WO2024066503A1 - 服务调用方法及装置 - Google Patents

服务调用方法及装置 Download PDF

Info

Publication number
WO2024066503A1
WO2024066503A1 PCT/CN2023/101382 CN2023101382W WO2024066503A1 WO 2024066503 A1 WO2024066503 A1 WO 2024066503A1 CN 2023101382 W CN2023101382 W CN 2023101382W WO 2024066503 A1 WO2024066503 A1 WO 2024066503A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
discovery data
cache
grid
calling
Prior art date
Application number
PCT/CN2023/101382
Other languages
English (en)
French (fr)
Inventor
李来
杨奕
Original Assignee
华为云计算技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为云计算技术有限公司 filed Critical 华为云计算技术有限公司
Publication of WO2024066503A1 publication Critical patent/WO2024066503A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
    • H04L67/5682—Policies or rules for updating, deleting or replacing the stored data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/63—Routing a service request depending on the request content or context

Definitions

  • the present application relates to the field of cloud computing technology, and in particular to a service calling method and device.
  • Microservices architecture is a cloud-native architecture. Based on microservices architecture, applications can be composed of many loosely coupled and independently deployable smaller components (usually called microservices), each with its own area of responsibility. When processing user requests, microservices-based applications call multiple internal microservices to jointly generate their responses.
  • a microservice architecture when a service is called, the loading of service discovery data is usually based on a network proxy implemented in a sidecar manner. In this way, the sidecar needs to occupy a separate container process, which consumes a lot of resources. In addition, since the sidecar and the service it mounts belong to different containers, the cross-process communication between the two will cause latency, affecting the processing efficiency when the service is called, thereby affecting the service performance.
  • an embodiment of the present application provides a service calling method, which is applied to a first service, wherein the first service is a microservice running in a service grid, and the method comprises: when the first service calls a second service, determining whether service discovery data of the second service exists in the cache of the first service; when the service discovery data of the second service does not exist in the cache, intercepting the calling request issued by the first service; obtaining the service discovery data of the second service from the control plane of the service grid, and storing the service discovery data of the second service in the cache; and executing a call from the first service to the second service based on the service discovery data of the second service.
  • the service calling method of the embodiment of the present application when the first service calls the second service, first determines whether there is service discovery data of the second service in the cache of the first service. If the service discovery data of the second service does not exist in the cache, the calling request issued by the first service is intercepted, and then the service discovery data of the second service is obtained from the control plane of the service grid, and the service discovery data of the second service is stored in the cache of the first service. Thereafter, based on the service discovery data of the second service, the first service executes the call to the second service.
  • the call request can be intercepted and the service discovery data of the second service can be obtained directly from the control plane of the service grid.
  • This not only enables lazy loading of service discovery data and reduces resource consumption when loading service discovery data, but also eliminates the need to use sidecar as a network proxy when loading service discovery data, thus avoiding cross-process communication. This reduces resource consumption when calling services, improves processing efficiency when calling services, and thus improves service performance.
  • the method further includes: when the service discovery data of the second service exists in the cache, according to the service discovery data of the second service in the cache, Now the data is available and the second service is called.
  • the first service when the service discovery data of the second service exists in the cache of the first service, the first service can directly call the second service according to the service discovery data of the second service in the cache of the first service, thereby improving the processing efficiency when calling the service.
  • the service discovery data includes an IP address and load information of an instance of a service
  • calling the second service according to the service discovery data of the second service in the cache includes: selecting a target instance from the instances of the second service according to a preset load balancing rule and the load information of the instance of the second service; and calling the target instance according to the IP address of the target instance.
  • the target instance when calling the second service according to the service discovery data of the second service in the cache, can be selected from the instances of the second service according to the preset load balancing rules and the load information of the instances of the second service; then the target instance is called according to the IP address of the target instance, so as to realize the calling of the second service by the first service. In this way, the processing efficiency of service calling can be improved.
  • the method also includes: during the operation of the first service, recording the link information of the first service through the link tracking service in the service grid.
  • the link information of the first service is recorded through the link tracking service in the service grid, so that when the first service is restarted later, the service discovery data is preloaded according to the historical link information of the first service.
  • the method further includes: during the restart process of the first service, obtaining historical calling information of the first service by calling the link tracking service, the historical calling information including identifiers of services called by the first service determined by the link tracking service according to the historical link information; after the first service is started, obtaining service discovery data of each service included in the historical calling information from the control plane of the service grid; and storing the obtained service discovery data in the cache.
  • the link tracking service can be called to obtain the historical call information of the first service, and after the first service is started, the service discovery data of each service included in the historical call information can be obtained from the control plane of the service grid, and the obtained service discovery data can be stored in the cache.
  • the service discovery data can be preloaded when the first service is started, thereby reducing the time cost of obtaining the service discovery data for the first time, improving the processing efficiency when the service is called, and thus improving the service performance.
  • the method further includes: during the operation of the first service, periodically calling the link tracking service to periodically update the service discovery data in the cache.
  • the service discovery data in the cache can be periodically updated by periodically calling the link tracking service, thereby realizing dynamic lazy loading of the service discovery data.
  • the method further includes: subscribing to the service discovery data in the cache to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated.
  • the service discovery data in the cache of the first service can be subscribed to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated, thereby enabling the service discovery data in the cache of the first service to be dynamically updated, thereby improving the accuracy of the service discovery data in the cache of the first service.
  • the method is implemented in a plug-in manner based on a Java agent, and the plug-in is mounted to the first service.
  • the plug-in is implemented based on the Java agent and mounted on the first service. Not only can the plug-in implementing the embodiment of the present application be dynamically loaded into the first service without intrusion, but the plug-in and the first service belong to the same process, reducing cross-process communication. Therefore, when the service is called, the Java plug-in can be used to replace the sidecar in the existing service grid, so that the service consumes less resources and has better performance.
  • an embodiment of the present application provides a service calling device, which is applied to a first service, wherein the first service is a microservice running in a service grid, and the device includes: a judgment module, which is used to judge whether there is service discovery data of the second service in the cache of the first service when the first service calls the second service; an interception module, which is used to intercept the call request issued by the first service when the service discovery data of the second service does not exist in the cache; a first acquisition module, which is used to obtain the service discovery data of the second service from the control plane of the service grid and store the service discovery data of the second service in the cache; and a first calling module, which executes the call of the first service to the second service based on the service discovery data of the second service.
  • the service calling device of an embodiment of the present application when the first service calls the second service, first determines whether there is service discovery data of the second service in the cache of the first service. If the service discovery data of the second service does not exist in the cache, the service calling device intercepts the calling request issued by the first service, then obtains the service discovery data of the second service from the control plane of the service grid, and stores the service discovery data of the second service in the cache of the first service. Thereafter, based on the service discovery data of the second service, the first service executes the calling of the second service.
  • the call request can be intercepted and the service discovery data of the second service can be obtained directly from the control plane of the service grid.
  • This not only enables lazy loading of service discovery data and reduces resource consumption when loading service discovery data, but also eliminates the need to use sidecar as a network proxy when loading service discovery data, thus avoiding cross-process communication. This reduces resource consumption when calling services, improves processing efficiency when calling services, and thus improves service performance.
  • the device also includes: a second calling module, which calls the second service according to the service discovery data of the second service in the cache when service discovery data of the second service exists in the cache.
  • the first service when the service discovery data of the second service exists in the cache of the first service, the first service can directly call the second service according to the service discovery data of the second service in the cache of the first service, thereby improving the processing efficiency when calling the service.
  • the service discovery data includes an IP address and load information of an instance of the service
  • the second calling module includes: a target instance selection submodule, used to select a target instance from the instances of the second service according to a preset load balancing rule and the load information of the instance of the second service; a target instance calling submodule, used to call the target instance according to the IP address of the target instance.
  • the target instance when calling the second service according to the service discovery data of the second service in the cache, can be selected from the instances of the second service according to the preset load balancing rules and the load information of the instances of the second service; then the target instance is called according to the IP address of the target instance, so as to realize the calling of the second service by the first service. In this way, the processing efficiency of service calling can be improved.
  • the device in a third possible implementation manner of the service calling device, also includes: a link information recording module, used to record the link information of the first service through the link tracking service in the service grid during the operation of the first service.
  • the link information of the first service is recorded through the link tracking service in the service grid, so that when the first service is restarted later, the service discovery data is preloaded according to the historical link information of the first service.
  • the device further includes: a historical call information acquisition module, used to acquire historical call information of the first service by calling the link tracking service during the restart of the first service, the historical call information including identifiers of services called by the first service as determined by the link tracking service based on the historical link information; a second acquisition module, used to acquire service discovery data of each service included in the historical call information from the control plane of the service grid after the first service is started; and a storage module, used to store the acquired service discovery data in the cache.
  • a historical call information acquisition module used to acquire historical call information of the first service by calling the link tracking service during the restart of the first service, the historical call information including identifiers of services called by the first service as determined by the link tracking service based on the historical link information
  • a second acquisition module used to acquire service discovery data of each service included in the historical call information from the control plane of the service grid after the first service is started
  • a storage module used to store the acquired service discovery data in the cache.
  • the link tracking service can be called to obtain the historical call information of the first service, and after the first service is started, the service discovery data of each service included in the historical call information can be obtained from the control plane of the service grid, and the obtained service discovery data can be stored in the cache.
  • the service discovery data can be preloaded when the first service is started, thereby reducing the time cost of obtaining the service discovery data for the first time, improving the processing efficiency when the service is called, and thus improving the service performance.
  • the device in the fifth possible implementation method of the service calling device, also includes: an update module, which is used to periodically update the service discovery data in the cache by periodically calling the link tracking service during the operation of the first service.
  • the service discovery data in the cache can be periodically updated by periodically calling the link tracking service, thereby realizing dynamic lazy loading of the service discovery data.
  • the device in a sixth possible implementation manner of the service calling device, also includes: a subscription module, used to subscribe to the service discovery data in the cache to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated.
  • a subscription module used to subscribe to the service discovery data in the cache to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated.
  • the service discovery data in the cache of the first service can be subscribed to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated, thereby enabling the service discovery data in the cache of the first service to be dynamically updated, thereby improving the accuracy of the service discovery data in the cache of the first service.
  • the device in a seventh possible implementation manner of the service calling device, is implemented in a plug-in manner based on a Java agent, and the plug-in is mounted to the first service.
  • the Java agent is implemented in a plug-in manner, and the plug-in is mounted to the first service.
  • the plug-in of the embodiment of the present application is dynamically loaded into the first service without intrusion, and the plug-in and the first service belong to the same process, reducing cross-process communication. Therefore, when the service is called, the Java plug-in can be used to replace the sidecar in the existing service grid, so that the service consumes less resources and has better performance.
  • an embodiment of the present application provides a computing device cluster, comprising at least one computing device, each computing device comprising a processor and a memory; the processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the service calling method of the above-mentioned first aspect or one or more of the multiple possible implementation methods of the first aspect.
  • an embodiment of the present application provides a computer program product comprising instructions, which, when executed by a computing device cluster, enables the computing device cluster to execute the service calling method of the above-mentioned first aspect or one or more of the multiple possible implementations of the first aspect.
  • an embodiment of the present application provides a computer-readable storage medium, comprising computer program instructions.
  • the computer program instructions When executed by a computing device cluster, the computing device cluster executes the service calling method of the above-mentioned first aspect or one or several of the multiple possible implementation methods of the first aspect.
  • FIG1 is a schematic diagram showing an application scenario of a service invocation method according to an embodiment of the present application.
  • FIG2 shows a flow chart of a service invoking method according to an embodiment of the present application.
  • FIG. 3 is a schematic diagram showing a processing procedure of a service calling method according to an embodiment of the present application.
  • FIG. 4 is a schematic diagram showing a processing procedure of a service calling method according to an embodiment of the present application.
  • FIG. 5 is a schematic diagram showing a processing procedure of a service calling method according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram showing a processing procedure of a service calling method according to an embodiment of the present application.
  • FIG. 7 shows a block diagram of a service invoking device according to an embodiment of the present application.
  • Service governance Manage and govern microservices, such as service registration, service discovery, load balancing, application routing, etc.
  • Service discovery refers to the ability of a client service in a microservice architecture to automatically discover a list of service addresses.
  • the current process can be seen as the process of automatically loading service discovery data for the client service.
  • microservices can communicate with each other without having to perceive the deployment location and IP address of the other end.
