WO2016110234A1 - 面向云平台应用的服务推荐方法、设备及系统 - Google Patents

面向云平台应用的服务推荐方法、设备及系统 Download PDF

Info

Publication number
WO2016110234A1
WO2016110234A1 PCT/CN2016/070057 CN2016070057W WO2016110234A1 WO 2016110234 A1 WO2016110234 A1 WO 2016110234A1 CN 2016070057 W CN2016070057 W CN 2016070057W WO 2016110234 A1 WO2016110234 A1 WO 2016110234A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
target
recommended
dimension
matching
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2016/070057
Other languages
English (en)
French (fr)
Inventor
潘方敏
刘赫伟
高超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP16734894.5A priority Critical patent/EP3232338A4/en
Publication of WO2016110234A1 publication Critical patent/WO2016110234A1/zh
Priority to US15/641,137 priority patent/US20170300497A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G06F16/24578—Query processing with adaptation to user needs using ranking
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • 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
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network

Definitions

  • the present invention relates to the field of cloud computing, and in particular, to a service recommendation method, device and system for a cloud platform application.
  • cloud platform In the era of cloud computing, services are relative to users. Developers use the cloud platform, which is the developer.
  • the service on the cloud platform refers to the capabilities provided by the cloud platform for applications, such as database, log, storage, etc. Users can directly use the services on the platform to achieve rapid development when developing applications.
  • the cloud platform provides the entire production process environment of the software development as a service to the user. It provides a series of convenient tools for the user to realize the lifecycle management of the application and service on the cloud platform, such as the application. Deployment, hosting; integration of services, publishing, etc.
  • the cloud platform provides a complete end-to-end ecosystem around the use of objects such as application developers and service providers. The goal is to enable application developers to quickly deploy online applications, service providers to quickly and easily host and publish services.
  • On the platform there are a large number of application developers and service developers. Service developers have developed various services and integrated them on the cloud platform. The users of these services are the developers of the applications.
  • Application developers use the various types of services available on the platform to rapidly develop cloud applications.
  • a ruby application on the cloud platform uses the mysql database. Developers can install and deploy a mysql, and then use the mysql application; you can also use the mysql service provided by the cloud platform, so application developers can not care about mysql installation and deployment, as well as operation and maintenance, as long as you pay attention to the ruby program. You can do it yourself.
  • the cloud platform cannot provide the corresponding template in time, and during the service use, if the application is not satisfied with the currently used service, it can only choose to manually re-bind a service that is considered suitable. And try the effect again, the application developer can't get the service matching the application to be developed in time and accurately.
  • the embodiment of the invention provides a service recommendation method, device and system for a cloud platform application, so as to enable a developer to obtain a service matched with an application to be developed in a timely and accurate manner.
  • a service recommendation method for a cloud platform application comprising:
  • Obtaining service state information of the cloud platform where the service state information includes services used by all applications on the cloud platform, and running states of the respective services;
  • Obtaining a target service type which is a service type required by the target application, where the target application is an application running on the cloud platform and requiring service recommendation;
  • the service status information a service matching the target service type is obtained, and the service is used as a service to be recommended.
  • the service that matches the target service type is obtained according to the service status information, and the service is used as a service to be recommended, including:
  • the target recommendation policy dimension corresponding to the target service type is generated according to the recommended policy configuration library, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one service state statistical dimension.
  • the service status statistic dimension is used to perform matching degree statistics with the service in the target service type; or, according to the recommended policy configuration library, a recommended policy dimension corresponding to various service types on the cloud platform is generated, and the recommended policy configuration library includes Corresponding relationship between the recommended policy dimension and the service type, the recommended policy dimension includes at least one service state statistical dimension, and the service state statistical dimension is used to perform matching degree statistics with the service in the target service type;
  • the recommended recommendation policy dimension is obtained;
  • the recommended recommendation policy dimension is the recommended policy dimension corresponding to the service type required by the application;
  • a matching score of the service status of each service in the target service type is performed under the target recommendation policy dimension, and a matching score of the service status of each service is obtained;
  • the matching score reflects each The degree of matching of the running state of the service with the statistical dimension of the at least one service state included in the recommended policy dimension, wherein the higher the matching score, the higher the matching degree;
  • the matching scores of the service status of each service are compared, and the service with the highest matching score is taken as the service to be recommended.
  • the service that matches the target service type is obtained according to the service status information, and the service is used as the service to be recommended, including:
  • the target recommendation policy dimension corresponding to the target service type is generated according to the recommended policy configuration library, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one statistical dimension of the service state.
  • the service status statistical dimension is used to perform matching degree statistics with the service in the target service type;
  • a matching score of the service status of each service in the target service type is performed under the target recommendation policy dimension, and a matching score of the service status of each service is obtained;
  • the matching score reflects each The degree of matching of the running state of the service with the statistical dimension of the at least one service state included in the recommended policy dimension, wherein the higher the matching score, the higher the matching degree;
  • Match scores for the service status of each service are arranged in descending order
  • At least one service is selected as the service to be recommended among the target service types.
  • the target service type is selected according to the service state information
  • Each service in the service performs matching score statistics of the service status, and obtains a matching score of the service status of each service, including:
  • the target is served under each of the target recommendation policy dimensions.
  • Each service in the service type performs service status statistics, and obtains the service status of each service under each dimension;
  • the status of each service under each dimension is scored, and the score of each service in each dimension is obtained;
  • the score of each service in each dimension is multiplied by a preset weight and summed to obtain a matching score of the service state of each service under the target recommendation strategy dimension.
  • the target service type is selected according to the service state information After each service in the service performs matching score statistics of the service status, and after obtaining the matching score of the service status of each service, the method further includes:
  • the matching scores of the service status of each service are compared, and the matching scores of the service status of each service are arranged in descending order, including:
  • the method further includes:
  • a service recommendation device for a cloud platform application includes:
  • An information obtaining module configured to obtain service state information of the cloud platform, where the service state information includes services used by all applications on the cloud platform, and an operating state of each service;
  • a service type obtaining module configured to obtain a target service type, where the target service type is a service type required by the target application, where the target application is an application running on the cloud platform and requiring service recommendation;
  • a dynamic analysis service recommendation module configured to obtain a service state of the cloud platform obtained by the module according to the information
  • the information obtains a service matching the target service type obtained by the service type obtaining module, and uses the service as a service to be recommended.
  • the dynamic analysis service recommendation module includes:
  • a generating unit configured to configure a library according to the recommended policy, and generate a target recommendation policy dimension corresponding to the target service type, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one a service status statistical dimension, the service status statistical dimension is used to compare the degree of matching with the service in the target service type;
  • a matching statistics unit configured to obtain, according to the information, the service status information of the cloud platform obtained by the module, and perform, in the target recommendation policy dimension generated by the generating unit, a matching score of the service status of each service in the target service type, and obtain a matching score of the service status of each service;
  • the matching score reflects a matching degree of the running status of each service with a statistical dimension of at least one service status included in the recommended policy dimension, wherein the higher the matching score, the more the matching degree high;
  • the determining unit is configured to compare the matching scores of the service status of each service obtained by the matching statistical unit, and use the service with the highest matching score as the service to be recommended.
  • the dynamic analysis service recommendation module includes:
  • a generating unit configured to configure a library according to the recommended policy, and generate a target recommendation policy dimension corresponding to the target service type, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one a service status statistical dimension, the service status statistical dimension is used to compare the degree of matching with the service in the target service type;
  • a matching statistics unit configured to obtain, according to the information, the service status information of the cloud platform obtained by the module, and perform, in the target recommendation policy dimension generated by the generating unit, a matching score of the service status of each service in the target service type, and obtain a matching score of the service status of each service;
  • the matching score reflects a matching degree of the running status of each service with a statistical dimension of at least one service status included in the recommended policy dimension, wherein the higher the matching score, the more the matching degree high;
  • the selecting unit is configured to select, from the highest ranked service, at least one service in the target service type as the service to be recommended according to the order in which the arranged units are arranged.
  • the device further includes:
  • a correction module configured to calculate a workload of the target application to migrate from the currently used service to the service to be recommended, and correct a matching score of the service state of each service according to the size of the workload, to obtain a corrected matching score
  • the arranging unit is specifically configured to compare the corrected matching scores of the service states of each service, and arrange the matching scores of the service states of each service in descending order.
  • the device further includes:
  • a correction module configured to calculate a workload of the target application to migrate from the currently used service to the service to be recommended, and correct a matching score of the service state of each service according to the size of the workload, to obtain a corrected matching score
  • the determining unit is specifically configured to compare the corrected matching scores obtained by the correction module, and use the service with the highest matching score as the service to be recommended.
  • the device further includes:
  • a push module for pushing the service to be recommended to the developer of the target application.
  • the generating unit may be replaced by the following two units:
  • the policy dimension generating unit is configured to generate a recommended policy dimension corresponding to various service types on the cloud platform according to the recommended policy configuration library, where the recommended policy configuration database includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension Included at least one service status statistical dimension, the service status statistical dimension is used for matching degree statistics with the service in the target service type;
  • the recommended policy dimension unit is configured to obtain a recommendation policy dimension to be counted according to the recommended policy dimension corresponding to the various services on the cloud platform generated by the policy dimension generation unit; the to-be-stated recommendation policy dimension is required by the application The recommended policy dimension corresponding to the service type.
  • a service recommendation system for a cloud platform application includes a cloud platform and a service recommendation device for the cloud platform application, and the service recommendation device for the cloud platform application is used to The application recommendation service on the cloud platform, the service recommendation device for the cloud platform application is as described in the second aspect and the first to the first possible implementation of the second aspect.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and dynamically analyzes the recommended matching service according to the service status information of the cloud platform, and the used service more and more matches the characteristics of the user's usage scenario, and improves the recommended service and the target application. suitability.
  • FIG. 1 is a flowchart of a service recommendation method for a cloud platform application according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of a policy configuration library and its organization manner according to an embodiment of the present invention
  • FIG. 3 is a schematic diagram of generating a corresponding recommended policy dimension for various service type configurations on a cloud platform according to an embodiment of the present invention
  • FIG. 4 is a flowchart of a service recommendation method for a cloud platform application according to an embodiment of the present invention
  • FIG. 5 is a flowchart of a service recommendation method for a cloud platform application according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of a dynamic configuration of a recommendation policy according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of dynamic configuration of a recommendation policy according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of dynamic configuration of a recommendation policy according to an embodiment of the present invention.
  • FIG. 9 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 10 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 11 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 12 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 13 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 14 is a schematic diagram of a service recommendation method for a cloud platform application according to an embodiment of the present invention.
  • FIG. 15 is a structural diagram of a service recommendation system for a cloud platform application according to an embodiment of the present invention.
  • FIG. 16 is a flowchart of a service recommendation method for a cloud platform application according to an embodiment of the present invention.
  • FIG. 17 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 18 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 19 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • FIG. 20 is a structural diagram of a service recommendation device for a cloud platform application according to an embodiment of the present invention.
  • the cloud platform provides a set of mechanisms for service binding and access.
  • the application binds the service to be used through the cloud platform, and the platform obtains the access mode of the service, and provides the cloud application access through a common interface form.
  • the application only needs to use the interface provided by the platform to use the service, and does not need to perceive the specific usage details of the service in advance.
  • the platform provides the ability to use the same or similar interfaces for the same type of service. For developers, using a common interface means that services can be switched quickly.
  • the embodiment of the present invention provides a cloud platform application for an application and a service on a cloud platform.
  • Service recommendation method By detecting the usage characteristics of the service to which the application is bound, the service that best matches the application is compared and pushed to the application developer in the same type of service as the service to which the application is bound.
  • the service recommendation method for the cloud platform application includes:
  • S100 Obtain service state information of the cloud platform, where the service state information includes services used by all applications on the cloud platform, and an operating state of each service;
  • the cloud application and the cloud service on the platform implement the binding of the relationship through the cloud controller, and the running information of the service and the application is sent to the cloud collector on the cloud through the hosted virtual machine/container.
  • the service recommendation device collects the binding information on the cloud controller and the running status information on the cloud collector, and obtains the services used by all the applications on the platform and the running status of each service according to the obtained information, thereby analyzing each cloud. The best service for the application.
  • the binding information on the cloud controller and the running status information on the cloud collector may also be pushed to the service recommendation device in an active push manner, so that the best service of the cloud application is analyzed according to the specific information situation.
  • S101 obtains a target service type, where the target service type is a service type required by the target application, where the target application is an application running on the cloud platform and needs to perform service recommendation;
  • the target service type can be obtained by analyzing the code of the target application
  • the target service type can be obtained by analyzing attribute information of the target application.
