CN117149405A - Resource allocation method and device - Google Patents
Resource allocation method and device Download PDFInfo
- Publication number
- CN117149405A CN117149405A CN202310908333.3A CN202310908333A CN117149405A CN 117149405 A CN117149405 A CN 117149405A CN 202310908333 A CN202310908333 A CN 202310908333A CN 117149405 A CN117149405 A CN 117149405A
- Authority
- CN
- China
- Prior art keywords
- service
- scene
- processing system
- information processing
- information
- 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.)
- Pending
Links
- 238000013468 resource allocation Methods 0.000 title claims abstract description 123
- 238000000034 method Methods 0.000 title claims abstract description 91
- 230000010365 information processing Effects 0.000 claims abstract description 244
- 238000004364 calculation method Methods 0.000 claims abstract description 84
- 230000008901 benefit Effects 0.000 claims description 82
- 230000006870 function Effects 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 18
- 238000005259 measurement Methods 0.000 claims description 8
- 238000009826 distribution Methods 0.000 claims description 7
- 230000004044 response Effects 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 7
- 238000012384 transportation and delivery Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000005457 optimization Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000005553 drilling Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000009530 blood pressure measurement Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0247—Calculate past, present or future revenues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/503—Resource availability
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a resource allocation method and a device, wherein the resource allocation method comprises the following steps: obtaining target flow accepted by a service scene in preset time and computing resource data usable by a service information processing system, wherein the service information processing system is used for throwing service information which can be processed by a user into the service scene in a user service system; according to the target flow accepted by the service scene in the preset time and the calculation resource data usable by the service information processing system, obtaining calculation resource data required to be distributed for a service module in the service information processing system when the income data obtained by the service information processing system for throwing service information into the service scene in the user service system in the preset time meets the preset income condition, wherein the service module is used for determining service information to be thrown for a user accessing the service scene. The method can improve the use efficiency of the computing resources.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a resource allocation method, a resource allocation device, an electronic device, and a computer readable storage medium.
Background
With the continuous development of advertisement showing service, the flow scene accessed by the online advertisement engine system is more and more, and the advertisement putting strategy is more and more complex, so that the machine resource amount used by the online advertisement engine system is continuously increased, and the demand on calculation power is continuously increased.
Due to the limited supply of machine resources (also known as computing resources) and the increased cost of computing power, it is becoming increasingly important in such contexts how to fully utilize currently existing machine resources and to increase the efficiency of use of machine resources.
Disclosure of Invention
The application provides a resource allocation method to solve the problem of low utilization efficiency of machine resources in the existing display advertisement service. The application further provides a resource allocation apparatus, an electronic device and a computer readable storage medium.
The application provides a resource allocation method, which comprises the following steps: obtaining target flow accepted by a service scene in a preset time, wherein the service scene is a service scene for providing service information for a user in a user service system; acquiring computing resource data usable by a service information processing system, wherein the service information processing system is used for throwing service information which can be processed by a user into a service scene in the user service system; according to the target flow accepted by the service scene in the preset time and the calculation resource data usable by the service information processing system, obtaining calculation resource data required to be distributed for a service module in the service information processing system when the income data obtained by the service information processing system for throwing service information into the service scene in the user service system in the preset time meets the preset income condition, wherein the service module is used for determining service information to be thrown for a user accessing the service scene.
Optionally, the method further comprises: obtaining unit flow rate gain data corresponding to the service scene, wherein the unit flow rate gain data is gain data obtained by putting service information into the service scene receiving unit flow rate in the preset time by the service information processing system, and the unit flow rate is a flow rate measurement standard; obtaining benefit data obtained by putting service information into a service scene in the user service system by the service information processing system in the preset time according to the flow accepted by the service scene in the preset time and the unit flow benefit data corresponding to the service scene.
Optionally, the obtaining unit flow benefit data corresponding to the service scenario includes: and obtaining unit flow gain data corresponding to the service scene according to the grade of the service strategy executed when the service module determines the service information to be put for the user accessing the service scene, wherein the grade is used for representing the condition grade of the computing resource data used when the service strategy is executed.