  • Service mesh is the infrastructure layer that handles communication between services in a microservice architecture. It supports reliable delivery of network requests of cloud-native applications in complex topology environments. Service mesh is deployed with applications but is transparent to them. Service mesh usually includes a data plane and a control plane.
  • Service visibility refers to the fact that the current service can obtain the service discovery data of the services it calls from the control plane of the service mesh.
  • the following uses the open source service mesh Istio as an example to illustrate the service mesh.
  • the open source service mesh Istio includes a data plane and a control plane.
  • the data plane of Istio refers to the Envoy proxy layer. Istio injects the Envoy proxy as a sidecar container next to the service container, and then the Envoy proxy intercepts all inbound and outbound traffic of the service.
  • the control plane of Istio refers to the Istiod component layer, which provides functions such as service discovery, configuration, and certificate management.
  • control plane delivers service information and governance rules to the data plane based on the service discovery (xDS) protocol; the data plane uses the xDS protocol as its application programming interface (API) and communicates with the control plane based on the xDS protocol.
  • xDS service discovery
  • API application programming interface
  • the x in the xDS protocol indicates that the xDS protocol does not refer to a specific protocol, but a general term for a group of service discovery protocols based on different data sources, including listener discovery service (LDS), cluster discovery service (CDS), endpoint discovery service (EDS), route discovery service (RDS), aggregated discovery service (ADS), health discovery service (HDS), secret discovery service (SDS), metric service (MS), rate limit service (RLS), etc.
  • LDS listener discovery service
  • CDS cluster discovery service
  • EDS endpoint discovery service
  • RDS route discovery service
  • ADS aggregated discovery service
  • HDS health discovery service
  • SDS secret discovery service
  • MS metric service
  • RLS rate limit service
  • the loading of service discovery data is usually based on a network proxy implemented in a sidecar manner.
  • each service container has a sidecar container for implementing the Envoy proxy.
  • the loading of service discovery data is all done through the sidecar container.
  • the sidecar needs to occupy a separate container process, which consumes a lot of resources.
  • the sidecar and the service it mounts belong to different containers. The cross-process communication between the two will cause latency, affecting the processing efficiency when the service is called, thereby affecting the service performance.
  • the loading of service discovery data based on sidecar is not suitable for large-scale scenarios and has complex deployment and operation and maintenance.
  • Istio uses a full delivery strategy to deliver service discovery data, that is, all sidecar processes in the service grid hold all service discovery data in the entire grid.
  • service discovery data is useless for workloads, and in large-scale scenarios, the full delivery of service discovery data will lead to serious resource consumption, that is, the full delivery strategy of service discovery data is not suitable for large-scale scenarios.
  • the loading of service discovery data adopts an on-demand loading strategy.
  • the dependencies between services are preconfigured in advance, and during the service call process, the service discovery data is loaded according to the preconfigured dependencies between services.
  • this method usually requires manual configuration, is not suitable for large-scale scenarios, and is difficult to maintain.
  • dedicated intermediate gateway services and controller services can be deployed in Istio to achieve on-demand loading of service discovery data.
  • this method requires starting additional intermediate gateway services and controller services in the service grid, which not only consumes resources, but also makes the deployment and operation of services more complicated.
  • an embodiment of the present application provides a service calling method, which is applied to a first service, where the first service is a microservice running in a service grid, and the method comprises: when the first service calls a second service, determining whether there is service discovery data of the second service in the cache of the first service; and if there is no service discovery data of the second service in the cache, determining whether there is service discovery data of the second service in the cache.
  • the call request issued by the first service is intercepted; the service discovery data of the second service is obtained from the control plane of the service grid, and the service discovery data of the second service is stored in the cache; based on the service discovery data of the second service, the first service executes the call to the second service.
  • the service calling method of the embodiment of the present application when the first service calls the second service, first determines whether there is service discovery data of the second service in the cache of the first service. If the service discovery data of the second service does not exist in the cache, the calling request issued by the first service is intercepted, and then the service discovery data of the second service is obtained from the control plane of the service grid, and the service discovery data of the second service is stored in the cache of the first service. Thereafter, based on the service discovery data of the second service, the first service executes the call to the second service.
  • the call request can be intercepted and the service discovery data of the second service can be obtained directly from the control plane of the service grid.
  • This not only enables lazy loading of service discovery data and reduces resource consumption when loading service discovery data, but also eliminates the need to use sidecar as a network proxy when loading service discovery data, thus avoiding cross-process communication. This reduces resource consumption when calling services, improves processing efficiency when calling services, and thus improves service performance.
  • lazy loading is delayed loading, which is a strategy that does not preload non-essential resources and only loads them when needed.
  • Lazy loading is an on-demand loading that can reduce resource usage.
  • the service calling method of the embodiment of the present application can be applied to the service governance scenario in the microservice architecture.
  • the service grid of the actual application scenario usually more than 100,000 or even millions of service instances are deployed.
  • the service calling method of the embodiment of the present application can be used to call the service, and lazy loading of service discovery data based on the sidecar can be realized when the service is called, which can not only reduce the resource consumption when the service is called, but also improve the processing efficiency when the service is called, thereby improving the service performance.
  • FIG1 shows a schematic diagram of an application scenario of a service call method according to an embodiment of the present application.
  • a terminal device 130 is connected to a cloud computing center 110 via a network (e.g., a wired network, a wireless network, etc.), and the cloud computing center 110 includes a server cluster 120, which provides services for applications on the terminal device 130 based on a microservice architecture.
  • the terminal device 130 can be a smart phone, a netbook, a tablet computer, a laptop computer, a wearable electronic device (such as a smart bracelet, a smart watch, etc.), a TV, a virtual reality device, a stereo, an electronic ink, and the like.
  • This application does not limit the specific type of the terminal device.
  • the server cluster 120 calls multiple internal microservices to jointly generate a response for the application.
  • the server cluster 120 calls multiple internal microservices, there will be calls between microservices, such as a call from one microservice to another microservice.
  • the service call method of the embodiment of the present application can be used to implement a call from one microservice to another microservice.
  • FIG2 shows a flow chart of a service calling method according to an embodiment of the present application.
  • the service calling method of the embodiment of the present application can be applied to a first service, where the first service is a microservice running in a service grid.
  • the service grid runs on the server cluster 120 shown in FIG1
  • the first service is a microservice running in the service grid as a call initiator.
  • the service calling method of the embodiment of the present application includes:
  • Step S210 When the first service calls the second service, it is determined whether service discovery data of the second service exists in the cache of the first service.
  • the first service and the second service are both microservices running in the service grid.
  • the first service and the second service both have corresponding caches (e.g., cache).
  • the service discovery data loaded by the first service can be stored in the cache of the first service, and the service discovery data loaded by the second service can be stored in the cache of the second service.
  • the second service can be regarded as the service provider, and the first service can be regarded as the service consumer or client.
  • the first service calls the second service, it can first determine whether the service discovery data of the second service exists in the cache of the first service. For example, assuming that the first service calls the second service through the identifier of the second service (for example, the ID, name, etc. that can uniquely identify the second service), when determining whether the service discovery data of the second service exists in the cache of the first service, it can be determined whether the service discovery data of the second service exists in the cache of the first service by searching, comparing, etc. based on the identifier of the second service.
  • the identifier of the second service for example, the ID, name, etc.
  • a service visibility list of the first service may be set in the cache of the first service.
  • the service visibility list of the first service may be used to store identifiers of services to which service discovery data stored in the cache of the first service belongs. For example, assuming that the cache of the first service S0 stores service discovery data of services S1, S2, S3, S4, and S5, the identifiers of services S1, S2, S3, S4, and S5 may be stored in the service visibility list of the first service S0.
  • determining whether service discovery data of a second service exists in the cache of a first service it may be determined whether an identifier of the second service exists in the service visibility list of the first service. If the identifier of the second service exists in the service visibility list of the first service, it may be considered that service discovery data of the second service exists in the cache of the first service. If the identifier of the second service does not exist in the service visibility list of the first service, it may be considered that service discovery data of the second service does not exist in the cache of the first service.
  • Step S220 When the service discovery data of the second service does not exist in the cache, intercept the call request issued by the first service.
  • the call request issued by the first service can be intercepted to obtain the service discovery data of the second service from the control plane of the service grid.
  • Step S230 Obtain service discovery data of the second service from the control plane of the service grid, and store the service discovery data of the second service in the cache.
  • the service discovery data of the second service can be obtained from the control plane of the service grid according to the identifier of the second service.
  • the service discovery data of the second service may include the identifier of the second service (for example, an ID and name that can uniquely identify the second service), the number of instances of the second service and the identifier of each instance (for example, an ID and name that can uniquely identify each instance), an IP address, load information, etc. It should be noted that the service discovery data of the second service may also include other information, and those skilled in the art may set the specific content of the service discovery data according to actual conditions, and this application does not limit this.
  • the obtained service discovery data of the second service may be stored in the cache of the first service.
  • the identifier of the second service may be stored in the service visibility list of the first service.
  • the service discovery data of the second service can also be subscribed to the control plane of the service grid.
  • the control plane of the service grid After subscribing to the service discovery data of the second service from the control plane of the service grid, when the control plane of the service grid finds that the service discovery data of the second service is updated, the control plane of the service grid pushes the updated service discovery data of the second service to the first service.
  • the first service When the first service receives the updated service discovery data (i.e., new service discovery data) of the second service pushed by the control plane of the service grid, it uses the service discovery data of the second service to
  • the updated service discovery data of the first service i.e., the new service discovery data
  • replaces the service discovery data of the second service in the cache i.e., the old service discovery data.
  • the service discovery data of the second service in the cache of the first service can be kept dynamically updated, thereby improving the accuracy of the service discovery data of the second service in the cache.
  • Step S240 Based on the service discovery data of the second service, the first service calls the second service.
  • the first service can call the second service according to the service discovery data of the second service in the cache of the first service.
  • the service discovery data of the second service can be obtained from the cache of the first service, and the load information of the instance of the second service can be obtained from the service discovery data; then, according to the preset load balancing rules (such as the lowest load priority, etc.) and the load information of the instance of the second service, the target instance is selected from the instance of the second service; then, the IP address of the target instance can be obtained from the service discovery data of the second service, and the target instance can be called according to the IP address of the target instance, so as to realize the call of the first service to the second service.
  • the preset load balancing rules such as the lowest load priority, etc.
  • a first service calls a second service
  • the second service can be directly called based on the service discovery data of the second service in the cache of the first service, thereby improving processing efficiency when calling services.
  • the service discovery data of the second service when calling the second service, can be obtained from the cache of the first service, and the load information of the instance of the second service can be obtained from the service discovery data; then, based on the preset load balancing rules (such as lowest load priority, etc.) and the load information of the instance of the second service, the target instance is selected from the instances of the second service; thereafter, the IP address of the target instance can be obtained from the service discovery data of the second service, and the target instance can be called based on the IP address of the target instance, thereby realizing the call of the first service to the second service.
  • the preset load balancing rules such as lowest load priority, etc.
  • Fig. 3 is a schematic diagram showing the processing process of the service invocation method according to an embodiment of the present application.
  • the cache 311 of the first service 310 stores the service discovery data of the second service 320 (service 320), as shown in the following example:
  • the first service 310 calls the second service 320, it is first determined whether the service discovery data of the second service 320 exists in the cache 311. After determination, if the service discovery data of the second service 320 exists in the cache 311, the first service 310 can obtain the service discovery data of the second service 320 from the cache 311, and directly call the second service 320 according to the obtained service discovery data of the second service 320.
  • Fig. 4 is a schematic diagram showing the processing process of the service invocation method according to an embodiment of the present application.
  • the cache 411 of the first service 410 stores the service discovery data of service F, as shown in the following example:
  • the first service 410 has a dependency on the second service 420.
  • the first service 410 calls the second service 420, it is first determined whether the service discovery data of the second service 420 (service420) exists in the cache 411. After determination, if the service discovery data of the second service 420 does not exist in the cache 411, in this case, the call request issued by the first service 410 can be intercepted, and then the service discovery data of the second service 420 can be obtained from the control plane 430 of the service grid, and the obtained service discovery data of the second service 420 is stored in the cache 411. In this process, the service discovery data of the second service 420 can also be subscribed to the control plane 430 of the service grid to obtain updates to the service discovery data of the second service 420.
  • the service discovery data in the cache 411 may be exemplified as follows:
  • the first service 410 may continue to call the second service 420 , specifically: the first service 410 obtains the service discovery data of the second service 420 from the cache 411 , and calls the second service 420 according to the obtained service discovery data of the second service 420 .
  • the link information of the first service may also be recorded through the link tracking service in the service grid.
  • the link information may include service dependency information (e.g., information about the service called by the first service).
  • the link information may also include other information such as the call link and the call request volume, which is not limited in this application.
  • the link tracking service may provide functions such as call link restoration, call request volume statistics, link topology, and dependency analysis.
  • the link information of the first service is tracked and recorded through the link tracking service in the service grid, so that when the first service is restarted later, the service discovery data of the service is preloaded according to the historical link information of the first service.
  • the historical call information of the first service may be obtained by calling the link tracking service in the service grid.