  • the recommended policy information of the recommended library may be configured according to the policy, and the recommended policy dimension corresponding to the various services on the cloud platform may be generated, where the recommended policy configuration library reflects the correspondence between the recommended policy dimension and the service type;
  • the statistical results of each service are compared and calculated, and the statistical results of the services are arranged in descending order;
  • At least from the service with the highest statistical score at least from all services of the same type a service as a service to be recommended;
  • the method may further include:
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • an embodiment of the present invention provides a policy configuration library and an organization manner thereof.
  • the recommended policy configuration repository some common recommended policy dimensions and configuration policies related to service recommendation are configured.
  • the recommended policy dimension includes a recommendation dimension related to the self-characteristic of the service type, a service-related recommendation dimension common to the platform, and a recommendation dimension related to the general service of the industry.
  • a recommendation dimension related to the self-characteristic of the service type includes a recommendation dimension related to the self-characteristic of the service type, a service-related recommendation dimension common to the platform, and a recommendation dimension related to the general service of the industry.
  • other preferred policy dimensions may be included, such as a user-defined recommendation dimension, a specific recommendation dimension for a certain type of service, and the like, which are not specifically limited in the embodiment of the present invention.
  • a number of specific recommendation dimensions are included in each type of recommendation strategy dimension.
  • a recommended dimension of read/write usage may be included; for a map service, The recommended dimension of the user distribution may be included; for the log service, the recommended dimension of the log data type and the amount of data may be included.
  • the recommended policy dimensions common to the platform may include recommended dimensions such as the stability of the service, user usage, and user migration.
  • the recommended recommendation strategy dimension it can include the recommended dimensions of the package price, new online service promotion and service bidding ranking of the service.
  • each recommendation policy is configured with a weighting factor of the service recommendation.
  • the specific recommendation strategy and the corresponding weight coefficient are obtained, and the final recommended service list is calculated and sent as a recommendation result to the developer of the application that needs the service recommendation.
  • database type services have read and write usage, user segmentation, data type, stability, usage and other recommended strategies.
  • This database service may have SQL database services, MSQL database services, or Oracle database services.
  • the statistical results of each database service on the above recommended policy dimensions may differ from each other. According to the characteristics of the application, for each service, each recommendation strategy is multiplied by a weight coefficient, and then compared, this result finds a service that is superior or meets the user's needs.
  • an embodiment of the present invention provides a service recommendation method for a cloud platform application, including:
  • S200 Obtain service state information of the cloud platform, where the service state information includes services used by all applications on the cloud platform, and an operating state of each service;
  • the target service type is a service type required by the target application, where the target application is an application running on the cloud platform and requiring service recommendation;
  • the recommended policy configuration database includes a correspondence between the recommended policy dimension and the service type, where the recommended policy dimension includes at least one service status statistics. Dimension, the service status statistical dimension is used to match the degree of the service in the target service type;
  • S203 Perform, according to the service status information, a matching score of the service status of each service in the target service type under the target recommendation policy dimension, and obtain a matching score of the service status of each service;
  • the matching score reflects the The degree of matching of the running status of each service with the statistical dimension of at least one service status included in the recommended policy dimension, wherein the higher the matching score, the higher the matching degree;
  • S203 may include:
  • S2031 Perform service status statistics for each service in the target service type in each dimension of the target recommendation policy dimension according to the service status information, and obtain a service status of each service in each dimension.
  • S2032 Perform a scoring calculation on the status of each service in each dimension according to a preset scoring standard, and obtain a score of each service in each dimension;
  • S204 Compare the matching scores of the service status of each service, and use the service with the highest matching score as the service to be recommended.
  • S204 may alternatively be the following steps:
  • the matching scores of the service status of each service are arranged in descending order; starting from the highest score service, at least one service is selected as the service to be recommended among the target service types.
  • the method may further include:
  • the matching score used in step S204 is a few corrected matching scores.
  • the method may further include:
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • an embodiment of the present invention provides a service recommendation method for a cloud platform application.
  • This embodiment is different from the embodiment corresponding to FIG. 4 in step S202.
  • steps S202 and S430 are used to replace step S202.
  • the remaining steps are the same as the embodiment corresponding to FIG.
  • the method includes:
  • S400 and S401 are the same as steps S200 and S201, respectively, and are not described herein again.
  • the recommendation policy configuration library is configured to generate a recommendation policy dimension corresponding to various service types on the cloud platform, where the recommendation policy configuration library includes a correspondence between a recommendation policy dimension and a service type, where the recommendation policy dimension includes at least one a service status statistical dimension, the service status statistical dimension is used to compare the degree of matching with the service in the target service type;
  • the recommended recommendation policy dimension is obtained;
  • the recommended recommendation policy dimension is a recommended policy dimension corresponding to the service type required by the application;
  • the recommended policy dimension corresponding to various services on the cloud platform is generated in S420, the recommended policy dimension corresponding to the same service type as the target service type row is matched, and the recommended policy dimension is to be The statistical recommendation strategy dimension.
  • the recommended policy dimension corresponding to the database service may be matched from the recommended policy dimension corresponding to the service, and the recommended policy dimension is used as the recommended recommendation policy dimension.
  • S403, S404, S405, and S410 are the same as S203, S204, S205, and S210, respectively, and are not described herein again.
  • S403 is the same as that of S203, that is, S4031-S4033 is the same as S2031-S2033, and details are not described herein again.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • the second is the service recommendation for applications that have been running for a while; the specific processing flow of these two processes is introduced below.
  • an embodiment of the present invention provides a service recommendation method for a cloud platform application, including:
  • step S300 the cloud platform application is applied.
  • the service recommendation device reads various recommended dimension information in the policy configuration recommendation during initialization. In this way, a recommendation policy dimension corresponding to various services on the cloud platform is generated.
  • the service recommendation device reads the recommendation dimension of the read/write usage, the recommendation strategy of the user distribution, the recommendation strategy of the DB stability, etc., thereby generating the corresponding recommendation policy dimension of the DB service;
  • the service recommendation device determines the recommended policy dimensions related to the log service by reading various recommended dimensions in the read policy configuration recommendation, such as:
  • This dimension includes: log data type; log data volume; log save time; log query function and other recommended dimensions;
  • the industry-wide recommendation strategy dimension of the Log service includes: package price, new online service, bidding ranking and other recommended dimensions;
  • step S300 by reading the recommended policy configuration library of FIG. 2, corresponding recommendation policy dimensions are generated for various service type configurations on the cloud platform.
  • its recommended policy dimensions include recommended policy dimensions such as database read/write usage, stability, user usage, and user migration.
  • the recommended policy dimensions such as stability and package price
  • the weighting coefficients of these aspects are recommended for each service recommendation strategy. different. In other words, each service will have an independent recommendation policy configuration.
  • the service recommendation policies used may also be different when applying initial deployments and dynamic analysis adjustments.
  • the main difference is that the monitoring information for the service running state can only be obtained when the dynamic analysis is adjusted, such as the log data type and data volume of the application, and the method of using the log service by the application. These policy dimensions are ignored during the initial deployment process. These policy dimensions have a weighting factor of 0 during the initial deployment.
  • Service status information includes services used by all applications on the cloud platform and services The running state.
  • the service status information may be represented by running status information of the service and the application;
  • the running status information of the service and the application includes binding information of the service and the application, and running information of the service and the application;
  • binding information from cloud controller services and applications and operational information of services and applications on the cloud collector can be obtained;
  • the sampling data from the cloud controller, the cloud collector, and the like may be received by means of message subscription; in an embodiment, the cloud controller, the cloud collector, and the like may also be actively queried and obtained. Sampling data from the source;
  • a search application by periodically obtaining the information of the service bound to it and the running information of the application and the service to which it is bound, a series of information of the service bound to the search application can be analyzed; The database service is bound to the search application.
  • the read/write usage, stability, and total user usage of the database service can be obtained.
  • the message carries attribute information of the newly deployed application
  • the cloud controller when an application is newly deployed on the cloud platform, the cloud controller sends a notification message to the service recommendation device, where the message carries attribute information of the application, such as a language stack used by the application and a usage framework.
  • the service recommendation device for the cloud platform application can know the potential service type required by the new application according to the attribute information of the new application, such as the language stack used and the usage framework.
  • the services required for a new application are carried in their attribute information.
  • a search application can know that the search application needs a database service through its attribute information.
  • a navigation application can know that the navigation application needs a map service through its attribute information.
  • the type of service required by the application may also be obtained by analyzing the public code information of the new application;
  • the affiliate message carried by the new application it is also possible to accept the affiliate message carried by the new application, the affiliate message or the type of service required by the application.
  • the developer of the application can develop the application.
  • the service type is carried in the attached message of the modified application.
  • the developer card method requires a map type service.
  • the developer can carry an affiliate message in the application, and the accessory message carries a message of a required service type, such as a map service.
  • the service recommendation device can or according to the type of service required by the application.
  • the recommended recommendation policy dimension is obtained according to the recommended policy dimension corresponding to the various services on the cloud platform generated in the S300.
  • the recommended recommendation policy dimension is a recommended policy dimension corresponding to the service type required by the application.
  • the recommended policy dimension of the database service needs to be analyzed, for example, the recommended policy dimensions of the database service that need to be analyzed include: read/write usage of the database, stability, user usage, and user migration.
  • the recommended policy information corresponding to various services on the cloud platform is generated according to the recommendation policy information of the policy configuration recommendation library. Then, by obtaining the service type required by the newly deployed application, the recommended policy dimension corresponding to the service type required by the newly deployed application is found in the recommended policy dimension corresponding to the generated various services.
  • the service type required by the newly deployed application may be obtained first, and then the library is configured according to the recommended policy, and the target recommendation policy dimension corresponding to the service type is generated, as in the previous steps. S202 should be the same.
  • the recommended policy dimensions for the statistical analysis required by the service type in S304 include: read/write usage of the database, stability, user usage, and user migration.
  • the database service types on the cloud platform are MYSQL, ACCESS, and BD2. Then, according to the service related information periodically obtained in S301, the statistical result of each dimension of the recommended policy dimension required for statistical analysis of each database service included in the database service type can be obtained:
  • the dimensions of the database read/write usage are: 50% read and 50% write;
  • the dimension of stability is: high stability;
  • the dimension of user usage is: 100 uses;
  • the dimension of user migration is: 50 users move out to stop using this service, and 50 users move in to use this service;
  • the dimensions of the database read/write usage are: 60% read and 40% write;
  • the dimension of stability is: medium stability;
  • the dimension of user usage is: 50 uses;
  • the dimension of the user migration volume is: 100 times the user moves out to stop using the service, and 50 times the user moves in to use the service;
  • the dimensions of the database read/write usage are: 40% read and 60% write;
  • the dimension of stability is: high stability;
  • the dimension of user usage is: 200 uses;
  • the dimension of user migration is: 50 users move out to stop using this service, and 100 users move in to use this service.
  • the recommended device defaults to the database type service at the time of recommendation, and the default rating criteria for each dimension is:
  • the dimensions of the database read/write usage are: 50% read and 50% write;
  • the dimension of stability is: high stability;
  • the dimension of user usage is: 100 uses;
  • the dimension of user migration is: the number of users moving in is greater than the amount of users moving out;
  • the preset scoring standard of the dimension of the read/write usage is: the frequency of reading is greater than the frequency of writing. Such as: 50% read and 50% write.
  • the invention is not particularly limited.
  • the score is 10 points. If it is not satisfied, it will be deducted by 2 in one unit according to the degree of dissatisfaction.
  • the dimension of the read/write usage of the database is: 50% read and 50% write - 10 points;
  • the dimensional score of stability is: high stability - 10 points;
  • the dimension of the user usage is scored as: 100 uses - 10 points;
  • the dimension of user migration is scored as follows: 50 users move out to stop using this service, and 50 users move in to use this service—8 points;
  • the dimension of the read/write usage of the database is: 60% read and 40% write - 8 points;
  • the dimensional score of stability is: medium stability - 8 points;
  • the dimension of the user usage is scored as: 50 uses - 6 points;
  • the dimension of the user migration volume is: 100 times the user moves out to stop using the service, and 50 times the user moves in to use the service—6 points;
  • the dimension of the read/write usage of the database is: 40% read and 60% write - 8 points;
  • the dimensional score of stability is: high stability - 10 points;
  • the dimension of the user usage is scored as: 200 uses - 10 points;
  • the dimension of user migration is scored as follows: 50 users move out to stop using this service, and 100 users move in to use this service—10 points.
  • the service recommendation list includes at least one service to be recommended
  • the weights preset in the initial recommendation are balanced, and the weights for each dimension of the service are the same. For example, for the appeal service, each service has 4 recommended dimensions, then the weight of each dimension is 1/4.
  • the weight may also be requested by the user in advance.
  • the user may pay more attention to stability, and the weight of the stability may be appropriately higher.
  • the dimension of the user migration amount may have a weight of 2/5, and other remaining dimensions. The weights are both 1/5.
  • the services are ranked in descending order according to the total score of each service to obtain a final service recommendation list.
  • the rankings of several services are: MYSQL service, BD2 service, and ACCESS service.
  • the ranking of several services is:
  • the statistics of the recommended dimensions of the bound service may change, periodically perform re-statistical analysis, and recommend the most appropriate service to the user based on the results of the re-statistical analysis.
  • the usage characteristics of the application are obtained based on the statistical results. For example, after re-stating for a search application, it is found that the search application frequently reads and writes the database, and the read-write ratio reaches 7:3.
  • the dimension of the service of the database service type is re-stated to obtain:
  • the dimensions of the database read/write usage are: 70% read and 30% write;
  • the dimension of stability is: high stability;
  • the dimension of user usage is: 80 uses;
  • the dimension of user migration is: 70 users move out to stop using this service, and 60 users move in to use this service;
  • the dimensions of the database read/write usage are: 50% read and 50% write;
  • the dimension of stability is: medium stability;