Optionally, the obtaining, according to the target flow received by the service scenario in the preset time and the computing resource data usable by the service information processing system, computing resource data to be allocated to a service module in the service information processing system when revenue data obtained by the service information processing system in the preset time for delivering service information to the service scenario in the user service system meets a preset revenue condition includes: obtaining calculation resource data required to be distributed for a service module in the service information processing system when the benefit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset benefit condition by utilizing a linear programming model; the linear programming model comprises an objective function and constraint conditions, wherein the objective function is a calculation function of revenue data obtained by putting service information into a service scene in the user service system by the service information processing system within the preset time, solution variables of the objective function comprise flow accepted by the service scene within the preset time, a grade of a service strategy executed by the service module when determining service information to be put for a user accessing the service scene, and calculation resource data required to be allocated for the service module in the service information processing system, and the constraint conditions comprise the objective flow accepted by the service scene within the preset time, calculation resource data available to the service information processing system and time available to the service module when determining the service information to be put for the user accessing the service scene.
Optionally, the method further comprises: obtaining, according to a target flow received by the service scene in a preset time and calculation resource data usable by the service information processing system, a reference level of a service policy executed when the service scene receives a reference flow and/or the service module determines service information to be released for a user accessing the service scene when revenue data obtained by the service information processing system releasing service information to the service scene in the user service system in the preset time meets a preset revenue condition.
Optionally, the method further comprises: according to the calculation resource data required to be distributed for a service module in the service information processing system, the reference flow accepted by the service scene in the preset time and the reference grade of a service strategy executed when the service module determines the service information to be put for a user accessing the service scene when the service information processing system puts the service information into the service scene in the user service system in the preset time and the obtained profit data meet the preset profit condition, the calculation resource data usable by the service module, the flow accepted by the service scene in the preset time and the grade of the executable service strategy when the service module determines the service information to be put for the user accessing the service scene are adjusted.
Optionally, the user service system has a plurality of service scenes, and the service information processing system comprises a plurality of service modules; the obtaining the target flow accepted by the service scene in the preset time comprises the following steps: obtaining target flow respectively accepted by the plurality of service scenes in the preset time; the obtaining, according to the target flow received by the service scenario in the preset time and the computing resource data usable by the service information processing system, computing resource data to be allocated to a service module in the service information processing system when revenue data obtained by the service information processing system delivering service information to the service scenario in the user service system in the preset time meets a preset revenue condition includes: according to the target flow respectively accepted by the service scenes and the computing resource data usable by the service information processing system in the preset time, obtaining computing resource data respectively required to be distributed for a plurality of service modules in the service information processing system when the profit data obtained by putting service information into the service scenes in the user service system in the preset time meets the preset profit condition.
Optionally, the method further comprises: obtaining unit flow gain data corresponding to the service scenes respectively, wherein the unit flow gain data is gain data obtained by putting service information into the service scene receiving the unit flow in the preset time by the service information processing system, and the unit flow is a flow measurement standard; obtaining the income data obtained by putting service information into a plurality of service scenes in the user service system by the service information processing system in the preset time according to the flow respectively accepted by the plurality of service scenes in the preset time and the unit flow income data respectively corresponding to the plurality of service scenes.
Optionally, the obtaining unit flow revenue data corresponding to each of the plurality of service scenarios includes: and obtaining unit flow gain data corresponding to the service scenes respectively according to the combination grades of the service policies executed by the service modules when the service information to be put is determined for the users accessing the service scenes, wherein the combination grades are the combination grades determined according to the grades of the service policies executed by the service modules respectively, and the grades are used for representing the condition grades of the computing resource data used when the service policies are executed.
Optionally, the obtaining the unit flow revenue data corresponding to the multiple service scenes respectively according to the combination level of the service policies executed when the multiple service modules determine the service information to be put for the user accessing the service scenes includes: obtaining reference revenue data obtained by putting the service information to be put determined by the plurality of service modules into the service scene receiving unit flow when the computing resource data used by the plurality of service modules for executing the service policies of the reference combination level meets the preset computing resource conditions through a heuristic iteration method; and determining the reference revenue data corresponding to the service scenes as unit flow revenue data corresponding to the service scenes.