  • the historical call information may include the identifier of the service called by the first service determined by the link tracking service according to the historical link information of the first service.
  • the service discovery data of each service called by the first service included in the historical call information may be obtained from the control plane of the service grid, and the obtained service discovery data may be stored in the cache of the first service.
  • the link tracking service in the service grid can be called to obtain the historical call information of the first service S6, and the historical call information includes the identifiers of the services called by the first service S6 within a preset time period before the current moment: the identifiers of service S7, service S8, and service S9; after the first service S6 is started, the service S7, service S8, and service S9 can be obtained from the control plane of the service grid according to the identifiers of service S7, service S8, and service S9.
  • the service discovery data of service S9 is obtained, and the obtained service discovery data of service S7, service S8 and service S9 are stored in the cache of the first service S6.
  • the historical call information of the first service can be obtained through the link tracking service, and the service discovery data can be preloaded based on the historical call information of the first service, thereby reducing the time cost of obtaining the service discovery data for the first time, improving the processing efficiency during service calls, and thus improving service performance.
  • the identifiers of each service called by the first service included in the historical call information may be stored in the service visibility list of the first service.
  • the service discovery data of each third service can also be subscribed to the control plane of the service grid.
  • the control plane of the service grid After subscribing to the service discovery data of each third service to the control plane of the service grid, for any third service, when the control plane of the service grid finds that the service discovery data of the third service is updated, the control plane of the service grid will push the updated service discovery data of the third service to the first service.
  • the first service receives the updated service discovery data (i.e., new service discovery data) of the third service pushed by the control plane of the service grid, it uses the updated service discovery data (i.e., new service discovery data) of the third service to replace the service discovery data (i.e., old service discovery data) of the third service in the cache.
  • the service discovery data of the third service in the cache of the first service can be kept dynamically updated, thereby improving the accuracy of the service discovery data of the third service in the cache.
  • FIG5 is a schematic diagram showing the processing process of the service calling method according to an embodiment of the present application.
  • the historical calling information of the first service 510 can be obtained by calling the link tracking service 520.
  • the service discovery data of each service included in the historical calling information of the first service 510 can be obtained from the control plane 530 of the service grid, and the obtained service discovery data is stored in the cache 511; in this process, the service discovery data of each service included in the historical calling information of the first service 510 can also be subscribed to the control plane 530 of the service grid to obtain updates to the service discovery data.
  • the link tracking service in the service grid may be periodically called to periodically update the service discovery data in the cache.
  • the link tracking service in the service grid may be called according to a preset period to obtain the historical call information of the first service in the most recent period, and then the service discovery data in the cache of the first service may be updated according to the historical call information in the most recent period.
  • the service discovery data in the cache of the first service D1 is the service discovery data of service D2, service D3, and service D4, during the operation of the first service D1, according to the preset cycle, the link tracking service in the service grid is called, and the historical call information of the first service D1 in the most recent cycle obtained includes: the identifiers of service D2, service D3, service D5, and service D6. Then, according to the historical call information of the first service D1 in the most recent cycle, the service discovery data in the cache of the first service D1 can be updated, as follows:
  • the service discovery data of the service D6 is obtained from the control plane of the service grid, and the obtained service discovery data of the service D6 is stored in the cache of the first service D1.
  • the service discovery data in the cache can be periodically updated by periodically calling the link tracking service, thereby realizing dynamic lazy loading of the service discovery data.
  • the service visibility list of the first service when the service visibility list of the first service is set in the cache of the first service, the service visibility list of the first service may be updated according to the historical call information in the most recent period to improve the accuracy of the service visibility list of the first service.
  • the service calling method of the embodiment of the present application can be implemented in a plug-in manner based on a Java agent, and the plug-in is mounted to the first service.
  • Loading the Java Agent at startup is a new feature introduced after the Java Development Kit (JDK) 1.5. This feature provides users with the ability to modify the bytecode after the Java virtual machine (JVM) reads the bytecode file into memory and before the JVM uses the corresponding byte stream to generate a Class object in the Java heap.
  • JVM Java virtual machine
  • the JVM will also use the bytecode modified by the user to create the Class object, thereby achieving the purpose of dynamically enhancing the program logic.
  • the service calling method of the embodiment of the present application is implemented in a plug-in manner based on the Java agent, and the plug-in is mounted to the first service. Not only can the plug-in implementing the embodiment of the present application be dynamically loaded into the first service without intrusion, but the plug-in and the first service belong to the same process, reducing cross-process communication. Therefore, when the service is called, the Java plug-in can be used to replace the sidecar in the existing service grid, so that the service consumes less resources and has better performance.
  • the plug-in may include a service discovery data subscription module and a service call interception module.
  • the service discovery data subscription module is mainly used to establish a connection with the control plane of the service grid, obtain and subscribe to service discovery data.
  • the service discovery data subscription module can also be used to implement service governance functions such as load balancing, flow control, current limiting and degradation.
  • the service discovery data subscription module can call the link tracking module in the service grid to implement preloading of service discovery data.
  • the service discovery data subscription module can also periodically call the link tracking service in the service grid to periodically update the service discovery data in the cache.
  • the service call interception module can intercept the call request of the first service when the first service calls the second service and the service discovery data of the second service does not exist in the cache of the first service, and notify the service call interception module to obtain the service discovery data of the second service.
  • FIG6 is a schematic diagram showing the processing process of the service calling method according to an embodiment of the present application.
  • the historical calling information of service A is obtained by calling the link tracking service, specifically: service A has called service B; after service A is started, the service discovery data of service B is obtained from the control plane of the service grid, and the service discovery data of service B is stored in the cache of service A to realize the preloading of service discovery data.
  • the service discovery data of service B can also be subscribed to the control plane of the service grid to obtain the update of the service discovery data of service B in a timely manner.
  • service A when service A calls service B, it can be determined whether there is service discovery data of service B in the cache of service A. After determination, if there is service discovery data of service B in the cache of service A, service A can directly call service B according to the service discovery data of service B in the cache.
  • service A when service A calls service C, it can be determined whether there is service discovery data of service C in the cache of service A. After determination, if there is no service discovery data of service C in the cache of service A, in this case, the call request issued by service A can be intercepted, and then the service discovery data of service C can be obtained from the control plane of the service grid, and the service discovery data of service C can be stored in the cache of service A to achieve lazy loading of service discovery data. After that, service A can call service C based on the service discovery data of service C in the cache of service A. In addition, while obtaining the service discovery data of service C from the control plane of the service grid, the service discovery data of service C can also be subscribed to the control plane of the service grid. According to, in order to obtain the update of service discovery data of service C in time.
  • the link information of service A can be recorded by calling the link tracking service.
  • the link tracking service can also be periodically called to periodically update the service discovery data in the cache of service A.
  • the service calling method of the embodiment of the present application is implemented in a plug-in manner based on Java Agent and mounted in the host service (i.e., the first service). It replaces the sidecar to interact with the control plane of the service grid for data, avoids frequent cross-process communications, and effectively improves the service performance in high concurrency and large-scale scenarios.
  • the service calling method of the embodiment of the present application realizes the preloading of service discovery data in the service startup phase and the lazy loading in the service operation phase. Without affecting the business of the host service itself, it realizes the minimized service visibility configuration and significantly reduces resource consumption.
  • the service calling method of the embodiment of the present application can use dependency analysis and subscription push mechanism to replace sidecar to realize communication and data interaction with the control plane of the service grid, and can use link tracking service to preload service discovery data in the startup phase and lazy load service discovery data in the service operation phase.
  • Fig. 7 shows a block diagram of a service invocation device according to an embodiment of the present application.
  • the service invocation device is applied to a first service, which is a microservice running in a service grid.
  • the service calling device includes:
  • a determination module 710 is configured to determine whether service discovery data of the second service exists in a cache of the first service when the first service calls the second service;
  • An interception module 720 configured to intercept a call request issued by the first service when the service discovery data of the second service does not exist in the cache;
  • a first acquisition module 730 configured to acquire service discovery data of the second service from a control plane of the service grid, and store the service discovery data of the second service in the cache;
  • the first calling module 740 executes the calling of the second service by the first service based on the service discovery data of the second service.
  • the device further includes: a second calling module, which calls the second service according to the service discovery data of the second service in the cache when the service discovery data of the second service exists in the cache.
  • the service discovery data includes the IP address and load information of the service instance
  • the second calling module includes: a target instance selection submodule, used to select the target instance from the instances of the second service according to a preset load balancing rule and the load information of the instance of the second service; a target instance calling submodule, used to call the target instance according to the IP address of the target instance.
  • the device further includes: a link information recording module, configured to record link information of the first service through a link tracking service in the service grid during operation of the first service.
  • the device also includes: a historical call information acquisition module, used to acquire historical call information of the first service by calling the link tracking service during the restart of the first service, the historical call information including identifiers of services called by the first service as determined by the link tracking service based on the historical link information; a second acquisition module, used to acquire service discovery data of each service included in the historical call information from the control plane of the service grid after the first service is started; and a storage module, used to store the acquired service discovery data in the cache.
  • a historical call information acquisition module used to acquire historical call information of the first service by calling the link tracking service during the restart of the first service, the historical call information including identifiers of services called by the first service as determined by the link tracking service based on the historical link information
  • a second acquisition module used to acquire service discovery data of each service included in the historical call information from the control plane of the service grid after the first service is started
  • a storage module used to store the acquired service discovery data in the cache.
  • the device further includes: an updating module, configured to periodically update the service discovery data in the cache by periodically calling the link tracking service during the operation of the first service.
  • the device also includes: a subscription module, used to subscribe to the service discovery data in the cache to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated.
  • a subscription module used to subscribe to the service discovery data in the cache to the control plane of the service grid, so that the control plane of the service grid pushes the updated service discovery data to the first service when the service discovery data is updated.
  • the device is implemented in a plug-in manner based on a Java agent, and the plug-in is mounted to the first service.
  • An embodiment of the present application provides a computing device cluster, including at least one computing device, each computing device including a processor and a memory; the processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster executes the above method.
  • An embodiment of the present application provides a computer program product including instructions.
  • the instructions are executed by a computing device cluster, the computing device cluster executes the above method.
  • An embodiment of the present application provides a computer-readable storage medium, including computer program instructions.
  • the computer program instructions When the computer program instructions are executed by a computing device cluster, the computing device cluster executes the above method.
  • a computer readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device.
  • a computer readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Computer readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing.
  • RAM random access memory
  • ROM read-only memory
  • EPROM or flash memory erasable programmable read-only memory
  • SRAM static random access memory
  • CD-ROM compact disc read-only memory
  • DVD digital versatile disc
  • memory stick a floppy disk
  • mechanical encoding device such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing.
  • the computer-readable program instructions or codes described herein can be downloaded from a computer-readable storage medium to each computing/processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and/or a wireless network.
  • the network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and/or edge servers.
  • the network adapter card or network interface in each computing/processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing/processing device.
  • the computer program instructions for performing the operations of the present application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, etc., and conventional procedural programming languages such as "C" language or similar programming languages.
  • the computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server.
  • the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet).
  • LAN local area network
  • WAN wide area network
  • an Internet service provider to connect via the Internet.
  • the state information of the computer-readable program instructions is used to personalize the electronic circuit, for example, A programmed logic circuit, a Field-Programmable Gate Array (FPGA) or a Programmable Logic Array (PLA), the electronic circuit can execute computer-readable program instructions to implement various aspects of the present application.
  • FPGA Field-Programmable Gate Array
  • PDA Programmable Logic Array
  • These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions/actions specified in one or more boxes in the flowchart and/or block diagram is generated.
  • These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and/or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions/actions specified in one or more boxes in the flowchart and/or block diagram.
  • Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions/actions specified in one or more boxes in the flowchart and/or block diagram.
  • each square frame in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification.
  • the functions marked in the square frame can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square frames can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the functions involved.
  • each box in the block diagram and/or flowchart, and the combination of boxes in the block diagram and/or flowchart can be implemented by hardware (such as circuits or ASICs (Application Specific Integrated Circuit)) that performs the corresponding function or action, or can be implemented by a combination of hardware and software, such as firmware.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请涉及一种服务调用方法及装置，所述方法应用于第一服务，所述第一服务为运行在服务网格中的微服务，所述方法包括：在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。本申请的实施例能够降低服务调用时的资源消耗，以及提高服务调用时的处理效率，从而提高服务性能。