  • the dimension of user usage is: 80 uses;
  • the dimension of user migration is: 70 users move out to stop using this service, and 100 users move in to use this service;
  • the dimensions of the database read/write usage are: 40% read and 60% write;
  • the dimension of stability is: high stability;
  • the dimension of user usage is: 100 uses;
  • the dimension of user migration is: 100 times the user moves out to stop using this service, and 100 times the user moves in to use this service.
  • S309 according to the update analysis result obtained in S308, perform recalculation of the scores of the recommended policy dimensions for each service of the service-bound service of the application and recalculate the weights, and obtain the updated service recommendation list and send the updated service recommendation list;
  • the updated service recommendation list includes at least one service to be recommended;
  • the weighting coefficient of each recommended dimension may be dynamically adjusted, and according to the state information of the service and the application, the weighting coefficients of some service dimensions may be reduced or improved;
  • a dimension such as a stability dimension
  • a service has experienced a crash, a crash, etc.
  • its weight will decrease over a period of time, thereby reducing the recommended ranking.
  • the MYSQL service is more focused on the read operation characteristics, and the updated rankings are MYSQL service, ACCESS, and BD2 services according to the actual usage of the read operation characteristics.
  • MYSQL is more suitable for this application than the BD2 service.
  • the sending of the service recommendation list may be sent to the user by using an email/sms/message/message push/web push/client message push, etc., and the user decides whether to select the handover service according to the recommended information. .
  • the recommendation device may also select a service with the highest score to send to the user by means of mail/sms/message push/web push/client message push.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • the application developer may use a whitelist to select a recommendation service to select only in a whitelist, or a blacklist. Block some services that developers are not willing to use.
  • the recommended device repurchases the configuration of the blacklist and whitelist of the office, selects the service in the recommended whitelist for the application, or removes the service in the blacklist.
  • the recommended device will remove the blacklisted total service from the service recommendation list, or further select the service in the whitelist in the service recommendation list.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • Service which improves the development efficiency of the application, and
  • the recommended matching service is used, and the used service more and more matches the characteristics of the user's usage scenario, and the matching degree between the recommended service and the target application is improved.
  • the recommended service can better match the developer's personal preferences.
  • FIG. 7 another scheme for recommending policy dynamic configuration is provided as shown in FIG. 7.
  • the configuration scheme is a weight assignment method for distinguishing roles, and provides weight configuration of different service dimensions for different application developers.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • FIG. 1 Another scheme for recommending policy dynamic configuration is provided as shown in FIG.
  • the solution is to customize the service dimensions and weights by the application developer, and the application developer configures the dimension priority of the services used by the applications deployed by the application developer.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • the application developer configures the services used by the deployed application. The priority of the dimension that the company pays attention to can better match the developer's application development habits and make the recommendation scheme more personalized.
  • the embodiment of the present invention discloses a service recommendation device for a cloud platform application, which is used to execute the foregoing method, and the device includes:
  • the information obtaining module 110 is configured to obtain service state information of the cloud platform, where the service state information includes services used by all applications on the cloud platform and operating states of the respective services;
  • the service type obtaining module 120 is configured to obtain a target service type, where the target service type is a service type required by the target application, where the target application is an application running on the cloud platform and needs to perform service recommendation;
  • the dynamic analysis service recommendation module 130 is configured to obtain, according to the information, the service status information of the cloud platform obtained by the module, and obtain the service that matches the target service type obtained by the service type obtaining module 120, and use the service as the service to be recommended. .
  • the device may further include:
  • the pushing module 140 is configured to push the service to be recommended to the developer of the target application.
  • the dynamic analysis service recommendation module 130 includes:
  • the generating unit 1301 is configured to generate a target recommendation policy dimension corresponding to the target service type according to the recommended policy configuration library, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one a service status statistical dimension, the service status statistical dimension is used to compare the degree of matching with the service in the target service type;
  • the matching statistics unit 1302 is configured to perform, according to the service state information of the cloud platform obtained by the module, the matching score statistics of the service state of each service in the target service type under the target recommendation policy dimension of the generating unit 1301 Obtaining a matching score of the service status of each service; the matching score reflects a matching degree of the running status of each service with a statistical dimension of at least one service status included in the recommended policy dimension, wherein the matching score is higher The higher the degree;
  • the determining unit 1303 is configured to compare the matching scores of the service status of each service obtained by the matching statistical unit 1302, and use the service with the highest matching score as the service to be recommended.
  • the device may further include:
  • a correction module 150 configured to calculate that the target application migrates from the currently used service to the to-be recommended service The workload of the service, according to the size of the workload, the matching score of the service status of each service is corrected, and the corrected matching score is obtained;
  • the arranging unit 1304 is specifically configured to compare the corrected matching scores of the service states of each service obtained by the correction module 150, and arrange the matching scores of the service states of each service in descending order.
  • the generating unit 1301 may generate another scheme for recommending policy dynamic configuration. This is a whitelist/blacklist method.
  • the application developer can use the whitelist to select the recommendation service to select only in the whitelist list, or use the blacklist method to block some services that developers are not willing to use. As shown in Figure 6 above.
  • the device may further include:
  • the service selection module 160 is configured to select a service in the whitelist according to the configuration of the black and white list of the generating unit 1301, or remove the service in the blacklist.
  • the service selection module 160 may perform black and white list screening before matching the statistical unit 1302 unit; in this case, the service selection module first filters the service according to the black and white list, and then The matching statistics unit 1302 performs matching based on the filtered services.
  • the line 2 is connected by a broken line in FIG. The matching process has been described in detail in the description of the above-mentioned matching statistic unit 1302, and details are not described herein again.
  • the service selection module 160 may also perform the screening of the black and white list before the dynamic analysis service recommendation module 130, such as the dotted line connection line in FIG. 12. In this case, the service selection The service finally selected by the module 160 is the service to be recommended, and the push module 140 pushes the developer to the target application.
  • service selection module 160 and the correction module 150 may be jointly set in the service recommendation device for the cloud platform application, or may be alternatively configured.
  • the dynamic analysis service recommendation module 130 includes:
  • the generating unit 1301 is configured to generate a target recommendation policy dimension corresponding to the target service type according to the recommended policy configuration library, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one a service status statistical dimension, the service status statistical dimension is used to compare the degree of matching with the service in the target service type;
  • the matching statistics unit 1302 is configured to perform, according to the service state information of the cloud platform obtained by the module, the matching score statistics of the service state of each service in the target service type under the target recommendation policy dimension of the generating unit 1301 Obtaining a matching score of the service status of each service; the matching score reflects a matching degree of the running status of each service with a statistical dimension of at least one service status included in the recommended policy dimension, wherein the matching score is higher The higher the degree;
  • the arranging unit 1304 is configured to sort the matching scores of the service states of each service obtained by the matching statistic unit 1302 in descending order;
  • the selecting unit 1305 is configured to select, according to the highest ranked service, at least one service among the target service types as the service to be recommended according to the order in which the arranged units are arranged.
  • the device may further include:
  • the correction module 150 is configured to calculate a workload of the target application to migrate from the currently used service to the service to be recommended, and correct the matching score of the service state of each service according to the workload, to obtain a corrected matching score;
  • the determining unit 1303 is specifically configured to compare the corrected matching scores obtained by the correcting module 150, and use the service with the highest matching score as the service to be recommended.
  • the generating unit 1301 may generate another scheme for recommending policy dynamic configuration. This is a whitelist/blacklist method.
  • the application developer can use the whitelist to select the recommendation service to select only in the whitelist list, or use the blacklist method to block some services that developers are not willing to use. As shown in Figure 6 above.
  • the device may further include:
  • the service selection module 160 is configured to select a service in the whitelist according to the configuration of the black and white list of the generating unit 1301, or remove the service in the blacklist.
  • the service selection module 160 may perform a black and white list screening before the matching statistics unit 1302 unit, such as the dotted line connection line 2 in FIG. 13; in this case, the service selection module The service is first filtered according to the black and white list, and then the matching statistic unit 1302 performs matching according to the filtered service.
  • the matching process has been described in detail in the description of the above-mentioned matching statistic unit 1302, and details are not described herein again.
  • the service selection module 160 may also be in the dynamic analysis service recommendation module. After the push module 140, the push module 140 performs the screening of the black and white list, such as the dotted line connection line 1 in FIG. 13. In this case, the service selected by the service selection module 160 is the service to be recommended, and has the push module 140. Push to the developer of the target app.
  • service selection module 160 and the correction module 150 may be jointly set in the service recommendation device for the cloud platform application, or may be alternatively configured.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • the application developer configures the dimension priority of the service used by the deployed application to better match the developer's application development habits.
  • the recommended scheme is more personalized.
  • the generating unit 1301 in the device embodiment corresponding to FIG. 10 to FIG. 13 above may be replaced by the following two units, and the remaining module units are unchanged:
  • the policy dimension generating unit 1320 is configured to: according to the recommended policy configuration library, generate a recommended policy dimension corresponding to various service types on the cloud platform, where the recommended policy configuration database includes a correspondence between the recommended policy dimension and the service type, and the recommended policy
  • the dimension includes at least one service status statistical dimension, and the service status statistical dimension is used to perform matching degree statistics with the service in the target service type;
  • the to-be-recommended policy dimension unit 1330 is configured to obtain a recommendation policy dimension to be counted according to the recommendation policy dimension corresponding to the various services on the cloud platform generated by the policy dimension generation unit 1320.
  • the to-be-stated recommendation policy dimension is required by the application.
  • the recommended policy dimension for the corresponding service type is required by the application.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • Service which improves the development efficiency of the application, and dynamically analyzes the recommended matching service according to the service status information of the cloud platform, and the service used is more and more
  • the user's usage scenario features improve the matching between the recommended service and the target application.
  • the application developer configures the dimension priority of the service used by the application deployed by the application developer, which can better match the developer's application development habits and make the recommendation scheme more personalized. Chemical.
  • an embodiment of the present invention provides a service recommendation system for a cloud platform application, where the system includes a cloud platform and a service recommendation device for the cloud platform application, and the service recommendation device for the cloud platform application is used to Application recommendation service on the cloud platform.
  • the service recommendation device for the cloud platform application is a relatively independent subsystem, and the location on the cloud platform is as shown in the figure.
  • the cloud application and the cloud service on the platform implement the binding of the relationship through the cloud controller, and the running information of the service and the application is sent to the information collector on the cloud through the hosted virtual machine/container.
  • the service recommendation system collects the binding information on the cloud controller and the running status information on the cloud collector, and analyzes the best service of the cloud application according to the obtained information.
  • the service recommendation device for cloud platform applications is used to:
  • Obtaining service state information of the cloud platform where the service state information includes services used by all applications on the cloud platform, and running states of the respective services;
  • Obtaining a target service type which is a service type required by the target application, where the target application is an application running on the cloud platform and requiring service recommendation;
  • the service recommendation device for the cloud platform application is specifically used to:
  • Obtaining service state information of the cloud platform where the service state information includes services used by all applications on the cloud platform, and running states of the respective services;
  • Obtaining a target service type which is a service type required by the target application, where the target application is an application running on the cloud platform and requiring service recommendation;
  • the target recommendation policy dimension corresponding to the target service type is generated according to the recommended policy configuration library, where the recommended policy configuration library includes a correspondence between the recommended policy dimension and the service type, and the recommended policy dimension includes at least one service state statistical dimension.
  • the service status statistical dimension is used to perform matching degree statistics with the service in the target service type;
  • a matching score of the service status of each service in the target service type is performed under the target recommendation policy dimension, and a matching score of the service status of each service is obtained;
  • the matching score reflects each The degree of matching of the running state of the service with the statistical dimension of the at least one service state included in the recommended policy dimension, wherein the higher the matching score, the higher the matching degree;
  • the matching scores of the service status of each service are arranged in descending order; starting from the highest score service, at least one service is selected as the service to be recommended among the target service types.
  • the service recommendation device for the cloud platform application is configured to perform, according to the service status information, a matching score of the service status of each service in the target service type under the target recommendation policy dimension, and obtain the When matching the service status of a service, the specificity is used to:
  • service status statistics are performed on each of the target service types in each of the target recommendation policy dimensions, and the service status of each service in each dimension is obtained;
  • the status of each service under each dimension is scored, and the score of each service in each dimension is obtained;
  • the score of each service in each dimension is multiplied by a preset weight and summed to obtain a matching score of the service state of each service under the target recommendation strategy dimension.
  • the embodiment of the present invention obtains a service that matches the target service type according to the service state information of the cloud platform, reduces the cost of the application developer to evaluate and use the service, and enables the application developer to obtain the application to be developed in time.
  • the service improves the development efficiency of the application, and the recommended service is analyzed according to the service status information of the cloud platform.
  • the service used is more and more matched with the characteristics of the user's usage scenario, and the matching between the recommended service and the target application is improved. degree.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computational Linguistics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种面向云平台应用的服务推荐方法，所述方法包括：获得云平台的服务状态信息（S100），所述服务状态信息包括所述云平台上的所有应用分别使用的服务以及各个服务的运行状态；获得目标服务类型（S101），所述目标服务类型为目标应用所需的服务类型，所述目标应用为运行在所述云平台上且需要进行服务推荐的应用；根据所述服务状态信息，获得与所述目标服务类型匹配的服务，将所述服务作为待推荐的服务（S102）。通过以上技术方案，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。