Optionally, the method further comprises: obtaining, according to target flows respectively accepted by the multiple service scenes in the preset time and computing resource data usable by the service information processing system, reference combination levels of service policies executed by the multiple service scenes when the profit data obtained by the service information processing system for delivering service information to the multiple service scenes in the user service system in the preset time meets preset profit conditions, wherein the reference flows respectively accepted by the multiple service scenes in the preset time and/or the reference combination levels of service policies executed by the multiple service modules when determining service information to be delivered for users accessing the service scenes.
Optionally, the method further comprises: according to the calculation resource data required to be distributed for a plurality of service modules in the service information processing system respectively when the profit data obtained by putting service information into a plurality of service scenes in the user service system within the preset time meets the preset profit condition, the reference flow respectively accepted by the plurality of service scenes within the preset time and the reference combination grade of the service strategy executed when the plurality of service modules determine the service information to be put for a user accessing the service scenes, the calculation resource data respectively usable by the plurality of service modules, the flow respectively accepted by the plurality of service scenes within the preset time and the grade of the service strategy respectively executable when the plurality of service modules determine the service information to be put for the user accessing the service scenes are adjusted.
The application also provides a resource allocation device, which comprises: the system comprises a first target flow obtaining unit, a second target flow obtaining unit and a control unit, wherein the first target flow obtaining unit is used for obtaining target flow accepted by a service scene in a preset time, and the service scene is a service scene for providing service information for a user in a user service system; the first computing resource data obtaining unit is used for obtaining computing resource data usable by a service information processing system, and the service information processing system is used for throwing service information which can be processed by a user into a service scene in the user service system; the first computing resource allocation unit is used for obtaining computing resource data required to be allocated for a service module in the service information processing system when the benefit data obtained by putting service information into the service scene in the user service system in the preset time meets a preset benefit condition according to the target flow accepted by the service scene in the preset time and the computing resource data usable by the service information processing system, wherein the service module is used for determining service information to be put into for a user accessing the service scene.
The application also provides a resource allocation method, which comprises the following steps: obtaining scene flow information which is sent by a client and used for representing target flow accepted by a service scene in preset time and system resource information which is used for representing computing resource data which can be used by a service information processing system, wherein the service scene is a service scene which provides service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system; obtaining calculation resource data required to be distributed for a service module in the service information processing system when the income data obtained by putting service information into a service scene in the user service system in the preset time meets a preset income condition according to the scene flow information and the system resource information, wherein the service module is used for determining service information to be put into for a user accessing the service scene; and returning resource allocation information for indicating calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition to the client.
The application also provides a resource allocation device, which comprises: the system comprises a first information obtaining unit, a second information obtaining unit and a service information processing unit, wherein the first information obtaining unit is used for obtaining scene flow information which is sent by a client and used for representing target flow accepted by a service scene in preset time, and system resource information which is used for representing calculation resource data which can be used by the service information processing system, the service scene is a service scene which provides service information for a user in a user service system, and the service information processing system is used for throwing the service information which can be processed by the user into the service scene in the user service system; the second computing resource allocation unit is used for obtaining computing resource data required to be allocated for a service module in the service information processing system when the benefit data obtained by putting the service information into the service scene in the user service system in the preset time meets a preset benefit condition according to the scene flow information and the system resource information, and the service module is used for determining service information to be put for a user accessing the service scene; and the first information sending unit is used for sending resource allocation information for representing calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition to the client.
The application also provides a resource allocation method, which comprises the following steps: displaying an information setting interface for setting target flow accepted by a service scene in a preset time and computing resource data usable by a service information processing system, wherein the service scene is a service scene for providing service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system; in response to detecting the target flow accepted by the service scene set by the information setting interface in the preset time and the computing resource data usable by the service information processing system, obtaining scene flow information for representing the target flow accepted by the service scene in the preset time and system resource information for representing the computing resource data usable by the service information processing system; the scene flow information and the system resource information are sent to a server; acquiring resource allocation information which is returned by the service end and used for representing calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition; and according to the resource allocation information, displaying the calculation resource data required to be allocated for the service module in the service information processing system when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset profit condition.