Description

服务调用方法及装置
本申请要求于2022年09月27日提交中国专利局、申请号为202211184674.2、申请名称为“服务调用方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及云计算技术领域，尤其涉及一种服务调用方法及装置。
背景技术
微服务架构是一种云原生架构。基于微服务架构，应用可由许多松散耦合且可独立部署的较小组件(通常称为微服务)组成，每个组件都有各自的责任领域。在处理用户请求时，基于微服务的应用会调用多个内部微服务来共同生成其响应。
在微服务架构中，进行服务调用时，服务发现数据的加载通常基于以挎斗(sidecar)方式实现的网络代理。在该方式下，sidecar需要单独占用一个容器的进程，资源消耗较大，而且由于sidecar与其挂载的服务属于不同的容器，两者间的跨进程通信会带来时延，影响服务调用时的处理效率，从而影响服务性能。
发明内容
有鉴于此，提出了一种服务调用方法及装置。
第一方面，本申请的实施例提供了一种服务调用方法，应用于第一服务，所述第一服务为运行在服务网格中的微服务，所述方法包括：在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
本申请的实施例的服务调用方法，在第一服务调用第二服务时，首先判断第一服务的缓存中是否存在第二服务的服务发现数据，在缓存中不存在第二服务的服务发现数据时，拦截第一服务发出的调用请求，然后从服务网格的控制面获取第二服务的服务发现数据，并将第二服务的服务发现数据存储到第一服务的缓存中，之后基于第二服务的服务发现数据，执行第一服务对第二服务的调用。
通过这种方式，能够在第一服务调用第二服务时，在第一服务的缓存中不存在第二服务的服务发现数据时，拦截调用请求并直接从服务网格的控制面获取第二服务的服务发现数据，不仅能够实现服务发现数据的懒加载，降低服务发现数据加载时的资源消耗，而且能够在加载服务发现数据时，无需使用sidecar作为网络代理，避免了跨进程通信，从而能够降低服务调用时的资源消耗，以及提高服务调用时的处理效率，进而提高服务性能。
根据第一方面，在所述服务调用方法的第一种可能的实现方式中，所述方法还包括：在所述缓存中存在所述第二服务的服务发现数据时，根据所述缓存中的所述第二服务的服务发 现数据，调用所述第二服务。
本实施例中，在第一服务的缓存中存在第二服务的服务发现数据时，第一服务能够根据第一服务的缓存中的第二服务的服务发现数据，直接调用第二服务，从而能够提高服务调用时的处理效率。
根据第一方面的第一种可能的实现方式，在所述服务调用方法的第二种可能的实现方式中，所述服务发现数据包括服务的实例的IP地址及负载信息，所述根据所述缓存中的所述第二服务的服务发现数据，调用所述第二服务，包括：根据预设的负载均衡规则及所述第二服务的实例的负载信息，从所述第二服务的实例中，选取目标实例；根据所述目标实例的IP地址，调用所述目标实例。
在本实施例中，在根据缓存中的第二服务的服务发现数据调用第二服务时，可首先根据预设的负载均衡规则及第二服务的实例的负载信息，从第二服务的实例中，选取目标实例；然后根据目标实例的IP地址，调用目标实例，从而实现第一服务对第二服务的调用。通过这种方式，能够提高服务调用时的处理效率。
根据第一方面或第一方面的第一种可能的实现方式或第一方面的第二种可能的实现方式，在所述服务调用方法的第三种可能的实现方式中，所述方法还包括：在所述第一服务运行过程中，通过所述服务网格中的链路追踪服务，记录所述第一服务的链路信息。
在本实施例中，在第一服务运行过程中，通过服务网格中的链路追踪服务，记录第一服务的链路信息，以便之后在第一服务重启时，根据第一服务的历史链路信息，对服务发现数据进行预加载。
根据第一方面的第三种可能的实现方式，在所述服务调用方法的第四种可能的实现方式中，所述方法还包括：在所述第一服务重新启动过程中，通过调用所述链路追踪服务，获取所述第一服务的历史调用信息，所述历史调用信息包括所述链路追踪服务根据历史链路信息确定的所述第一服务调用过的服务的标识；在所述第一服务启动后，从所述服务网格的控制面，获取所述历史调用信息包括的各个服务的服务发现数据；将获取的服务发现数据存储到所述缓存中。
在本实施例中，能够在第一服务重新启动过程中，通过调用链路追踪服务，获取第一服务的历史调用信息，并在第一服务启动后，从服务网格的控制面，获取历史调用信息包括的各个服务的服务发现数据，以及将获取的服务发现数据存储到缓存中。通过这种方式，能够在第一服务启动时，实现服务发现数据的预加载，从而能够减少首次获取服务发现数据的时间成本，提高服务调用时的处理效率，进而提高服务性能。
根据第一方面的第三种可能的实现方式，在所述服务调用方法的第五种可能的实现方式中，所述方法还包括：在所述第一服务运行过程中，通过周期调用所述链路追踪服务，对所述缓存中的服务发现数据进行周期更新。
本实施例中，能够在第一服务运行过程中，通过周期调用链路追踪服务，对缓存中的服务发现数据进行周期更新，从而能够实现服务发现数据的动态懒加载。
根据第一方面或者第一方面的第一种可能的实现方式至第一方面的第五种可能的实现方式中的任意一种，在所述服务调用方法的第六种可能的实现方式中，所述方法还包括：向所述服务网格的控制面订阅所述缓存中的服务发现数据，以使所述服务网格的控制面在所述服务发现数据更新时，向所述第一服务推送更新后的服务发现数据。
在本实施例中，能够向服务网格的控制面订阅第一服务的缓存中的服务发现数据，以使服务网格的控制面在服务发现数据更新时，向第一服务推送更新后的服务发现数据，从而能够使得第一服务的缓存中的服务发现数据保持动态更新，提高第一服务的缓存中的服务发现数据的准确性。
根据第一方面或者第一方面的第一种可能的实现方式至第一方面的第六种可能的实现方式中的任意一种，在所述服务调用方法的第七种可能的实现方式中，所述方法基于Java代理以插件方式实现，所述插件挂载到所述第一服务。
在本实施例中，基于Java代理以插件方式实现，并将插件挂载到第一服务，不仅能够在无侵入的前提下将实现本申请实施例的插件动态加载到第一服务，而且插件与第一服务同属一个进程，减少了跨进程通信，从而能够在服务调用时，使用Java插件替换现有的服务网格中的sidecar，使得服务的资源消耗更少，性能更优。
第二方面，本申请的实施例提供了一种服务调用装置，应用于第一服务，所述第一服务为运行在服务网格中的微服务，所述装置包括：判断模块，用于在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；拦截模块，用于在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；第一获取模块，用于从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；第一调用模块，基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
本申请的实施例的服务调用装置，在第一服务调用第二服务时，首先判断第一服务的缓存中是否存在第二服务的服务发现数据，在缓存中不存在第二服务的服务发现数据时，拦截第一服务发出的调用请求，然后从服务网格的控制面获取第二服务的服务发现数据，并将第二服务的服务发现数据存储到第一服务的缓存中，之后基于第二服务的服务发现数据，执行第一服务对第二服务的调用。
通过这种方式，能够在第一服务调用第二服务时，在第一服务的缓存中不存在第二服务的服务发现数据时，拦截调用请求并直接从服务网格的控制面获取第二服务的服务发现数据，不仅能够实现服务发现数据的懒加载，降低服务发现数据加载时的资源消耗，而且能够在加载服务发现数据时，无需使用sidecar作为网络代理，避免了跨进程通信，从而能够降低服务调用时的资源消耗，以及提高服务调用时的处理效率，进而提高服务性能。
根据第二方面，在所述服务调用装置的第一种可能的实现方式中，所述装置还包括：第二调用模块，在所述缓存中存在所述第二服务的服务发现数据时，根据所述缓存中的所述第二服务的服务发现数据，调用所述第二服务。
本实施例中，在第一服务的缓存中存在第二服务的服务发现数据时，第一服务能够根据第一服务的缓存中的第二服务的服务发现数据，直接调用第二服务，从而能够提高服务调用时的处理效率。
根据第二方面的第一种可能的实现方式，在所述服务调用装置的第二种可能的实现方式中，所述服务发现数据包括服务的实例的IP地址及负载信息，所述第二调用模块，包括：目标实例选取子模块，用于根据预设的负载均衡规则及所述第二服务的实例的负载信息，从所述第二服务的实例中，选取目标实例；目标实例调用子模块，用于根据所述目标实例的IP地址，调用所述目标实例。
在本实施例中，在根据缓存中的第二服务的服务发现数据调用第二服务时，可首先根据预设的负载均衡规则及第二服务的实例的负载信息，从第二服务的实例中，选取目标实例；然后根据目标实例的IP地址，调用目标实例，从而实现第一服务对第二服务的调用。通过这种方式，能够提高服务调用时的处理效率。
根据第二方面或第二方面的第一种可能的实现方式或第二方面的第二种可能的实现方式，在所述服务调用装置的第三种可能的实现方式中，所述装置还包括：链路信息记录模块，用于在所述第一服务运行过程中，通过所述服务网格中的链路追踪服务，记录所述第一服务的链路信息。
在本实施例中，在第一服务运行过程中，通过服务网格中的链路追踪服务，记录第一服务的链路信息，以便之后在第一服务重启时，根据第一服务的历史链路信息，对服务发现数据进行预加载。
根据第二方面的第三种可能的实现方式，在所述服务调用装置的第四种可能的实现方式中，所述装置还包括：历史调用信息获取模块，用于在所述第一服务重新启动过程中，通过调用所述链路追踪服务，获取所述第一服务的历史调用信息，所述历史调用信息包括所述链路追踪服务根据历史链路信息确定的所述第一服务调用过的服务的标识；第二获取模块，用于在所述第一服务启动后，从所述服务网格的控制面，获取所述历史调用信息包括的各个服务的服务发现数据；存储模块，用于将获取的服务发现数据存储到所述缓存中。
在本实施例中，能够在第一服务重新启动过程中，通过调用链路追踪服务，获取第一服务的历史调用信息，并在第一服务启动后，从服务网格的控制面，获取历史调用信息包括的各个服务的服务发现数据，以及将获取的服务发现数据存储到缓存中。通过这种方式，能够在第一服务启动时，实现服务发现数据的预加载，从而能够减少首次获取服务发现数据的时间成本，提高服务调用时的处理效率，进而提高服务性能。
根据第二方面的第三种可能的实现方式，在所述服务调用装置的第五种可能的实现方式中，所述装置还包括：更新模块，用于在所述第一服务运行过程中，通过周期调用所述链路追踪服务，对所述缓存中的服务发现数据进行周期更新。
本实施例中，能够在第一服务运行过程中，通过周期调用链路追踪服务，对缓存中的服务发现数据进行周期更新，从而能够实现服务发现数据的动态懒加载。
根据第二方面或第二方面的第一种可能的实现方式至第二方面的第五种可能的实现方式中的任意一种，在所述服务调用装置的第六种可能的实现方式中，所述装置还包括：订阅模块，用于向所述服务网格的控制面订阅所述缓存中的服务发现数据，以使所述服务网格的控制面在所述服务发现数据更新时，向所述第一服务推送更新后的服务发现数据。