Description

面向云平台应用的服务推荐方法、设备及系统 技术领域
本发明涉及云计算领域，特别涉及一种面向云平台应用的服务推荐方法、设备及系统。
背景技术
在云计算时代，服务是相对于用户而言的。开发者使用云平台，云平台就是服务于开发者。云平台上的服务，指的是云平台面向应用所提供了的一些能力，比如数据库、日志、存储等，用户在进行应用开发时可以直接使用平台上的服务来实现快速开发。
云平台将整个软件开发的生产流程环境作为一种服务提供给了用户，它通过服务的方式，为用户提供一系列的便利工具，以实现云平台上应用、服务的生命周期管理，如应用的部署、托管；服务的集成，发布等。围绕应用开发者、服务提供者等使用对象，云平台提供了一套完整的端到端的生态系统。其目标是为了实现应用开发者快速部署上线应用，服务提供者方便快速的托管和发布服务等。在平台上，有大量的应用开发者和服务开发者。服务开发者开发了各种服务并集成托管到了云平台上，这些服务的用户就是应用的开发者。应用开发者利用平台上提供着的各类型的服务，进行云应用的快速开发。举个例子，云平台上一个ruby的应用，要使用mysql数据库。开发者可以自己安装部署一个mysql，然后应用去使用这个mysql；也可以使用云平台提供的mysql服务，这样应用开发者就可以不去关心mysql的安装部署，以及运维等情况，只要关注ruby程序自身即可。
在云平台上有大量的服务提供者，这些提供者会针对某种服务类型，提供服务，同服务类型有时候就会存在大量的服务。这些同类型的服务实现了相近的基础功能，又存在一些服务自身独有的特性以及使用特点。现有技术中，针对云平台上服务众多的特点，对一些典型的应用和服务使用场景，提供典型应用模板，根据模板建立相关资源和绑定关系。开发者选择其中的某个应用模板 进行部署应用，云平台在创建应用的同时，自动创建服务实例并与应用进行绑定。
这样，对于一些新兴的应用和服务，云平台就无法及时提供出相应的模板，而且服务使用过程中，如果应用对当前使用的服务不满意，只能选择手动重新绑定一个自认为合适的服务，再尝试效果，应用开发者不能够及时和准确地获得到和待开发应用匹配的服务。
发明内容
本发明实施例提供一种面向云平台应用的服务推荐方法、设备和系统，用以让开发者及时和准确地获得和待开发应用匹配的服务。
根据本发明实施例的第一方面，提供了一种面向云平台应用的服务推荐方法，该方法包括：
获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
根据该服务状态信息，获得与该目标服务类型匹配的服务，将该服务作为待推荐的服务。
结合第一方面，在第一方面的第一种可能的实现方式中，该根据该服务状态信息，获得与该目标服务类型匹配的服务，将该服务作为待推荐的服务，包括：
根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；或者，根据推荐策略配置库，生成与云平台上各种服务类型相对应的推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计； 根据生成的与云平台上各种服务相对应的推荐策略维度，得到待统计推荐策略维度；该待统计推荐策略维度为该应用所需的服务类型相对应的推荐策略维度；
根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
对该每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务。
结合第一方面，在第一方面的第二种可能的实现方式中，该根据该服务状态信息，获得与该目标服务类型匹配的服务，将该服务作为待推荐的服务，包括：
根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态的统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
对该每个服务的服务状态的匹配分数按照从高到低的顺序排列；
从分数最高的服务开始，在该目标服务类型中选择至少一个服务作为待推荐的服务。
结合第一方面的第一种或第二种可能的实现方式，在第一方面的第三种可能的实现方式中，该根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数，包括：
根据该服务状态信息，在该目标推荐策略维度中的每个维度下对该目标服 务类型中每个服务进行服务状态统计，得到每个服务在该每个维度下的服务状态；
根据预设的评分标准，对该每个服务在该每个维度下的状态进行评分计算，得到每个服务在每个维度下的分数；
将该每个服务在每个维度下的分数乘以预先设置的权重并求和，得到每个服务在该目标推荐策略维度下的服务状态的匹配分数。
结合第一方面的第一种或第二种可能的实现方式，在第一方面的第四种可能的实现方式中，该根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数之后，该方法还包括：
计算该目标应用从当前使用的服务迁移到该待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
该对该每个服务的服务状态的匹配分数进行比较，将该每个服务的服务状态的匹配分数按照从高到低的顺序排列，包括：
对该每个服务的服务状态的修正后的匹配分数进行比较，将该每个服务的服务状态的匹配分数按照从高到低的顺序排列；
结合第一方面以及第一方面的第一种至第四种任一种可能的实现方式，在第一方面的第五种可能的实现方式中该方法还包括：
将该待推荐的服务推送给该目标应用的开发者。
根据本发明实施例的第二方面，提供了一种面向云平台应用的服务推荐设备，该设备包括：
信息获得模块，用于获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
服务类型获得模块，用于获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
动态分析服务推荐模块，用于根据该信息获得模块获得的云平台的服务状 态信息，获得与该服务类型获得模块获得的目标服务类型匹配的服务，将该服务作为待推荐的服务。
结合第二方面，在第二方面的第一种可能的实现方式中，该动态分析服务推荐模块包括：
生成单元，用于根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
匹配统计单元，用于根据该信息获得模块获得的云平台的服务状态信息，在该生成单元生成的目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
确定单元，用于对该匹配统计单元得到的每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务。
结合第二方面，在第二方面的第二种可能的实现方式中，该动态分析服务推荐模块包括：
生成单元，用于根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
匹配统计单元，用于根据该信息获得模块获得的云平台的服务状态信息，在该生成单元生成的目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
排列单元，用于对该每个服务的服务状态的匹配分数按照从高到低的顺序排列；
选择单元，用于从分数最高的服务开始，按照该排列单元排列的顺序在该目标服务类型中选择至少一个服务作为待推荐的服务。
结合在第二方面的第一种可能的实现方式，在第二方面的第三种可能的实现方式，该设备还包括：
修正模块，用于计算该目标应用从当前使用的服务迁移到该待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
该排列单元具体用于，对该每个服务的服务状态的修正后的匹配分数进行比较，将该每个服务的服务状态的匹配分数按照从高到低的顺序排列。
结合在第二方面的第二种可能的实现方式，在第二方面的第四种可能的实现方式，该设备还包括：
修正模块，用于计算该目标应用从当前使用的服务迁移到该待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
该确定单元具体用于，对该修正模块得到的修正后的匹配分数进行比较，将修正后的匹配分数最高的服务作为待推荐的服务。
结合第二方面以及第二方面的第一种至第四种任一种可能的实现方式，在第二方面的第五种可能的实现方式中，其特征在于，该设备还包括：
推送模块，用于将该待推荐的服务推送给该目标应用的开发者。
结合第二方面以及第二方面的任一种可能的实现方式，生成单元可以由下述两个单元替换：
策略维度生成单元，用于根据推荐策略配置库，生成与云平台上各种服务类型相对应的推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
待推荐策略维度单元，用于根据该策略维度生成单元中生成的与云平台上各种服务相对应的推荐策略维度，得到待统计推荐策略维度；该待统计推荐策略维度为该应用所需的服务类型相对应的推荐策略维度。
根据本发明实施例的第三方面，提供了一种面向云平台应用的服务推荐系统，该系统包括云平台及面向云平台应用的服务推荐设备，该面向云平台应用的服务推荐设备用于向该云平台上的应用推荐服务，该面向云平台应用的服务推荐设备如第二方面以及第二方面的第一种至第无种任一种可能的实现方式所述。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与所述目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1本发明实施例提供一种面向云平台应用的服务推荐方法流程图；
图2本发明实施例提供一种策略配置库及其组织方式示意图；
图3本发明实施例提供一种针对云平台上的各种服务类型配置生成相应的推荐策略维度的示意图；
图4本发明实施例提供一种面向云平台应用的服务推荐方法流程图；
图5本发明实施例提供一种面向云平台应用的服务推荐方法流程图；
图6本发明实施例提供一种推荐策略动态配置示意图；
图7本发明实施例提供一种推荐策略动态配置示意图；
图8本发明实施例提供一种推荐策略动态配置示意图；
图9本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图10本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图11本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图12本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图13本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图14本发明实施例提供一种面向云平台应用的服务推荐方法情景示意图；
图15本发明实施例提供一种面向云平台应用的服务推荐系统结构图；
图16本发明实施例提供一种面向云平台应用的服务推荐方法流程图。
图17本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图18本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图19本发明实施例提供一种面向云平台应用的服务推荐设备结构图；
图20本发明实施例提供一种面向云平台应用的服务推荐设备结构图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
通常，云平台提供了一套服务绑定和使用访问的机制。应用通过云平台绑定要使用的服务，由平台获得得到服务的访问方式，并通过通用的接口形式提供给云应用访问使用。在这一整套服务使用流程中，应用只需要使用平台提供的接口方式即可使用服务，不需要提前感知该服务的具体使用细节。平台就提供了这样一种能力：对于同一类型的服务，应用使用相同或相近的接口方式，对于开发者而言，使用通用的接口就意味着可以快速实现服务的切换使用。
对于应用开发者而言，在应用开发之前需要明确会使用到的平台上的服务类型，但是具体使用某种服务，一方面需要明确自身应用的使用特点，还要了解每个平台服务的特点；另一方面在对应用的使用场景预估不准，或应用的使用人群或场景等出现变化时，还需要考虑怎样重新选择一个适合应用的服务。
本发明实施例针对云平台上的应用和服务，提供了一种面向云平台应用的 服务推荐方法。通过检测应用所绑定的服务的使用特点，在与该应用所绑定的服务同类型的服务中，比较与该应用最为匹配的服务并将之推送给应用开发者。
如图1所示，该面向云平台应用的服务推荐方法包括：
S100，获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
平台上的云应用与云服务，通过云控制器实现关系的绑定，而服务与应用的运行信息，通过其托管的虚拟机/容器，发送到云上的云收集器中。服务推荐设备采集云控制器上的绑定信息和云收集器上的运行状态信息，根据得到的信息情况，获得平台上的所有应用分别使用的服务以及各个服务的运行状态，从而分析每个云应用的最佳服务。
可选地，云控制器上的绑定信息和云收集器上的运行状态信息也可以主动推送的方式推送给服务推荐设备，以使其根据具体的信息情况分析云应用的最佳服务。
S101获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
可选地，目标服务类型，可以通过分析目标应用的代码获得；
可选地，目标服务类型，可以通过分析目标应用的属性信息获得。
S102，根据该服务状态信息，获得与该目标服务类型匹配的服务，将该服务作为待推荐的服务；
可选地，可以根据策略配置推荐库的推荐策略信息，生成与云平台上各种服务相对应的推荐策略维度，该推荐策略配置库体现了推荐策略维度和服务类型的对应关系；
可选地，基于上述应用与服务的运行状态信息在生成的推荐策略维度下对该服务类型的每个服务进行统计分析，得到每个服务的统计结果；
可选地，对每个服务的统计结果进行比较计算，将各个服务的统计结果进行从高到低的顺序排列；
可选地，从统计分数最高的服务开始，从所有同类型的服务中筛选出至少 一个服务作为待推荐的服务；
在一个实施例中，如图1的虚线框所示，该方法还可以包括：
S103，将上述待推荐的服务推送给该应用的开发者；
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
为使本领域普通技术人员更为形象的理解本发明，本发明实施例对上述策略配置库进行介绍。如图2所示，本发明实施例提供一种策略配置库及其组织方式。在该推荐策略配置库中，配置了一些通用的与服务推荐相关的推荐策略维度和配置策略。
可选地，根据图2，推荐策略维度包括了服务类型的自身特性相关的推荐维度、平台通用的服务相关的推荐维度以及业界通用服务相关的推荐维度。当然可选地，还可以包括其它推荐策略维度，例如用户定制的推荐维度，特定的针对某一类服务的推荐维度等，本发明实施例并不做特别地限定。
