running the application, when the loading state information indicates that the node in the target node cluster finishes loading the application indicated by the application identifier; wherein the target node cluster is different from the original node cluster.

In this embodiment, the functions and effects achieved by the modules of the server may be explained in comparison with other embodiments, and are not described in detail.

Embodiments of the present description also provide a computer storage medium storing computer program instructions that, when executed, implement: determining a target node cluster, wherein the target node cluster comprises a plurality of nodes; sending an application loading request to a node in the target node cluster, wherein the application loading request is attached with an application identifier; receiving loading state information fed back by the nodes in the target node cluster; and under the condition that the loading state information indicates that the nodes in the target node cluster finish loading the application represented by the application identifier, sending an application offline request to an original node cluster running the application represented by the application identifier so as to stop running the application by the nodes in the original node cluster, wherein the target node cluster is different from the original node cluster.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.

The foregoing description of various embodiments of the present specification is provided for the purpose of illustration to those skilled in the art. It is not intended to be exhaustive or to limit the invention to a single disclosed embodiment. As described above, various alternatives and modifications of the present specification will be apparent to those skilled in the art to which the above-described technology pertains. Thus, while some alternative embodiments have been discussed in detail, other embodiments will be apparent or relatively easy to derive by those of ordinary skill in the art. The description is intended to embrace all alternatives, modifications, and variations of the present invention that have been discussed herein, as well as other embodiments that fall within the spirit and scope of the above-mentioned application.

The description is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

While the specification has been described with respect to the embodiments, those skilled in the art will appreciate that there are numerous variations and permutations of the specification that fall within the spirit and scope of the specification, and it is intended that the appended claims include such variations and modifications as fall within the spirit and scope of the specification.

Claims

1. An application calling method, comprising:

receiving a calling request, wherein the calling request is attached with an application identifier;

determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of nodes;

determining a target node in a node cluster corresponding to the target business service;

sending a calling request to the target node to call the application represented by the application identifier;

determining a target business service in a plurality of business services according to the calling request, wherein the determining the target business service comprises the following steps:

determining a client identifier of a client sending the call request;

inquiring an access record of the client identification; wherein, the access record records the access data of the client;

and determining the target business service corresponding to the calling request according to the access record.

2. The method of claim 1, wherein determining a target node in the node cluster corresponding to the target business service comprises:

sending a node determination request to a management server; wherein the node determination request is accompanied by a cluster identifier; the cluster identifier is used for representing a node cluster corresponding to the target business service;

and receiving the node identification of the node determined by the management server in the node cluster represented by the cluster identification, wherein the node represented by the node identification is used as the target node.

3. The method of claim 1, wherein determining a target node in the node cluster corresponding to the target business service comprises:

reading node list information corresponding to the target service in a configuration module; the node list information comprises at least one node identification; the node identification is used for representing a node;

and determining a target node in the nodes represented by the node list information.

4. An application calling method, comprising:

determining a first target business service and a second target business service in a plurality of business services according to the calling request;

correspondingly, determining a target node in the node cluster corresponding to the target service includes:

determining a first target node in a node cluster corresponding to the first target business service; determining a second target node in the node cluster corresponding to the second target business service;

correspondingly, sending a call request to the target node includes:

sending a first calling request to the first target node to call the application represented by the application identification;

receiving first result data fed back by the first target node;

and sending a second calling request to the second target node, wherein the second calling request is accompanied by the first result data, so that the application represented by the application identifier in the second target node generates feedback data based on the first result data.

5. The method of claim 4, wherein determining the first target node in the node cluster corresponding to the first target business service comprises:

sending a first node determination request to a management server; the first node determines that the request is attached with a cluster identifier of a node cluster corresponding to the first target business service;

and receiving the node identification of the node determined by the management server in the node cluster represented by the cluster identification, wherein the node represented by the node identification is used as the first target node.

6. The method of claim 4, wherein determining the first target node in the node cluster corresponding to the first target business service comprises:

reading node list information corresponding to the first target service in a configuration module; the node list information comprises at least one node identification; the node identification is used for representing a node;

and determining a first target node in the nodes represented by the node list information.

7. The method of claim 4, wherein the invocation request is further accompanied by a service identifier;

determining a first target business service and a second target business service in a plurality of business services according to the calling request, wherein the method comprises the following steps:

determining a first target business service and a second target business service in the plurality of business services according to the business identification; and the service identification corresponds to the first target service and the second target service.

8. The method of claim 4, wherein the invocation request is accompanied by user-entered dialog data;

and determining a first target business service and a second target business service which respond to the calling request according to the dialogue data.

9. A routing device, comprising:

the receiving module is used for receiving a calling request, and the calling request is attached with an application identifier;

the service determining module is used for determining a target business service in a plurality of business services according to the calling request; the business service corresponds to the node cluster; the node cluster comprises a plurality of nodes;

the node determining module is used for determining a target node in a node cluster corresponding to the target business service;

the request sending module is used for sending a calling request to the target node so as to call the application represented by the application identifier;

wherein the service determination module is specifically configured to: determining a client identifier of a client sending the call request; inquiring the access record of the client identification; wherein, the access record records the access data of the client; and determining the target business service corresponding to the call request according to the access record.

10. A routing device, comprising:

the node determining module is used for determining a target node in the node cluster corresponding to the target business service;

wherein the service determination module is specifically configured to: determining a first target business service and a second target business service in a plurality of business services according to the calling request;

correspondingly, the node determining module is specifically configured to: determining a first target node in a node cluster corresponding to the first target business service; determining a second target node in the node cluster corresponding to the second target business service;

correspondingly, the request sending module is specifically configured to: sending a first calling request to the first target node to call the application represented by the application identification; receiving first result data fed back by the first target node; and sending a second calling request to the second target node, wherein the second calling request is accompanied by the first result data, so that the application represented by the application identifier in the second target node generates feedback data based on the first result data.

11. A computer storage medium having computer program instructions stored thereon that when executed implement: receiving a calling request, wherein the calling request is attached with an application identifier; determining a target business service in a plurality of business services according to the calling request, wherein the business service corresponds to a node cluster, and the node cluster comprises a plurality of nodes; determining a target node in a node cluster corresponding to the target business service; sending a calling request to the target node to call the application represented by the application identifier;

determining a client identifier of a client sending the call request;

inquiring the access record of the client identification; wherein, the access record records the access data of the client;

and determining the target business service corresponding to the call request according to the access record.

12. A computer storage medium having computer program instructions stored thereon that when executed implement: receiving a calling request, wherein the calling request is attached with an application identifier; determining a target business service in a plurality of business services according to the calling request, wherein the business service corresponds to a node cluster, and the node cluster comprises a plurality of nodes; determining a target node in a node cluster corresponding to the target business service; sending a calling request to the target node to call the application represented by the application identifier;

determining a target business service in a plurality of business services according to the call request, wherein the determining the target business service comprises the following steps:

correspondingly, sending a call request to the target node includes:

receiving first result data fed back by the first target node;