在本实施例中，能够向服务网格的控制面订阅第一服务的缓存中的服务发现数据，以使服务网格的控制面在服务发现数据更新时，向第一服务推送更新后的服务发现数据，从而能够使得第一服务的缓存中的服务发现数据保持动态更新，提高第一服务的缓存中的服务发现数据的准确性。
根据第二方面或第二方面的第一种可能的实现方式至第二方面的第六种可能的实现方式中的任意一种，在所述服务调用装置的第七种可能的实现方式中，所述装置基于Java代理以插件方式实现，所述插件挂载到所述第一服务。
在本实施例中，基于Java代理以插件方式实现，并将插件挂载到第一服务，不仅能够在 无侵入的前提下将实现本申请实施例的插件动态加载到第一服务，而且插件与第一服务同属一个进程，减少了跨进程通信，从而能够在服务调用时，使用Java插件替换现有的服务网格中的sidecar，使得服务的资源消耗更少，性能更优。
第三方面，本申请的实施例提供了一种计算设备集群，包括至少一个计算设备，每个计算设备包括处理器和存储器；所述至少一个计算设备的处理器用于执行所述至少一个计算设备的存储器中存储的指令，以使得所述计算设备集群执行上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的服务调用方法。
第四方面，本申请的实施例提供了一种包含指令的计算机程序产品，当所述指令被计算设备集群运行时，使得所述计算设备集群执行上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的服务调用方法。
第五方面，本申请的实施例提供了一种计算机可读存储介质，包括计算机程序指令，当所述计算机程序指令由计算设备集群执行时，所述计算设备集群执行上述第一方面或者第一方面的多种可能的实现方式中的一种或几种的服务调用方法。
本申请的这些和其他方面在以下(多个)实施例的描述中会更加简明易懂。
附图说明
包含在说明书中并且构成说明书的一部分的附图与说明书一起示出了本申请的示例性实施例、特征和方面，并且用于解释本申请的原理。
图1示出根据本申请一实施例的服务调用方法的应用场景的示意图。
图2示出根据本申请一实施例的服务调用方法的流程图。
图3示出根据本申请一实施例的服务调用方法的处理过程的示意图。
图4示出根据本申请一实施例的服务调用方法的处理过程的示意图。
图5示出根据本申请一实施例的服务调用方法的处理过程的示意图。
图6示出根据本申请一实施例的服务调用方法的处理过程的示意图。
图7示出根据本申请一实施例的服务调用装置的框图。
具体实施方式
以下将参考附图详细说明本申请的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面，但是除非特别指出，不必按比例绘制附图。
在这里专用的词“示例性”意为“用作例子、实施例或说明性”。这里作为“示例性”所说明的任何实施例不必解释为优于或好于其它实施例。
另外，为了更好的说明本申请，在下文的具体实施方式中给出了众多的具体细节。本领域技术人员应当理解，没有某些具体细节，本申请同样可以实施。在一些实例中，对于本领域技术人员熟知的方法、手段、元件和电路未作详细描述，以便于凸显本申请的主旨。
为了便于理解，首先对本申请的实施例涉及的相关术语进行说明：
服务治理：对微服务进行管理和治理，例如服务注册、服务发现、负载均衡、应用路由等。
服务发现：是指微服务架构中作为客户端的服务自动发现服务地址列表的能力。服务发 现的过程可以看作是作为客户端的服务自动加载服务发现数据的过程。借助于自动化的服务发现，微服务之间可以在无需感知对端部署位置及IP地址的情况下实现通信。
服务网格：即service mesh，是微服务架构中处理服务之间的通信的基础设施层，其支持云原生应用的网络请求在复杂的拓扑环境中可靠地传递。服务网格与应用部署在一起，但对应用透明。服务网格通常包括数据面(data plane)和控制面(control plane)。
服务可见性：是指当前服务可从服务网格的控制面获得其调用的服务的服务发现数据。
下面以开源服务网格Istio为例，对服务网格进行示例性地说明。
开源服务网格Istio包括数据面和控制面。Istio的数据面是指Envoy代理(Envoy proxy)这一层。Istio将Envoy代理作为一个sidecar容器注入到服务容器旁边，然后Envoy代理拦截该服务的所有入站和出站流量。Istio的控制面是指Istiod组件这一层，其提供服务发现(discovery)、配置(configuration)及证书管理(certificates)等功能。
在Istio中，控制面基于服务发现(x discovery service，xDS)协议向数据面传递服务信息和治理规则；数据面将xDS协议作为其应用编程接口(application programming interface，API)，并基于xDS协议与控制面进行通信。
其中，xDS协议中的x表示xDS协议不是指具体的某个协议，而是一组基于不同数据源的服务发现协议的总称，包括监听器发现服务(listener discovery service，LDS)、集群发现服务(cluster discovery service，CDS)、集群成员发现服务(endpoint discovery service，EDS)、路由发现服务(route discovery service，RDS)、聚合发现服务(aggregated discovery service，ADS)、健康度发现服务(health discovery service，HDS)、密钥发现服务(secret discovery service，SDS)、指标发现服务(metric service，MS)、限流发现服务(rate limit service，RLS)等。
在微服务架构中，进行服务调用时，服务发现数据的加载通常基于以sidecar方式实现的网络代理。例如，开源服务网格Istio中，每个服务容器旁边均有用于实现Envoy代理的sidecar容器，进行服务调用时，服务发现数据的加载均通过sidecar容器。然而，sidecar需要单独占用一个容器的进程，资源消耗较大，而且sidecar与其挂载的服务属于不同的容器，两者间的跨进程通信会带来时延，影响服务调用时的处理效率，从而影响服务性能。
此外，基于sidecar(即基于以sidecar方式实现的网络代理)的服务发现数据的加载，还存在不适用于大规模场景、部署及运维复杂等问题。在一个示例中，Istio下发服务发现数据采用全量下发策略，即服务网格中的所有sidecar进程都持有整个网格内所有的服务发现数据，但是，大部分服务发现数据对工作负载来说是无用的，且在大规模场景下，服务发现数据的全量下发会导致严重的资源消耗，即服务发现数据的全量下发策略并不适用于大规模场景。
在另一个示例中，服务发现数据的加载采用按需加载策略。例如，提前预配置服务间的依赖关系，在服务调用过程中，根据预配置的服务间的依赖关系，加载服务发现数据，但是，该方式通常需要手动配置，不适用于大规模场景且维护困难。再例如，可在Istio中部署专门的中间网关服务及控制器服务，来实现服务发现数据的按需加载，但是，该方式需要在服务网格中启动额外的中间网关服务及控制器服务，不仅会带来资源消耗，而且使得服务的部署及运维变得更为复杂。
为了解决上述技术问题，本申请的实施例提供了一种服务调用方法，应用于第一服务，所述第一服务为运行在服务网格中的微服务，所述方法包括：在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；在所述缓存中不存 在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
本申请实施例的服务调用方法，在第一服务调用第二服务时，首先判断第一服务的缓存中是否存在第二服务的服务发现数据，在缓存中不存在第二服务的服务发现数据时，拦截第一服务发出的调用请求，然后从服务网格的控制面获取第二服务的服务发现数据，并将第二服务的服务发现数据存储到第一服务的缓存中，之后基于第二服务的服务发现数据，执行第一服务对第二服务的调用。
通过这种方式，能够在第一服务调用第二服务时，在第一服务的缓存中不存在第二服务的服务发现数据时，拦截调用请求并直接从服务网格的控制面获取第二服务的服务发现数据，不仅能够实现服务发现数据的懒加载，降低服务发现数据加载时的资源消耗，而且能够在加载服务发现数据时，无需使用sidecar作为网络代理，避免了跨进程通信，从而能够降低服务调用时的资源消耗，以及提高服务调用时的处理效率，进而提高服务性能。
其中，懒加载即延迟加载，是一种非必要资源不预先加载且仅在需要时加载它们的策略。懒加载是一种按需加载，能够降低资源占用。
本申请实施例的服务调用方法可应用于微服务架构中的服务治理场景。在实际应用场景的服务网格中，通常会部署十万以上甚至上百万个服务的实例，在这种大规模场景下，通过本申请实施例的服务调用方法进行服务调用，能够在服务调用时，实现无需基于sidecar的服务发现数据的懒加载，不仅能够降低服务调用时的资源消耗，而且能够提高服务调用时的处理效率，从而提升服务性能。
图1示出根据本申请一实施例的服务调用方法的应用场景的示意图。如图1所示，终端设备130通过网络(例如有线网络、无线网络等)与云计算中心110连接，云计算中心110包括服务器集群120，服务器集群120基于微服务架构，为终端设备130上的应用提供服务。其中，终端设备130可以是智能手机、上网本、平板电脑、笔记本电脑、可穿戴电子设备(如智能手环、智能手表等)、TV、虚拟现实设备、音响、电子墨水，等等。本申请对终端设备的具体类型不作限制。
用户在终端设备130上使用某一应用时，服务器集群120会调用多个内部微服务来共同生成该应用的响应。其中，服务器集群120在调用多个内部微服务时，会存在微服务之间的调用，例如一个微服务对另一个微服务的调用，该情况下，可使用本申请实施例的服务调用方法来实现一个微服务对另一个微服务的调用。
图2示出根据本申请一实施例的服务调用方法的流程图。本申请实施例的服务调用方法可应用于第一服务，第一服务为运行在服务网格中的微服务。例如，服务网格运行在图1所示的服务器集群120上，第一服务为运行在该服务网格中的作为调用发起方的微服务。
如图2所示，本申请实施例的服务调用方法包括：
步骤S210，在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据。
其中，第一服务与第二服务均为运行在服务网格中的微服务。第一服务及第二服务均有对应的缓存(例如cache)。可将第一服务加载的服务发现数据存储在第一服务的缓存中，将第二服务加载的服务发现数据存储在第二服务的缓存中。
在第一服务调用第二服务的场景中，可将第二服务看作服务提供方，将第一服务看作服务消费方或客户方。在第一服务调用第二服务时，可首先判断第一服务的缓存中是否存在第二服务的服务发现数据。例如，假设第一服务通过第二服务的标识(例如可以唯一标识第二服务的ID、名称等)来调用第二服务，判断第一服务的缓存中是否存在第二服务的服务发现数据时，可根据第二服务的标识，通过查找、比对等方式，确定第一服务的缓存中是否存在第二服务的服务发现数据。
在一种可能的实现方式中，可在第一服务的缓存中设置第一服务的服务可见性列表。第一服务的服务可见性列表可用于存储第一服务的缓存中存储的服务发现数据所属的服务的标识。例如，假设第一服务S0的缓存中存储了服务S1、服务S2、服务S3、服务S4及服务S5的服务发现数据，则可将服务S1、服务S2、服务S3、服务S4及服务S5的标识，存储到第一服务S0的服务可见性列表中。