可选地，在每一类推荐策略维度上，还包括了许多种具体的推荐维度。如图2所示，以自身特性相关的推荐策略维度来说，对于数据库(Data Base，简称DB)服务来说，可以包括读/写使用情况的推荐维度；对于地图(Map)服务来说，可以包括用户分布情况的推荐维度；对于日志(Log)服务来说，可以包括日志数据类型与数据量的推荐维度。
针对平台通用的推荐策略维度来说，可以包括该类服务的稳定性，用户使用量以及用户迁移情况等推荐维度。
针对业界通用的推荐策略维度来说，可以包括该类服务的套餐价格，新上线服务推广以及服务竞价排名等推荐维度。
可选地，每一个推荐策略都配置了服务推荐的权重系数。在具体地服务推荐过程中，获得具体地推荐策略和对应的权重系数，计算出最终推荐服务列表，将之作为推荐结果发送给需要服务推荐的应用的开发者。
比如说，数据库类型的服务，都有读写使用情况，用户分部情况，数据类型，稳定性，使用量等推荐策略。这个数据库服务可能有SQL数据库服务，也可能有MSQL数据库服务，也可能有Oracle数据库服务。每个数据库服务在上述推荐策略维度上的统计结果可能互不一样。根据应用的特点，针对每个服务，将每个推荐策略乘以一个权重系数，然后相比较，这个结果找出较优的或者符合用户需求的服务。
可选地，如图3所示，通过读取图2的推荐策略配置库，针对云平台上的各种服务类型配置生成了相应的推荐策略维度。
如图4所示，本发明实施例提供一种面向云平台应用的服务推荐方法，包括：
S200，获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
S201，获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
S202，根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
S203，根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
可选地，如图4中的虚线框所示，S203可以包括：
S2031，根据该服务状态信息，在该目标推荐策略维度中的每个维度下对该目标服务类型中每个服务进行服务状态统计，得到每个服务在该每个维度下的服务状态；
S2032，根据预设的评分标准，对该每个服务在该每个维度下的状态进行评分计算，得到每个服务在每个维度下的分数；
S2033，将该每个服务在每个维度下的分数乘以预先设置的权重并求和，得到每个服务在该目标推荐策略维度下的服务状态的匹配分数。
S204，对该每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务。
可选地，在一个实施例中，S204还可以替换地为以下步骤：
对该每个服务的服务状态的匹配分数按照从高到低的顺序排列；从分数最高的服务开始，在该目标服务类型中选择至少一个服务作为待推荐的服务。
可选地，在S203之后还可以包括：
S210，计算该目标应用从当前使用的服务迁移到该待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
此时，步骤S204中用到的匹配分数几位修正后的匹配分数。
可选地，如图4中的虚线框所示，该方法还可以包括：
S205，将该待推荐的服务推送给该目标应用的开发者。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
如图16所示，本发明实施例提供一种面向云平台应用的服务推荐方法，此实施例和图4对应的实施例不同的是，S202步骤，本实施例用S420和S430步骤替换S202步骤，其余步骤和图4对应的实施例相同。
该方法包括：
S400和S401分别同步骤S200和S201，在此不再赘述。
S420，根据推荐策略配置库，生成与云平台上各种服务类型相对应的推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
S430，根据S420中生成的与云平台上各种服务相对应的推荐策略维度，得到待统计推荐策略维度；该待统计推荐策略维度为该应用所需的服务类型相对应的推荐策略维度；
很好理解的是，当S420中生成了与云平台上各种服务相对应的推荐策略维度后，匹配出和目标服务类型行相同的服务类型对应的推荐策略维度，该推荐策略维度即为待统计推荐策略维度。
举例来说，目标服务类型为数据库服务，那么这时候可以从个中服务相对应的推荐策略维度中匹配出数据库服务所对应的推荐策略维度，将该该推荐策略维度作为待统计推荐策略维度。
S403，S404，S405和S410分别同S203，S204，S205和S210，在此不再赘述。
其中S403的具体实现方式同S203的具体实现方式，即S4031-S4033分别同S2031-S2033，在此不再赘述。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
在实际应用中，服务的推荐过程可以分为两种情况：
一面向针对初次部署应用的服务推荐；
二是面向已经运行一段时间的应用的服务推荐；下面对这种两种过程的具体处理流程进行介绍。
如图5所示，本发明实施例提供一种面向云平台应用的服务推荐方法，包括：
S300，根据策略配置推荐库的推荐策略信息，生成与云平台上各种服务类型相对应的推荐策略维度，该推荐策略配置库包括对应于各种服务类型的推荐策略维度；
以图2中介绍的策略配置推荐库为例，在步骤S300中，面向云平台应用 的服务推荐设备在初始化时，读取策略配置推荐中的各种推荐维度信息。以此生成与云平台上各种服务相对应的推荐策略维度。
如针对数据库服(DB)服务，服务推荐设备会读取读/写使用情况的推荐维度，用户分布情况的推荐策略，DB稳定性的推荐策略等，以此生成DB服务相应的推荐策略维度；
针对日志(Log)服务，服务推荐设备通过读取读取策略配置推荐中的各种推荐维度，确定Log服务相关的推荐策略维度，如：
Log服务的自身特性相关的推荐策略维度，这个维度下包括：日志数据类型；日志数据量；日志保存时间；日志查询功能等推荐维度；
Log服务的平台通用的推荐策略维度，这个维度下包括：Log服务稳定性，用户使用量，用户迁移量等推荐维度；
Log服务的业界通用的推荐策略维度，这个维度下包括：套餐价格，新上线服务，竞价排名等推荐维度；
可选地，如图3所示，经过步骤S300，通过读取图2的推荐策略配置库，针对云平台上的各种服务类型配置生成了相应的推荐策略维度。
例如，针对DB服务，其推荐策略维度包括数据库的读/写使用情况、稳定性、用户使用量以及用户迁移量等推荐策略维度。
可选地，如稳定性、套餐价格等推荐策略维度，是整个云平台公用的，但是针对不同的应用类型，其关注的重点不同，所以每个服务推荐策略上，这些方面的权重系数都是不同。也就是说，每种服务，都会有一个独立的推荐策略配置。
在应用初次部署，和动态分析调整时，使用的服务推荐策略也可能是不同的。在一个实施例中，主要的差别在于对于服务运行状态的监控信息，只有动态分析调整时才能获得得到，例如应用的日志数据类型和数据量，以及应用使用日志服务的方法等。在初次部署过程中，这些策略维度都会被忽略。相应的在初次部署的过程中，这些策略维度的权重系数为0。
S301，每隔设定的时间，获得云平台上的服务运行状态信息；
服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务 的运行状态。可选地，服务状态信息可以通过服务和应用的运行状态信息来表示；
可选地，服务和应用的运行状态信息包括服务和应用的绑定信息以及服务和应用的运行信息；
可选地，可以获得来自云控制器服务和应用的绑定信息和云收集器上的服务和应用的运行信息；
在一个实施例中，可以通过消息订阅的方式，接收来自云控制器、云收集器等信息来源的采样数据；在一实施例中，也可以主动查询并获得云控制器、云收集器等信息来源的采样数据；
比如对一个搜索应用来说，通过定期的获得与其绑定的服务的信息以及该应用和其绑定的服务的运行信息，可以分析出，与该搜索应用绑定的服务的一系列信息；比如，和该搜索应用绑定了数据库服务，通过步骤S301的分析，可以得到该数据库服务的读/写使用情况，稳定性以及总的用户使用量等。
S302，接收来自云平台的新部署应用通知消息；
可选地，该消息携带有该新部署应用的属性信息；
可选地，当云平台上新部署了一个应用时，云控制器会发送通知消息到服务推荐设备，该消息携带有该应用的属性信息，如应用所使用的语言栈以及使用框架等；
S303，获知该应用所需的服务类型；
可选地，面向云平台应用的服务推荐设备根据新应用的属性信息，比如所使用的语言栈以及使用框架等，可以获知该新应用所需要的潜在的服务类型。一般来说，新应用所需的服务会携带在其属性信息里。
比如一个搜索应用，通过其属性信息可以获知该搜索应用需要数据库服务。比如一个导航应用，通过其属性信息可以获知该导航应用需要地图服务。
可选地，还可以通过分析新应用的公开代码信息，来获知该应用所需的服务类型；
可选地，还可以通过接受新应用携带的附属消息，通过该附属消息或者该应用所需的服务类型。在一个实例中，应用的开发者可以将开发这个应用需要 的服务类型携带在改应用的附属消息中。
比如，开发者卡法一个导航应用需要用到地图类型的服务，此时开发者就可以在应用里携带一个附属消息，该附属消息里面携带有所需要的服务类型的消息，比如需要地图服务。开发者将该应用部署在云平台上的时候，服务推荐设备根据该附属消息就能或者该应用所需要的服务类型。
S304，根据S300中生成的与云平台上各种服务相对应的推荐策略维度，得到待统计推荐策略维度；该待统计推荐策略维度为该应用所需的服务类型相对应的推荐策略维度；
例如，如果S303通过该应用的属性信息获知该应用需要数据库服务，那么在S304中，就能根据S300的生成结果(如图3所示)，得到数据库服务的需要分析的推荐策略维度，例如在一个实施例中，该数据库服务的需要分析的推荐策略维度包括：数据库的读/写使用情况、稳定性、用户使用量以及用户迁移量。
需要说明的是，在本实施例中，先根据策略配置推荐库的推荐策略信息生成了与云平台上各种服务相对应的推荐策略维度。然后再通过获得新部署的应用所需的服务类型，去生成的各种服务相对应的推荐策略维度中找出和新部署的应用所需的服务类型对应的推荐策略维度。
当然很好理解的是，在另一个实施例中，也可以先获得新部署的应用所需要的服务类型，然后根据推荐策略配置库，生成与该服务类型对应的目标推荐策略维度，如前面步骤S202该的一样。
这两种实现方式均可，不做特别的限定。
S305，根据S301中获得到的云平台上的服务状态信息，获得该服务类型中每个服务在该待统计推荐策略维度的中每个维度的统计结果；
比如，在S304中已经得到该服务类型所需要统计分析的推荐策略维度包括：数据库的读/写使用情况、稳定性、用户使用量以及用户迁移量。
云平台上数据库服务类型有，MYSQL、ACCESS和BD2。那么根据S301中定期获得的服务相关信息，可以得到该数据库服务类型所包含的每个数据库服务的需要统计分析的推荐策略维度的每个维度的统计结果：
MYSQL服务：
数据库的读/写使用情况的维度为：50％读和50％写；
稳定性的维度为：高稳定性；
用户使用量的维度为：100次使用；
用户迁移量的维度为：50次用户迁出停止使用此服务，50次用户迁入使用此服务；
ACCESS服务：
数据库的读/写使用情况的维度为：60％读和40％写；
稳定性的维度为：中稳定性；
用户使用量的维度为：50次使用；
用户迁移量的维度为：100次用户迁出停止使用此服务，50次用户迁入使用此服务；
BD2服务：
数据库的读/写使用情况的维度为：40％读和60％写；
稳定性的维度为：高稳定性；
用户使用量的维度为：200次使用；
用户迁移量的维度为：50次用户迁出停止使用此服务，100次用户迁入使用此服务。
S306，根据S305中获得每个服务的每个维度的统计结果，根据预设的评分标准计算各维度的分数；
可选地，推荐设备在推荐的时候默认对于数据库类型的服务来说，各维度的预设评分标准是：
数据库的读/写使用情况的维度为：50％读和50％写；
稳定性的维度为：高稳定性；
用户使用量的维度为：100次使用；
用户迁移量的维度为：迁入用户量要大于迁出用户量；
当然，可选地，还可以有其它不同的预设评分标准，比如在一个实施例中，读/写使用情况的维度的预设评分标准为：读的使用频率要大于写的使用频率， 如：50％读和50％写。本发明不做特别地限定。
当一个维度满足相应的上述预设标准时，评分为10分。不满足的话根据不满足的程度以2分为一个单位扣除。
这样，根据上述预设标准，对于S305中获得的每个服务的每个维度的统计结果，评分如下：
MYSQL服务：
数据库的读/写使用情况的维度评分为：50％读和50％写—10分；
稳定性的维度评分为：高稳定性—10分；
用户使用量的维度评分为：100次使用—10分；
用户迁移量的维度评分为：50次用户迁出停止使用此服务，50次用户迁入使用此服务—8分；
总分：38分；
ACCESS服务：
数据库的读/写使用情况的维度评分为：60％读和40％写—8分；
稳定性的维度评分为：中稳定性—8分；
用户使用量的维度评分为：50次使用—6分；
用户迁移量的维度为：100次用户迁出停止使用此服务，50次用户迁入使用此服务—6分；
总分：28分；
BD2服务：
数据库的读/写使用情况的维度评分为：40％读和60％写—8分；
稳定性的维度评分为：高稳定性—10分；
用户使用量的维度评分为：200次使用—10分；
用户迁移量的维度评分为：50次用户迁出停止使用此服务，100次用户迁入使用此服务—10分。
总分：38分；
S307，将每个服务的每个维度的分数乘以预先设置的权重后，并将所有维度的分数求和，求和后按照总分数从高到低的顺序进行排名并得到服务推荐清 单，该服务推荐清单中包括至少一个待推荐的服务；
可选地，在初次推荐时预设的权重是均衡的，针对每个服务各维度的权重是一样的，比如对于上诉服务来说每个服务有4个推荐维度，那么每个维度的权重是1/4.
可选地，权重也可以是用户事先要求的，比如用户可能更注重稳定性，那么稳定性的权重就会适当的高一些，例如用户迁移量的维度的权重可以为2/5，其它剩余维度的权重均为1/5。
将S306中得到的每个维度的分数乘以预先获得的权重过后，根据各服务的总分将各服务按照从高到低的顺序进行排名得到最终的服务推荐清单。
可选地，如果每个维度的权重是1/4，那么根据S307，几个服务的排名为：MYSQL服务、BD2服务和ACCESS服务。
可选地，如果稳定性的权重可以为2/5，其它剩余维度的权重均为1/5，那么根据S307，几个服务的排名为：
BD2服务、MYSQL服务和ACCESS。
S308，在应用运行预设的时间后，重新对上述待统计推荐策略维度进行统计并获得更新分析结果；
应用运行一段时间后，所绑定服务的各项推荐维度的统计指标可能会发生变化，定期进行重新统计分析，并根据重新统计分析的结果，向用户推荐最合适的服务。
重新统计后会根据统计的结果，获得该应用的使用特点。例如对于一个搜索应用来说重新统计后，发现该搜索应用会频繁的读写数据库，读写比例达到7:3.
例如，假设用户选择了BD2服务，运行一段时间后，对该数据库服务类型的服务的维度进行重新统计得到：
MYSQL服务：
数据库的读/写使用情况的维度为：70％读和30％写；
稳定性的维度为：高稳定性；
用户使用量的维度为：80次使用；
用户迁移量的维度为：70次用户迁出停止使用此服务，60次用户迁入使用此服务；
ACCESS服务：
数据库的读/写使用情况的维度为：50％读和50％写；
稳定性的维度为：中稳定性；
用户使用量的维度为：80次使用；
用户迁移量的维度为：70次用户迁出停止使用此服务，100次用户迁入使用此服务；
BD2服务：
数据库的读/写使用情况的维度为：40％读和60％写；
稳定性的维度为：高稳定性；
用户使用量的维度为：100次使用；
用户迁移量的维度为：100次用户迁出停止使用此服务，100次用户迁入使用此服务。
S309，根据S308中获得的更新分析结果，对该应用绑定的服务同类型的每个服务进行推荐策略维度的分数进行重新计算以及权重的重新计算累加，得到更新后的服务推荐清单并发送；该更新后的服务推荐清单中包括至少一个待推荐的服务；
例如，根据S301中最新获得的服务和应用的运行信息，发现该应用在实际使用中更偏重于数据库的读操作特性，那么这时候推荐设备，就需要根据S308中的最新统计分析的结果对同类型的服务进行比较排序，得到更新后的推荐清单。
更为形象地，以上流程可以用图14这种方法的应用的情景示意图来形象展示。
可选地，每个推荐维度的权重系数可以是动态调整的，根据服务和应用的状态信息，可以降低或者提高一些服务维度的权重系数；
例如，维度中，如稳定性维度，如果服务出现过崩溃、死机等问题，则在一段时间内，其权重会降低，从而降低推荐的排名。
例如，通过统计发现该应用更注重数据库的读/写空能，那么数据库的读/写功能对应的权重就会升高，从而提升推荐的排名。
另外，在初次计算排序出结果后，需要根据排名，再计算一遍从当前服务迁移到目标服务的工作量(代码修改量/数据迁移工作)等信息。进行一次排序结果的修正。比如如果工作量很大，那么相应的服务排名就会降低。
例如，经过比较排序后，MYSQL服务由于更偏重于读操作特性，那么按照读操作特性的实际使用量，更新后的排名为MYSQL服务、ACCESS和BD2服务。显然，MYSQL比BD2服务更适合该应用。
可选地，在一个实施例中，服务推荐清单的发送，可以使用邮件/短信/消息推送/网页推送/客户端消息推送等方式发送给用户，用户根据推荐的信息，自行决定是否选择切换服务。
可选地，在一个实施例中，推荐设备也可选择一个得分最高的服务通过邮件/短信/消息推送/网页推送/客户端消息推送等方式发送给用户。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
可选地，除了上述推荐方法外，在一个实施例中，如图6所示，应用开发者可以使用白名单的方式选择推荐服务只在白名单列表中选择，也可以使用黑名单的方式，屏蔽一些开发者不愿意使用的服务。
这样，推荐设备回购恩局黑白名单的配置情况，选择对应用推荐白名单中的服务，或者去掉黑名单中的服务。
例如，推荐设备会在服务推荐清单中去掉黑名单总的服务，或者，在服务推荐清单中进一步选择出白名单中的服务。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经 过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。且进一步根据白买名单的设置，推荐的服务能更匹配开发者的个人喜好。
可选地，除了上述推荐方法外，在一个实施例中，如图7所示提供了另一种推荐策略动态配置的方案。该配置方案是区分角色的权重配置方法，针对不同的应用开发者，提供不同的服务维度的权重配置。
例如，对于普通开发者，存在竞价排名/新上线服务推广等维度的权重，在企业用户/VIP用户下可以适当减低甚至取消，以提供更针对应用本身特点的服务推荐。
在进行动态分析的时候，仅针对各个用户集合进行区别的计算。这种情况下，可以通过分布式的动态分析计算，也可以整体统一计算。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
可选地，除了上述推荐方法外，在一个实施例中，如图8所示提供了另一种推荐策略动态配置的方案。该方案是由应用开发者定制服务维度和权重，由应用开发者配置其部署的应用所使用的服务所关注的维度优先级。
开发者在选择使用某类型的服务时，可以配置服务特性维度的权重系数(平台级的维度权重如推广、价格等，不可配置。权重的总和是固定的)。根据开发者自身评估的应用需要服务的关注点，进行服务推荐的参考。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。并且通过由应用开发者定制服务维度和权重，由应用开发者配置其部署的应用所使用的服 务所关注的维度优先级，能更好的匹配开发者的应用开发习惯，使推荐方案更具有个性化。
与上述方法对应，如图9所示，本发明实施例公开了一种面向云平台应用的服务推荐设备，用以执行上述方法，该设备包括：
信息获得模块110，用于获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
服务类型获得模块120，用于获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
动态分析服务推荐模块130，用于根据该信息获得模块获110得的云平台的服务状态信息，获得与该服务类型获得模块120获得的目标服务类型匹配的服务，将该服务作为待推荐的服务。
在一个实施例中，如图9中的虚线框所示，该设备还可以包括：
推送模块140，用于将该待推荐的服务推送给该目标应用的开发者。
在一个实施例中，如图10所示，该动态分析服务推荐模块130包括：
生成单元1301，用于根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
匹配统计单元1302，用于根据该信息获得模块获得的云平台的服务状态信息，在生成单元生1301成的目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
确定单元1303，用于对该匹配统计单元1302得到的每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务。
在一个实施例中，如图10中的虚线框所示，该设备还可以包括：
修正模块150，用于计算该目标应用从当前使用的服务迁移到该待推荐服 务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
该排列单元1304具体用于，对修正模块150得到的每个服务的服务状态的修正后的匹配分数进行比较，将该每个服务的服务状态的匹配分数按照从高到低的顺序排列。
可选地，在一个实施例中，生成单元1301可以生成另一种推荐策略动态配置的方案。这是一种白名单/黑名单的方法，应用开发者可以使用白名单的方式选择推荐服务只在白名单列表中选择，也可以使用黑名单的方式，屏蔽一些开发者不愿意使用的服务，如前面图6所示。
相应地，如图12中的虚线框所示，该设备还可以包括：
服务选择模块160，用于根据生成单元1301的黑白名单的配置情况，选择白名单中的服务，或者去掉黑名单中的服务。
当然很好理解的是，在一个实施例中，服务选择模块160可以在匹配统计单元1302单元之前进行黑白名单的筛选；在此种情况下，服务选择模块先根据黑白名单进行服务的筛选，然后匹配统计单元1302在根据筛选出的服务进行匹配。如图12中的虚线连接线2。匹配的过程在上述匹配统计单元1302的描述中已经详细描述，在此不再赘述。
在另一个实施例中，服务选择模块160也可以在动态分析服务推荐模块130之后，推送模块140之前进行黑白名单的筛选，如图12中的虚线连接线1.在此种情况下，服务选择模块160最终选择出的服务即是待推荐的服务，并有推送模块140推送给目标应用的开发者。
需要说明的是，服务选择模块160和修正模块150可以共同设置在面向云平台应用的服务推荐设备中，也可以择一的设置。
在一个实施例中，如图11所示，该动态分析服务推荐模块130包括：
生成单元1301，用于根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
匹配统计单元1302，用于根据该信息获得模块获得的云平台的服务状态信息，在生成单元生1301成的目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
排列单元1304，用于对匹配统计单元1302得到的每个服务的服务状态的匹配分数按照从高到低的顺序排列；
选择单元1305，用于从分数最高的服务开始，按照该排列单元排列的顺序在该目标服务类型中选择至少一个服务作为待推荐的服务。
在一个实施例中，如图11中的虚线框所示，该设备还可以包括：
修正模块150，用于计算该目标应用从当前使用的服务迁移到该待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
该确定单元1303具体用于，对该修正模块150得到的修正后的匹配分数进行比较，将修正后的匹配分数最高的服务作为待推荐的服务。
可选地，在一个实施例中，生成单元1301可以生成另一种推荐策略动态配置的方案。这是一种白名单/黑名单的方法，应用开发者可以使用白名单的方式选择推荐服务只在白名单列表中选择，也可以使用黑名单的方式，屏蔽一些开发者不愿意使用的服务，如前面图6所示。
相应地，如图13中的虚线框所示，该设备还可以包括：
服务选择模块160，用于根据生成单元1301的黑白名单的配置情况，选择白名单中的服务，或者去掉黑名单中的服务。
当然很好理解的是，在一个实施例中，服务选择模块160可以在匹配统计单元1302单元之前进行黑白名单的筛选，如图13中的虚线连接线2；在此种情况下，服务选择模块先根据黑白名单进行服务的筛选，然后匹配统计单元1302在根据筛选出的服务进行匹配。匹配的过程在上述匹配统计单元1302的描述中已经详细描述，在此不再赘述。
在另一个实施例中，服务选择模块160也可以在动态分析服务推荐模块 130之后，推送模块140之前进行黑白名单的筛选，如图13中的虚线连接线1.在此种情况下，服务选择模块160最终选择出的服务即是待推荐的服务，并有推送模块140推送给目标应用的开发者。
需要说明的是，服务选择模块160和修正模块150可以共同设置在面向云平台应用的服务推荐设备中，也可以择一的设置。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
进一步，通过设置黑白名单以及由应用开发者定制服务维度和权重，由应用开发者配置其部署的应用所使用的服务所关注的维度优先级，能更好的匹配开发者的应用开发习惯，使推荐方案更具有个性化。
可选地，如图17-图20所示，以上图10-图13对应的设备实施例中的生成单元1301，可以替换的由下述两个单元来代替，其余模块单元不变：
策略维度生成单元1320，用于根据推荐策略配置库，生成与云平台上各种服务类型相对应的推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
待推荐策略维度单元1330，用于根据策略维度生成单元1320中生成的与云平台上各种服务相对应的推荐策略维度，得到待统计推荐策略维度；该待统计推荐策略维度为该应用所需的服务类型相对应的推荐策略维度。
鉴于其余模块单元的功能，其余模块和单元的功能和链接关系在图10-图13对应的设备实施例中已经详细说明，在此不再赘述。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹 配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。并且通过由应用开发者定制服务维度和权重，由应用开发者配置其部署的应用所使用的服务所关注的维度优先级，能更好的匹配开发者的应用开发习惯，使推荐方案更具有个性化。
如图15所示，本发明实施例提供一种面向云平台应用的服务推荐系统，该系统包括云平台及面向云平台应用的服务推荐设备，该面向云平台应用的服务推荐设备用于向该云平台上的应用推荐服务。
该面向云平台应用的服务推荐设备为一个相对独立的子系统，在云平台上的位置，如图所示。平台上的云应用与云服务，通过云控制器实现关系的绑定，而服务与应用的运行信息，通过其托管的虚拟机/容器，发送到云上的信息收集器中。服务推荐系统采集云控制器上的绑定信息和云收集器上的运行状态信息，根据得到的信息情况，分析云应用的最佳服务。
该面向云平台应用的服务推荐设备用于：
获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
根据该服务状态信息，获得与该目标服务类型匹配的服务，将该服务作为待推荐的服务；
可选地，该面向云平台应用的服务推荐设备具体用于：
获得云平台的服务状态信息，该服务状态信息包括该云平台上的所有应用分别使用的服务以及各个服务的运行状态；
获得目标服务类型，该目标服务类型为目标应用所需的服务类型，该目标应用为运行在该云平台上且需要进行服务推荐的应用；
根据推荐策略配置库，生成与该目标服务类型对应的目标推荐策略维度，该推荐策略配置库包括了推荐策略维度和服务类型的对应关系，该推荐策略维度包括了至少一种服务状态统计维度，该服务状态统计维度用于和该目标服务类型中的服务进行匹配程度统计；
根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数；该匹配分数反应该每个服务的运行状态的与该推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
对该每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务；或者，
对该每个服务的服务状态的匹配分数按照从高到低的顺序排列；从分数最高的服务开始，在该目标服务类型中选择至少一个服务作为待推荐的服务。
可选地，该面向云平台应用的服务推荐设备在用于根据该服务状态信息，在该目标推荐策略维度下对该目标服务类型中的每个服务进行服务状态的匹配分数统计，得到该每个服务的服务状态的匹配分数时，具体用于：
根据该服务状态信息，在该目标推荐策略维度中的每个维度下对该目标服务类型中每个服务进行服务状态统计，得到每个服务在该每个维度下的服务状态；
根据预设的评分标准，对该每个服务在该每个维度下的状态进行评分计算，得到每个服务在每个维度下的分数；
将该每个服务在每个维度下的分数乘以预先设置的权重并求和，得到每个服务在该目标推荐策略维度下的服务状态的匹配分数。
本发明实施例通过以上技术方案，根据云平台的服务状态信息获得与该目标服务类型匹配的服务，降低应用开发者学习评估使用服务的成本，使应用开发者及时获得到和待开发应用匹配的服务，从而提高了应用的开发效率，且经过动态分根据云平台的服务状态信息析推荐匹配的服务，使用的服务越来越匹配用户的使用场景特点，提高了推荐的服务和目标应用的匹配度。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，该的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，该的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
以上该仅为本发明的几个实施例，本领域的技术人员依据申请文件公开的可以对本发明进行各种改动或变型而不脱离本发明的精神和范围。