在判断第一服务的缓存中是否存在第二服务的服务发现数据时，可判断第一服务的服务可见性列表中是否存在第二服务的标识。如果第一服务的服务可见性列表中存在第二服务的标识，则可认为第一服务的缓存中存在第二服务的服务发现数据。如果第一服务的服务可见性列表中不存在第二服务的标识，则可认为第一服务的缓存中不存在第二服务的服务发现数据。
需要说明的是，本领域技术人员还可通过其他方式来判断第一服务的缓存中是否存在第二服务的服务发现数据，本申请对此不作限制。
步骤S220，在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求。
在第一服务的缓存中不存在第二服务的服务发现数据时，可拦截第一服务发出的调用请求，以便从服务网格的控制面获取第二服务的服务发现数据。
步骤S230，从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中。
拦截第一服务发出的调用请求后，可根据第二服务的标识，从服务网格的控制面获取第二服务的服务发现数据。第二服务的服务发现数据可包括第二服务的标识(例如可以唯一标识第二服务的ID、名称等)、第二服务的实例的数量以及各个实例的标识(例如可以唯一标识各个实例的ID、名称等)、IP地址、负载信息等。需要说明的是，第二服务的服务发现数据还可包括其他信息，本领域技术人员可根据实际情况对服务发现数据包括的具体内容进行设置，本申请对此不作限制。
从服务网格的控制面获取到第二服务的服务发现数据后，可将获取的第二服务的服务发现数据存储到第一服务的缓存中。
在一种可能的实现方式中，在第一服务的缓存中设置第一服务的服务可见性列表的情况下，可将第二服务的标识，存储到第一服务的服务可见性列表中。
在一种可能的实现方式中，在从服务网格的控制面获取第二服务的服务发现数据的同时，还可向服务网格的控制面订阅第二服务的服务发现数据。在向服务网格的控制面订阅第二服务的服务发现数据后，在服务网格的控制面发现第二服务的服务发现数据更新时，服务网格的控制面会向第一服务推送第二服务的更新后的服务发现数据。第一服务在接收到服务网格的控制面推送的第二服务的更新后的服务发现数据(即新的服务发现数据)时，使用第二服 务的更新后的服务发现数据(即新的服务发现数据)，替换缓存中的第二服务的服务发现数据(即旧的服务发现数据)。通过这种方式，能够使得第一服务的缓存中的第二服务的服务发现数据保持动态更新，从而提高缓存中的第二服务的服务发现数据的准确性。
步骤S240，基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
在将第二服务的服务发现数据存储到第一服务的缓存中后，可根据第一服务的缓存中的第二服务的服务发现数据，执行第一服务对第二服务的调用。具体可例如，可从第一服务的缓存中获取第二服务的服务发现数据，并从该服务发现数据中，获取第二服务的实例的负载信息；然后根据预设的负载均衡规则(例如负载最低优先等)及第二服务的实例的负载信息，从第二服务的实例中，选取目标实例；之后可从第二服务的服务发现数据中，获取目标实例的IP地址，并根据目标实例的IP地址，调用目标实例，从而实现第一服务对第二服务的调用。
需要说明的是，第二服务的实例可以为多个，本申请对第二服务的实例的具体数量不作限制。
在一种可能的实现方式中，第一服务调用第二服务时，在第一服务的缓存中存在第二服务的服务发现数据时，可根据第一服务的缓存中的第二服务的服务发现数据，直接调用第二服务，从而能够提高服务调用时的处理效率。
在一种可能的实现方式中，根据第一服务的缓存中的第二服务的服务发现数据，调用第二服务时，可从第一服务的缓存中获取第二服务的服务发现数据，并从该服务发现数据中，获取第二服务的实例的负载信息；然后根据预设的负载均衡规则(例如负载最低优先等)及第二服务的实例的负载信息，从第二服务的实例中，选取目标实例；之后可从第二服务的服务发现数据中，获取目标实例的IP地址，并根据目标实例的IP地址，调用目标实例，从而实现第一服务对第二服务的调用。
图3示出根据本申请一实施例的服务调用方法的处理过程的示意图。如图3所示，第一服务310的缓存311中存储了第二服务320(service320)的服务发现数据，示例如下：
xds{
cds[service320]
eds[service320]
……
}
在第一服务310调用第二服务320时，首先判断缓存311中是否存在第二服务320的服务发现数据。经判断，缓存311中存在第二服务320的服务发现数据，该情况下，第一服务310可从缓存311中获取第二服务320的服务发现数据，并根据获取的第二服务320的服务发现数据，直接调用第二服务320。
图4示出根据本申请一实施例的服务调用方法的处理过程的示意图。如图4所示，第一服务410的缓存411中存储了服务F(serviceF)的服务发现数据，示例如下：
xds{
cds[serviceF]
eds[serviceF]
……
}
在第一服务410运行过程中，由于热更新等行为，第一服务410对第二服务420产生了依赖关系。在第一服务410调用第二服务420时，首先判断缓存411中是否存在第二服务420(service420)的服务发现数据。经判断，缓存411中不存在第二服务420的服务发现数据，该情况下，可拦截第一服务410发出的调用请求，然后从服务网格的控制面430中获取第二服务420的服务发现数据，并将获取的第二服务420的服务发现数据存储到缓存411中，在该过程中，还可向服务网格的控制面430订阅第二服务420的服务发现数据，以获取第二服务420的服务发现数据的更新。
将第二服务420的服务发现数据存储到缓存411中后，缓存411中的服务发现数据可示例如下：
xds{
cds[serviceF]
eds[serviceF]
……
cds[service420]
eds[service420]
……
}
之后，可继续执行第一服务410对第二服务420的调用，具体为：第一服务410从缓存411中获取第二服务420的服务发现数据，并根据获取的第二服务420的服务发现数据，调用第二服务420。
在一种可能的实现方式中，在第一服务运行过程中，还可通过服务网格中的链路追踪服务，记录第一服务的链路信息。其中，链路信息可包括服务依赖信息(例如第一服务调用的服务的信息)。在一些示例中，链路信息还可包括调用链路、调用请求量等其他信息，本申请对此不作限制。链路追踪服务可提供调用链路还原、调用请求量统计、链路拓扑、依赖分析等功能。
在第一服务运行过程中，通过服务网格中的链路追踪服务，追踪并记录第一服务的链路信息，以便之后在第一服务重启时，根据第一服务的历史链路信息，对服务的服务发现数据进行预加载。
在一种可能的实现方式中，在第一服务重新启动过程中，可通过调用服务网格中的链路追踪服务，获取第一服务的历史调用信息。其中，历史调用信息可包括链路追踪服务根据第一服务的历史链路信息确定的第一服务调用过的服务的标识。在第一服务启动后，可从服务网格的控制面，获取历史调用信息包括的第一服务调用过的各个服务的服务发现数据，并将获取的服务发现数据存储到第一服务的缓存中。
例如，假设在第一服务S6启动过程中，可通过调用服务网格中的链路追踪服务，获得第一服务S6的历史调用信息，该历史调用信息包括第一服务S6在当前时刻之前的预设时长内调用过的服务的标识：服务S7、服务S8及服务S9的标识；在第一服务S6启动后，可分别根据服务S7、服务S8及服务S9的标识，从服务网格的控制面获取服务S7、服务S8及服务 S9的服务发现数据，并将获取的服务S7、服务S8及服务S9的服务发现数据存储到第一服务S6的缓存中。
通过这种方式，能够在第一服务重新启动时，通过链路追踪服务获取第一服务的历史调用信息，并根据第一服务的历史调用信息，实现服务发现数据的预加载，从而减少首次获取服务发现数据的时间成本，提高服务调用时的处理效率，进而提高服务性能。
在一种可能的实现方式中，在第一服务的缓存中设置第一服务的服务可见性列表的情况下，可将历史调用信息包括的第一服务调用过的各个服务的标识，存储到第一服务的服务可见性列表中。
在一种可能的实现方式中，在从服务网格的控制面获取历史调用信息包括的第一服务调用过的各个服务(这里称为第三服务)的服务发现数据的同时，还可向服务网格的控制面订阅各个第三服务的服务发现数据。
在向服务网格的控制面订阅各个第三服务的服务发现数据后，对于任一第三服务，在服务网格的控制面发现所述第三服务的服务发现数据更新时，服务网格的控制面会向第一服务推送所述第三服务的更新后的服务发现数据。第一服务在接收到服务网格的控制面推送的所述第三服务的更新后的服务发现数据(即新的服务发现数据)时，使用所述第三服务的更新后的服务发现数据(即新的服务发现数据)，替换缓存中的所述第三服务的服务发现数据(即旧的服务发现数据)。通过这种方式，能够使得第一服务的缓存中的第三服务的服务发现数据保持动态更新，从而提高缓存中的第三服务的服务发现数据的准确性。
图5示出根据本申请一实施例的服务调用方法的处理过程的示意图。如图5所示，在第一服务510重新启动过程中，可通过调用链路追踪服务520，获取第一服务510的历史调用信息。在第一服务510启动后，可从服务网格的控制面530，获取第一服务510的历史调用信息包括的各个服务的服务发现数据，并将获取的服务发现数据存储到缓存511中；在该过程中，还可向服务网格的控制面530订阅第一服务510的历史调用信息包括的各个服务的服务发现数据，以获取该服务发现数据的更新。
在一种可能的实现方式中，在第一服务运行过程中，还可通过周期调用服务网格中的链路追踪服务，对缓存中的服务发现数据进行周期更新。具体地，在第一服务运行过程中，可按照预设周期，调用服务网格中的链路追踪服务，获取第一服务在最近一个周期内的历史调用信息，然后根据最近一个周期内的历史调用信息，对第一服务的缓存中的服务发现数据进行更新。
例如，假设第一服务D1的缓存中的服务发现数据为服务D2、服务D3、服务D4的服务发现数据，在第一服务D1的运行过程中，按照预设周期，调用服务网格中的链路追踪服务，获取的第一服务D1在最近一个周期内的历史调用信息包括：服务D2、服务D3、服务D5、服务D6的标识，那么，可根据第一服务D1在最近一个周期内的历史调用信息，对第一服务D1的缓存中的服务发现数据进行更新，具体如下：
从第一服务D1的缓存中删除服务D4的服务发现数据；
从服务网格的控制面获取服务D5的服务发现数据，并将获取的服务D5的服务发现数据存储到第一服务D1的缓存中；
从服务网格的控制面获取服务D6的服务发现数据，并将获取的服务D6的服务发现数据存储到第一服务D1的缓存中。
通过这种方式，能够在第一服务运行过程中，通过周期调用链路追踪服务，对缓存中的服务发现数据进行周期更新，从而实现服务发现数据的动态懒加载。
在一种可能的实现方式中，在第一服务的缓存中设置第一服务的服务可见性列表的情况下，还可根据最近一个周期内的历史调用信息，对第一服务的服务可见性列表进行更新，以提高第一服务的服务可见性列表的准确性。