Claims (13)

  1. 一种面向云平台应用的服务推荐方法，其特征在于，所述方法包括：
    获得云平台的服务状态信息，所述服务状态信息包括所述云平台上的所有应用分别使用的服务以及各个服务的运行状态；
    获得目标服务类型，所述目标服务类型为目标应用所需的服务类型，所述目标应用为运行在所述云平台上且需要进行服务推荐的应用；
    根据所述服务状态信息，获得与所述目标服务类型匹配的服务，将所述服务作为待推荐的服务。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述服务状态信息，获得与所述目标服务类型匹配的服务，将所述服务作为待推荐的服务，包括：
    根据推荐策略配置库，生成与所述目标服务类型对应的目标推荐策略维度，所述推荐策略配置库包括了推荐策略维度和服务类型的对应关系，所述推荐策略维度包括了至少一种服务状态统计维度，所述服务状态统计维度用于和所述目标服务类型中的服务进行匹配程度统计；
    根据所述服务状态信息，在所述目标推荐策略维度下对所述目标服务类型中的每个服务进行服务状态的匹配分数统计，得到所述每个服务的服务状态的匹配分数；所述匹配分数反应所述每个服务的运行状态的与所述推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
    对所述每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务。
  3. 根据权利要求1所述的方法，其特征在于，所述根据所述服务状态信息，获得与所述目标服务类型匹配的服务，将所述服务作为待推荐的服务，包括：
    根据推荐策略配置库，生成与所述目标服务类型对应的目标推荐策略维度，所述推荐策略配置库包括了推荐策略维度和服务类型的对应关系，所述推荐策略维度包括了至少一种服务状态的统计维度，所述服务状态统计维度用于 和所述目标服务类型中的服务进行匹配程度统计；
    根据所述服务状态信息，在所述目标推荐策略维度下对所述目标服务类型中的每个服务进行服务状态的匹配分数统计，得到所述每个服务的服务状态的匹配分数；所述匹配分数反应所述每个服务的运行状态的与所述推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
    对所述每个服务的服务状态的匹配分数按照从高到低的顺序排列；
    从分数最高的服务开始，在所述目标服务类型中选择至少一个服务作为待推荐的服务。
  4. 根据权利要求2或3所述的服务推荐方法，其特征在于，所述根据所述服务状态信息，在所述目标推荐策略维度下对所述目标服务类型中的每个服务进行服务状态的匹配分数统计，得到所述每个服务的服务状态的匹配分数，包括：
    根据所述服务状态信息，在所述目标推荐策略维度中的每个维度下对所述目标服务类型中每个服务进行服务状态统计，得到每个服务在所述每个维度下的服务状态；
    根据预设的评分标准，对所述每个服务在所述每个维度下的状态进行评分计算，得到每个服务在每个维度下的分数；
    将所述每个服务在每个维度下的分数乘以预先设置的权重并求和，得到每个服务在所述目标推荐策略维度下的服务状态的匹配分数。
  5. 根据权利要求2或3所述的服务推荐方法，其特征在于，所述根据所述服务状态信息，在所述目标推荐策略维度下对所述目标服务类型中的每个服务进行服务状态的匹配分数统计，得到所述每个服务的服务状态的匹配分数之后，所述方法还包括：
    计算所述目标应用从当前使用的服务迁移到所述待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
    所述对所述每个服务的服务状态的匹配分数进行比较，将所述每个服务的 服务状态的匹配分数按照从高到低的顺序排列，包括：
    对所述每个服务的服务状态的修正后的匹配分数进行比较，将所述每个服务的服务状态的匹配分数按照从高到低的顺序排列；
  6. 根据权利要求1-5任一项所述的服务推荐方法，其特征在于，所述方法还包括：
    将所述待推荐的服务推送给所述目标应用的开发者。
  7. 一种面向云平台应用的服务推荐设备，其特征在于，所述设备包括：
    信息获得模块，用于获得云平台的服务状态信息，所述服务状态信息包括所述云平台上的所有应用分别使用的服务以及各个服务的运行状态；
    服务类型获得模块，用于获得目标服务类型，所述目标服务类型为目标应用所需的服务类型，所述目标应用为运行在所述云平台上且需要进行服务推荐的应用；
    动态分析服务推荐模块，用于根据所述信息获得模块获得的云平台的服务状态信息，获得与所述服务类型获得模块获得的目标服务类型匹配的服务，将所述服务作为待推荐的服务。
  8. 根据权利要求7所述的服务推荐设备，其特征在于，所述动态分析服务推荐模块包括：
    生成单元，用于根据推荐策略配置库，生成与所述目标服务类型对应的目标推荐策略维度，所述推荐策略配置库包括了推荐策略维度和服务类型的对应关系，所述推荐策略维度包括了至少一种服务状态统计维度，所述服务状态统计维度用于和所述目标服务类型中的服务进行匹配程度统计；
    匹配统计单元，用于根据所述信息获得模块获得的云平台的服务状态信息，在所述生成单元生成的目标推荐策略维度下对所述目标服务类型中的每个服务进行服务状态的匹配分数统计，得到所述每个服务的服务状态的匹配分数；所述匹配分数反应所述每个服务的运行状态的与所述推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
    确定单元，用于对所述匹配统计单元得到的每个服务的服务状态的匹配分数进行比较，将匹配分数最高的服务作为待推荐的服务。
  9. 根据权利要求7所述的方法，其特征在于，所述动态分析服务推荐模块包括：
    生成单元，用于根据推荐策略配置库，生成与所述目标服务类型对应的目标推荐策略维度，所述推荐策略配置库包括了推荐策略维度和服务类型的对应关系，所述推荐策略维度包括了至少一种服务状态统计维度，所述服务状态统计维度用于和所述目标服务类型中的服务进行匹配程度统计；
    匹配统计单元，用于根据所述信息获得模块获得的云平台的服务状态信息，在所述生成单元生成的目标推荐策略维度下对所述目标服务类型中的每个服务进行服务状态的匹配分数统计，得到所述每个服务的服务状态的匹配分数；所述匹配分数反应所述每个服务的运行状态的与所述推荐策略维度包括的至少一种服务状态的统计维度的匹配程度，其中匹配分数越高匹配程度越高；
    排列单元，用于对所述每个服务的服务状态的匹配分数按照从高到低的顺序排列；
    选择单元，用于从分数最高的服务开始，按照所述排列单元排列的顺序在所述目标服务类型中选择至少一个服务作为待推荐的服务。
  10. 根据权利要求8所述的方法，其特征在于，所述设备还包括：
    修正模块，用于计算所述目标应用从当前使用的服务迁移到所述待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
    所述排列单元具体用于，对所述每个服务的服务状态的修正后的匹配分数进行比较，将所述每个服务的服务状态的匹配分数按照从高到低的顺序排列。
  11. 根据权利要求9所述的方法，其特征在于，所述设备还包括：
    修正模块，用于计算所述目标应用从当前使用的服务迁移到所述待推荐服务的工作量，根据工作量的大小对每个服务的服务状态的匹配分数进行修正，得到修正后的匹配分数；
    所述确定单元具体用于，对所述修正模块得到的修正后的匹配分数进行比较，将修正后的匹配分数最高的服务作为待推荐的服务。
  12. 根据权利要求7-11任一项所述的服务推荐设备，其特征在于，所述 设备还包括：
    推送模块，用于将所述待推荐的服务推送给所述目标应用的开发者。
  13. 一种面向云平台应用的服务推荐系统，其特征在于，所述系统包括云平台及面向云平台应用的服务推荐设备，所述面向云平台应用的服务推荐设备用于向所述云平台上的应用推荐服务，所述面向云平台应用的服务推荐设备如权利要求7-12任一项所述。
PCT/CN2016/070057 2015-01-05 2016-01-04 面向云平台应用的服务推荐方法、设备及系统 Ceased WO2016110234A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP16734894.5A EP3232338A4 (en) 2015-01-05 2016-01-04 Cloud platform application-oriented service recommendation method, device and system
US15/641,137 US20170300497A1 (en) 2015-01-05 2017-07-03 Cloud Platform Application-Orientated Service Recommendation Method, Device, and System