在一种可能的实现方式中，本申请实施例的服务调用方法可基于Java代理(Java Agent)以插件方式实现，该插件挂载到第一服务。启动时加载Java Agent是Java开发工具包(Java Development Kit，JDK)1.5之后引入的新特性，该特性为用户提供了在Java虚拟机(Java virtual machine，JVM)将字节码文件读入内存之后，JVM使用对应的字节流在Java堆中生成一个Class对象之前，用户可以对其字节码进行修改的能力，从而JVM也将会使用用户修改过之后的字节码进行Class对象的创建，达到动态增强程序逻辑的目的。
本申请实施例的服务调用方法，基于Java代理以插件方式实现，并将插件挂载到第一服务，不仅能够在无侵入的前提下将实现本申请实施例的插件动态加载到第一服务，而且插件与第一服务同属一个进程，减少了跨进程通信，从而能够在服务调用时，使用Java插件替换现有的服务网格中的sidecar，使得服务的资源消耗更少，性能更优。
在一种可能的实现方式中，在本申请实施例的服务调用方法基于Java代理(Java Agent)以插件方式实现时，插件可包括服务发现数据订阅模块及服务调用拦截模块。服务发现数据订阅模块主要用于与服务网格的控制面建立连接，获取并订阅服务发现数据。服务发现数据订阅模块还可用于实现负载均衡、流量控制、限流降级等服务治理功能。在第一服务重新启动过程中，服务发现数据订阅模块可调用服务网格中的链路追踪模块，实现服务发现数据的预加载。在第一服务运行过程中，服务发现数据订阅模块还可周期调用服务网格中的链路追踪服务，对缓存中的服务发现数据进行周期更新。服务调用拦截模块可在第一服务调用第二服务且第一服务的缓存中不存在第二服务的服务发现数据时，拦截第一服务的调用请求，并通知服务调用拦截模块获取第二服务的服务发现数据。
图6示出根据本申请一实施例的服务调用方法的处理过程的示意图。如图6所示，在服务A启动过程中，通过调用链路追踪服务，获取服务A的历史调用信息，具体为：服务A调用过服务B；在服务A启动后，从服务网格的控制面获取服务B的服务发现数据，并将服务B的服务发现数据存储到服务A的缓存中，实现服务发现数据的预加载。在从服务网格的控制面获取服务B的服务发现数据的同时，还可向服务网格的控制面订阅服务B的服务发现数据，以及时获得服务B的服务发现数据的更新。
在服务A运行过程中，在服务A调用服务B时，可判断服务A的缓存中是否存在服务B的服务发现数据。经判断，服务A的缓存中存在服务B的服务发现数据，则服务A可根据缓存中的服务B的服务发现数据，直接调用服务B。
在服务A运行过程中，在服务A调用服务C时，可判断服务A的缓存中是否存在服务C的服务发现数据。经判断，服务A的缓存中不存在服务C的服务发现数据，该情况下，可拦截服务A发出的调用请求，然后从服务网格的控制面获取服务C的服务发现数据，并将服务C的服务发现数据存储到服务A的缓存中，实现服务发现数据的懒加载。之后可根据服务A的缓存中的服务C的服务发现数据，执行服务A对服务C的调用。此外，在从服务网格的控制面获取服务C的服务发现数据的同时，还可向服务网格的控制面订阅服务C的服务发现数 据，以及时获得服务C的服务发现数据的更新。
在服务A运行过程中，可通过调用链路追踪服务，记录服务A的链路信息。在服务A运行过程中，还可通过周期调用链路追踪服务，对服务A的缓存中的服务发现数据进行周期更新。
本申请实施例的服务调用方法，基于Java Agent以插件方式实现，并挂载到宿主服务(即第一服务)中，替换sidecar与服务网格的控制面进行数据交互，避免了频繁的跨进程通信，有效提高了高并发、大规模等场景下的服务性能。
本申请实施例的服务调用方法，实现了服务发现数据在服务启动阶段的预加载以及在服务运行阶段的懒加载，在不影响宿主服务自身业务的前提下，实现了最小化的服务可见性配置，显著降低了资源消耗。
与相关技术相比，本申请实施例的服务调用方法，能够使用依赖分析和订阅推送机制替换sidecar来实现与服务网格的控制面之间的通信和数据交互，以及能够采用链路追踪服务来进行启动阶段的服务发现数据的预加载以及在服务运行阶段进行服务发现数据的懒加载。
图7示出本申请一实施例的服务调用装置的框图。所述服务调用装置应用于第一服务，所述第一服务为运行在服务网格中的微服务。
如图7所示，所述服务调用装置包括：
判断模块710，用于在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；
拦截模块720，用于在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；
第一获取模块730，用于从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；
第一调用模块740，基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
在一种可能的实现方式中，所述装置还包括：第二调用模块，在所述缓存中存在所述第二服务的服务发现数据时，根据所述缓存中的所述第二服务的服务发现数据，调用所述第二服务。
在一种可能的实现方式中，所述服务发现数据包括服务的实例的IP地址及负载信息，所述第二调用模块，包括：目标实例选取子模块，用于根据预设的负载均衡规则及所述第二服务的实例的负载信息，从所述第二服务的实例中，选取目标实例；目标实例调用子模块，用于根据所述目标实例的IP地址，调用所述目标实例。
在一种可能的实现方式中，所述装置还包括：链路信息记录模块，用于在所述第一服务运行过程中，通过所述服务网格中的链路追踪服务，记录所述第一服务的链路信息。
在一种可能的实现方式中，所述装置还包括：历史调用信息获取模块，用于在所述第一服务重新启动过程中，通过调用所述链路追踪服务，获取所述第一服务的历史调用信息，所述历史调用信息包括所述链路追踪服务根据历史链路信息确定的所述第一服务调用过的服务的标识；第二获取模块，用于在所述第一服务启动后，从所述服务网格的控制面，获取所述历史调用信息包括的各个服务的服务发现数据；存储模块，用于将获取的服务发现数据存储到所述缓存中。
在一种可能的实现方式中，所述装置还包括：更新模块，用于在所述第一服务运行过程中，通过周期调用所述链路追踪服务，对所述缓存中的服务发现数据进行周期更新。
在一种可能的实现方式中，所述装置还包括：订阅模块，用于向所述服务网格的控制面订阅所述缓存中的服务发现数据，以使所述服务网格的控制面在所述服务发现数据更新时，向所述第一服务推送更新后的服务发现数据。
在一种可能的实现方式中，所述装置基于Java代理以插件方式实现，所述插件挂载到所述第一服务。
本申请的实施例提供了一种计算设备集群，包括至少一个计算设备，每个计算设备包括处理器和存储器；所述至少一个计算设备的处理器用于执行所述至少一个计算设备的存储器中存储的指令，以使得所述计算设备集群执行上述方法。
本申请的实施例提供了一种包含指令的计算机程序产品，当所述指令被计算设备集群运行时，使得所述计算设备集群执行上述方法。
本申请的实施例提供了一种计算机可读存储介质，包括计算机程序指令，当所述计算机程序指令由计算设备集群执行时，所述计算设备集群执行上述方法。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是――但不限于――电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(Random Access Memory，RAM)、只读存储器(Read Only Memory，ROM)、可擦式可编程只读存储器(Electrically Programmable Read-Only-Memory，EPROM或闪存)、静态随机存取存储器(Static Random-Access Memory，SRAM)、便携式压缩盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、数字多功能盘(Digital Video Disc，DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。
这里所描述的计算机可读程序指令或代码可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本申请操作的计算机程序指令可以是汇编指令、指令集架构(Instruction Set Architecture，ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(Local Area Network，LAN)或广域网(Wide Area Network，WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可 编程逻辑电路、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或可编程逻辑阵列(Programmable Logic Array，PLA)，该电子电路可以执行计算机可读程序指令，从而实现本申请的各个方面。
这里参照根据本申请实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本申请的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其它可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
也可以把计算机可读程序指令加载到计算机、其它可编程数据处理装置、或其它设备上，使得在计算机、其它可编程数据处理装置或其它设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其它可编程数据处理装置、或其它设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本申请的多个实施例的装置、系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，所述模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。
也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行相应的功能或动作的硬件(例如电路或ASIC(Application Specific Integrated Circuit，专用集成电路))来实现，或者可以用硬件和软件的组合，如固件等来实现。
尽管在此结合各实施例对本发明进行了描述，然而，在实施所要求保护的本发明过程中，本领域技术人员通过查看所述附图、公开内容、以及所附权利要求书，可理解并实现所述公开实施例的其它变化。在权利要求中，“包括”(comprising)一词不排除其他组成部分或步骤，“一”或“一个”不排除多个的情况。单个处理器或其它单元可以实现权利要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施，但这并不表示这些措施不能组合起来产生良好的效果。
以上已经描述了本申请的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术的改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。