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510004844.8A CN104615661B (zh) 2015-01-05 2015-01-05 面向云平台应用的服务推荐方法、设备及系统
CN201510004844.8 2015-01-05

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/641,137 Continuation US20170300497A1 (en) 2015-01-05 2017-07-03 Cloud Platform Application-Orientated Service Recommendation Method, Device, and System

Publications (1)

Publication Number Publication Date
WO2016110234A1 true WO2016110234A1 (zh) 2016-07-14

Family

ID=53150104

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/070057 Ceased WO2016110234A1 (zh) 2015-01-05 2016-01-04 面向云平台应用的服务推荐方法、设备及系统

Country Status (4)

Country Link
US (1) US20170300497A1 (zh)
EP (1) EP3232338A4 (zh)
CN (1) CN104615661B (zh)
WO (1) WO2016110234A1 (zh)

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104615661B (zh) * 2015-01-05 2019-02-19 华为技术有限公司 面向云平台应用的服务推荐方法、设备及系统
CN104980517B (zh) * 2015-06-26 2018-04-06 深圳市腾讯计算机系统有限公司 基于快照的集群感知系统、方法和装置
US10901812B2 (en) * 2017-09-18 2021-01-26 Rapyuta Robotics Co., Ltd. Managing communication between cloud and heterogeneous devices across networks
US10972404B2 (en) * 2017-09-18 2021-04-06 Rapyuta Robotics Co., Ltd. Generate deploy and provision a customized cloud device application using a software service and device store
US20190158367A1 (en) * 2017-11-21 2019-05-23 Hewlett Packard Enterprise Development Lp Selection of cloud service providers to host applications
CN108230162B (zh) * 2017-12-29 2022-01-11 泰康保险集团股份有限公司 保险服务推荐的方法、装置、存储介质及电子设备
CN108596353A (zh) * 2018-05-03 2018-09-28 苏州工业园区服务外包职业学院 共享泊位预定方法和平台
CN109582863B (zh) * 2018-11-19 2020-08-04 珠海格力电器股份有限公司 一种推荐方法及服务器
US20210304158A1 (en) * 2020-03-30 2021-09-30 Bank Of America Corporation System for implementing a resource evaluation engine within a technical environment
CN111552571B (zh) * 2020-04-30 2024-05-24 深信服科技股份有限公司 应用反馈方法、计算机设备及计算机存储介质
CN113052565A (zh) * 2020-10-17 2021-06-29 严怀华 基于云服务的智慧社区业务信息处理方法及云服务设备
CN114640706B (zh) * 2020-11-30 2023-09-12 华为技术有限公司 数据传输方法、电子设备及计算机可读存储介质
CN112800333B (zh) * 2021-02-04 2023-10-27 北京信息科技大学 企业用户服务的推荐方法、装置、设备及存储介质
US20230004938A1 (en) * 2021-07-01 2023-01-05 Google Llc Detecting Inactive Projects Based On Usage Signals And Machine Learning
CN113781235B (zh) * 2021-09-01 2024-07-26 微民保险代理有限公司 一种数据处理方法、装置、计算机设备以及存储介质
US11562043B1 (en) * 2021-10-29 2023-01-24 Shopify Inc. System and method for rendering webpage code to dynamically disable an element of template code
CN115174158B (zh) * 2022-06-14 2024-04-16 阿里云计算有限公司 基于多云管理平台的云产品配置检查方法
CN116308170B (zh) * 2023-03-21 2023-10-13 北京中关村软件园孵化服务有限公司 一种应用于数字孵化服务平台的管理方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102917077A (zh) * 2012-11-20 2013-02-06 无锡城市云计算中心有限公司 云计算系统中的资源分配方法
CN103051730A (zh) * 2013-01-15 2013-04-17 合肥工业大学 一种云计算商务环境下多源信息服务资源分配系统及IA-Min分配方法
US20130166646A1 (en) * 2011-12-27 2013-06-27 Nokia Corporation Method and apparatus for providing social network services based on connectivity information
CN104615661A (zh) * 2015-01-05 2015-05-13 华为技术有限公司 面向云平台应用的服务推荐方法、设备及系统

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9489647B2 (en) * 2008-06-19 2016-11-08 Csc Agility Platform, Inc. System and method for a cloud computing abstraction with self-service portal for publishing resources
CN102255933B (zh) * 2010-05-20 2016-03-30 中兴通讯股份有限公司 云服务中介、云计算方法及云系统
US9140188B2 (en) * 2011-10-25 2015-09-22 United Technologies Corporation Gas turbine engine with intercooling turbine section
CN102523247B (zh) * 2011-11-24 2014-09-24 合肥工业大学 一种基于多属性匹配的云服务推荐方法及装置
US9967159B2 (en) * 2012-01-31 2018-05-08 Infosys Limited Systems and methods for providing decision time brokerage in a hybrid cloud ecosystem
CN102880501B (zh) * 2012-07-24 2016-05-25 北京奇虎科技有限公司 应用推荐的实现方法、装置和系统
CN102904824B (zh) * 2012-09-25 2015-04-29 中国联合网络通信集团有限公司 服务提供实体选择方法及装置
US9582780B1 (en) * 2013-01-30 2017-02-28 Skyhigh Networks, Inc. Cloud service usage risk assessment
US9552232B2 (en) * 2013-03-12 2017-01-24 Xerox Corporation System and process to recommend cloud service cloud configuration based on service similarity
US20140278807A1 (en) * 2013-03-15 2014-09-18 Cloudamize, Inc. Cloud service optimization for cost, performance and configuration
US9813318B2 (en) * 2013-03-15 2017-11-07 International Business Machines Corporation Assessment of best fit cloud deployment infrastructures
US9571516B1 (en) * 2013-11-08 2017-02-14 Skyhigh Networks, Inc. Cloud service usage monitoring system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130166646A1 (en) * 2011-12-27 2013-06-27 Nokia Corporation Method and apparatus for providing social network services based on connectivity information
CN102917077A (zh) * 2012-11-20 2013-02-06 无锡城市云计算中心有限公司 云计算系统中的资源分配方法
CN103051730A (zh) * 2013-01-15 2013-04-17 合肥工业大学 一种云计算商务环境下多源信息服务资源分配系统及IA-Min分配方法
CN104615661A (zh) * 2015-01-05 2015-05-13 华为技术有限公司 面向云平台应用的服务推荐方法、设备及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3232338A4 *

Also Published As

Publication number Publication date
EP3232338A1 (en) 2017-10-18
CN104615661A (zh) 2015-05-13
EP3232338A4 (en) 2017-10-18
US20170300497A1 (en) 2017-10-19
CN104615661B (zh) 2019-02-19

Similar Documents

Publication Publication Date Title
WO2016110234A1 (zh) 面向云平台应用的服务推荐方法、设备及系统
US11055646B2 (en) Automated ticket resolution
CN106445652B (zh) 用于智能云计划和停用的方法和系统
US20210304074A1 (en) Method and system for target based hyper-parameter tuning
KR102409347B1 (ko) 정책 기반 자원 관리 및 할당 시스템
US10719769B2 (en) Systems and methods for generating and communicating application recommendations at uninstall time
US20180276689A1 (en) Analyzing big data to determine a data plan
JP2018515844A (ja) データ処理方法及びシステム
US12400246B2 (en) Facilitating responding to multiple product or service reviews associated with multiple sources
US12271906B2 (en) System and method for managing issues through resource optimization
CN119718362A (zh) 用于虚拟系统升级的方法、设备和计算机程序产品
CN113656046B (zh) 一种应用部署方法和装置
US20250086153A1 (en) Auto recognition of big data computation engine for optimized query runs on cloud platforms
US20240211829A1 (en) System and method for managing issues using skill matching
Chatziprimou et al. Surrogate-assisted online optimisation of cloud iaas configurations
US20240211831A1 (en) System and method for managing issues through proficiency analysis
CN103914573A (zh) 一种测点迁移方法及装置
US11907230B1 (en) System and method for distributed management of hardware based on intent
US20250139203A1 (en) Selection and use of blueprints in device management
US20260086516A1 (en) Distributed system management with predictive control
KR101189808B1 (ko) 광고 시스템 및 그의 광고 제어 방법
US20250356367A1 (en) Flexible scoring mechanism for enabling sustainable deployments in multicloud
KR20260049935A (ko) 학습 모델을 관리하는 방법, 전자 장치 및 기록 매체
WO2024254022A1 (en) Large artificial intelligence model prediction and capacity
CN115509728A (zh) 一种云服务解决方案提供方法及装置

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: 16734894

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2016734894

Country of ref document: EP