Claims (19)

  1. 一种服务调用方法，其特征在于，应用于第一服务，所述第一服务为运行在服务网格中的微服务，所述方法包括：
    在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；
    在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；
    从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；
    基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    在所述缓存中存在所述第二服务的服务发现数据时，根据所述缓存中的所述第二服务的服务发现数据，调用所述第二服务。
  3. 根据权利要求2所述的方法，其特征在于，所述服务发现数据包括服务的实例的IP地址及负载信息，
    所述根据所述缓存中的所述第二服务的服务发现数据，调用所述第二服务，包括：
    根据预设的负载均衡规则及所述第二服务的实例的负载信息，从所述第二服务的实例中，选取目标实例；
    根据所述目标实例的IP地址，调用所述目标实例。
  4. 根据权利要求1-3中任意一项所述的方法，其特征在于，所述方法还包括：
    在所述第一服务运行过程中，通过所述服务网格中的链路追踪服务，记录所述第一服务的链路信息。
  5. 根据权利要求4所述的方法，其特征在于，所述方法还包括：
    在所述第一服务重新启动过程中，通过调用所述链路追踪服务，获取所述第一服务的历史调用信息，所述历史调用信息包括所述链路追踪服务根据历史链路信息确定的所述第一服务调用过的服务的标识；
    在所述第一服务启动后，从所述服务网格的控制面，获取所述历史调用信息包括的各个服务的服务发现数据；
    将获取的服务发现数据存储到所述缓存中。
  6. 根据权利要求4所述的方法，其特征在于，所述方法还包括：
    在所述第一服务运行过程中，通过周期调用所述链路追踪服务，对所述缓存中的服务发现数据进行周期更新。
  7. 根据权利要求1-6中任意一项所述的方法，其特征在于，所述方法还包括：
    向所述服务网格的控制面订阅所述缓存中的服务发现数据，以使所述服务网格的控制面在所述服务发现数据更新时，向所述第一服务推送更新后的服务发现数据。
  8. 根据权利要求1-7中任意一项所述的方法，其特征在于，所述方法基于Java代理以插件方式实现，所述插件挂载到所述第一服务。
  9. 一种服务调用装置，其特征在于，应用于第一服务，所述第一服务为运行在服务网格中的微服务，
    所述装置包括：
    判断模块，用于在所述第一服务调用第二服务时，判断所述第一服务的缓存中是否存在所述第二服务的服务发现数据；
    拦截模块，用于在所述缓存中不存在所述第二服务的服务发现数据时，拦截所述第一服务发出的调用请求；
    第一获取模块，用于从所述服务网格的控制面获取所述第二服务的服务发现数据，并将所述第二服务的服务发现数据存储到所述缓存中；
    第一调用模块，基于所述第二服务的服务发现数据，执行所述第一服务对所述第二服务的调用。
  10. 根据权利要求9所述的装置，其特征在于，所述装置还包括：
    第二调用模块，在所述缓存中存在所述第二服务的服务发现数据时，根据所述缓存中的所述第二服务的服务发现数据，调用所述第二服务。
  11. 根据权利要求10所述的装置，其特征在于，所述服务发现数据包括服务的实例的IP地址及负载信息，
    所述第二调用模块，包括：
    目标实例选取子模块，用于根据预设的负载均衡规则及所述第二服务的实例的负载信息，从所述第二服务的实例中，选取目标实例；
    目标实例调用子模块，用于根据所述目标实例的IP地址，调用所述目标实例。
  12. 根据权利要求9-11中任意一项所述的装置，其特征在于，所述装置还包括：
    链路信息记录模块，用于在所述第一服务运行过程中，通过所述服务网格中的链路追踪服务，记录所述第一服务的链路信息。
  13. 根据权利要求12所述的装置，其特征在于，所述装置还包括：
    历史调用信息获取模块，用于在所述第一服务重新启动过程中，通过调用所述链路追踪服务，获取所述第一服务的历史调用信息，所述历史调用信息包括所述链路追踪服务根据历史链路信息确定的所述第一服务调用过的服务的标识；
    第二获取模块，用于在所述第一服务启动后，从所述服务网格的控制面，获取所述历史调用信息包括的各个服务的服务发现数据；
    存储模块，用于将获取的服务发现数据存储到所述缓存中。
  14. 根据权利要求12所述的装置，其特征在于，所述装置还包括：
    更新模块，用于在所述第一服务运行过程中，通过周期调用所述链路追踪服务，对所述缓存中的服务发现数据进行周期更新。
  15. 根据权利要求9-14中任意一项所述的装置，其特征在于，所述装置还包括：
    订阅模块，用于向所述服务网格的控制面订阅所述缓存中的服务发现数据，以使所述服务网格的控制面在所述服务发现数据更新时，向所述第一服务推送更新后的服务发现数据。
  16. 根据权利要求9-15中任意一项所述的装置，其特征在于，所述装置基于Java代理以插件方式实现，所述插件挂载到所述第一服务。
  17. 一种计算设备集群，其特征在于，包括至少一个计算设备，每个计算设备包括处理器和存储器；
    所述至少一个计算设备的处理器用于执行所述至少一个计算设备的存储器中存储的 指令，以使得所述计算设备集群执行如权利要求1-8中任意一项所述的方法。
  18. 一种包含指令的计算机程序产品，其特征在于，当所述指令被计算设备集群运行时，使得所述计算设备集群执行如权利要求的1-8中任意一项所述的方法。
  19. 一种计算机可读存储介质，其特征在于，包括计算机程序指令，当所述计算机程序指令由计算设备集群执行时，所述计算设备集群执行如权利要求1-8中任意一项所述的方法。
PCT/CN2023/101382 2022-09-27 2023-06-20 服务调用方法及装置 WO2024066503A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211184674.2A CN117834739A (zh) 2022-09-27 2022-09-27 服务调用方法及装置
CN202211184674.2 2022-09-27

Publications (1)

Publication Number Publication Date
WO2024066503A1 true WO2024066503A1 (zh) 2024-04-04

Family

ID=90475946

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/101382 WO2024066503A1 (zh) 2022-09-27 2023-06-20 服务调用方法及装置

Country Status (2)

Country Link
CN (1) CN117834739A (zh)
WO (1) WO2024066503A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118524143A (zh) * 2024-07-22 2024-08-20 亿次网联(杭州)科技有限公司 基于服务网格的数据处理方法、设备及计算机可读介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110401696A (zh) * 2019-06-18 2019-11-01 华为技术有限公司 一种去中心化处理的方法、通信代理、主机以及存储介质
CN113055421A (zh) * 2019-12-27 2021-06-29 南京亚信软件有限公司 一种服务网格治理方法及系统
CN113364885A (zh) * 2021-06-29 2021-09-07 中国电信股份有限公司 微服务调用方法、装置、电子设备和可读存储介质
WO2022111313A1 (zh) * 2020-11-24 2022-06-02 深圳前海微众银行股份有限公司 一种请求处理方法及微服务系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110401696A (zh) * 2019-06-18 2019-11-01 华为技术有限公司 一种去中心化处理的方法、通信代理、主机以及存储介质
CN113055421A (zh) * 2019-12-27 2021-06-29 南京亚信软件有限公司 一种服务网格治理方法及系统
WO2022111313A1 (zh) * 2020-11-24 2022-06-02 深圳前海微众银行股份有限公司 一种请求处理方法及微服务系统
CN113364885A (zh) * 2021-06-29 2021-09-07 中国电信股份有限公司 微服务调用方法、装置、电子设备和可读存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118524143A (zh) * 2024-07-22 2024-08-20 亿次网联(杭州)科技有限公司 基于服务网格的数据处理方法、设备及计算机可读介质

Also Published As

Publication number Publication date
CN117834739A (zh) 2024-04-05

Similar Documents

Publication Publication Date Title
US10630779B2 (en) System and method for using VoIP session continuity information using logical scalable units
AU2017363366B2 (en) On-demand code execution in a localized device coordinator
US9917889B2 (en) Enterprise service bus routing system
US11258667B2 (en) Network management method and related device
US9361087B1 (en) Device driver aggregation in operating system deployment
EP3886399A1 (en) Method for invoking server and proxy server
US20180316589A1 (en) System and method for volte session continuation using logical scalable units
US20150213134A1 (en) Data query method and system and storage medium
CN108718347B (zh) 一种域名解析方法、系统、装置及存储介质
US9432321B2 (en) Method and apparatus for messaging in the cloud
US10666503B1 (en) Network connection and termination system
US10581993B2 (en) Method for forwarding traffic in application on mobile intelligent terminal
US9794360B1 (en) Estimating unique viewers of media content without cookies
CN110601981A (zh) 服务路由方法、服务提供方云域及服务调用方云域
US20140337471A1 (en) Migration assist system and migration assist method
WO2024066503A1 (zh) 服务调用方法及装置
CN114448895B (zh) 一种应用访问方法、装置、设备及介质
CN108093036B (zh) 一种获取资源的方法及装置
CN115242882A (zh) 一种基于传输层路由访问k8s容器环境的方法及装置
US10178191B1 (en) Using multiple cookies
CN112702441B (zh) 基于容器的访问数据处理方法、装置、系统及存储介质
EP3065379A1 (en) System and method for volte session continuation using logical scalable units
CN109445966B (zh) 事件处理方法、装置、介质和计算设备
US10791088B1 (en) Methods for disaggregating subscribers via DHCP address translation and devices thereof
US11481397B1 (en) Aggregating and emitting database activity record batches

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23869751

Country of ref document: EP

Kind code of ref document: A1