The application also provides a resource allocation device, which comprises: the system comprises a first display unit, a second display unit and a service information processing system, wherein the first display unit is used for displaying an information setting interface for setting target flow accepted by a service scene in preset time and computing resource data usable by the service information processing system, the service scene is a service scene for providing service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system; a second information obtaining unit configured to obtain, in response to detection of a target flow rate received by a service scenario set through the information setting interface within a preset time and calculation resource data usable by a service information processing system, scenario flow rate information for representing the target flow rate received by the service scenario within the preset time, and system resource information for representing the calculation resource data usable by the service information processing system; the second information sending unit is used for sending the scene flow information and the system resource information to a server; a third information obtaining unit, configured to obtain resource allocation information returned by the server, where the resource allocation information is used to indicate computing resource data that needs to be allocated to a service module in the service information processing system when revenue data obtained by putting service information into a service scenario in the user service system in the preset time meets a preset revenue condition; and the second display unit is used for displaying the calculation resource data required to be distributed for the service module in the service information processing system when the benefit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset benefit condition according to the resource distribution information.
The application also provides a resource allocation method, which comprises the following steps: obtaining the target flow received by a target scene in a preset time; obtaining computing resource data usable by an information processing system, wherein the information processing system is used for throwing information into the target scene; according to the target flow received by the target scene in the preset time and the calculation resource data usable by the information processing system, when the income data obtained by the information processing system for throwing information into the target scene in the preset time meets the preset income condition, calculating resource data required to be distributed for a processing module in the information processing system, wherein the processing module is used for determining target information for a user accessing the target scene.
The application also provides a resource allocation device, which comprises: the second target flow obtaining unit is used for obtaining the target flow received by the target scene in the preset time; the second computing resource data obtaining unit is used for obtaining computing resource data usable by an information processing system, and the information processing system is used for throwing information into the target scene; the third computing resource allocation unit is used for obtaining computing resource data required to be allocated to a processing module in the information processing system when the benefit data obtained by the information processing system throwing information into the target scene in the preset time meets the preset benefit condition according to the target flow received by the target scene in the preset time and the computing resource data usable by the information processing system, and the processing module is used for determining target information for a user accessing the target scene.
The application also provides an electronic device, which comprises a processor and a memory; the memory is used for storing programs and data, and the processor calls the programs stored in the memory to execute the resource allocation method.
The present application also provides a storage medium storing a program and data, the program being executed by a processor for implementing the above-described resource allocation method.
Compared with the prior art, the application has the following advantages:
according to the resource allocation method provided by the application, the calculation resource data which can be used by the service information processing system and is carried by the service scene in the preset time is obtained, and the calculation resource data which is required to be allocated for the service module in the service information processing system is obtained when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset profit condition according to the target flow carried by the service scene in the preset time and the calculation resource data which can be used by the service information processing system. The method aims at providing the service information processing system with the service scene in the user service system to obtain the benefit data meeting the preset benefit condition, and takes the target flow accepted by the service scene in the preset time and the calculation resource data available to the service information processing system as the constraint condition to obtain the calculation resource data required to be distributed for the service module in the service information processing system, namely the distribution result of the calculation resource data.
Drawings
Fig. 1 is a schematic diagram of a business scenario in the present application.
Fig. 2 is a flow chart of a method for implementing resource allocation in the present application.
Fig. 3 is a schematic diagram of an information setting interface in the present application.
Fig. 4 is a schematic diagram of a resource allocation result display interface according to the present application.
Fig. 5 is a flowchart of a resource allocation method according to a first embodiment of the present application.
Fig. 6 is a schematic diagram of a resource allocation apparatus according to a second embodiment of the present application.
Fig. 7 is a flowchart of a resource allocation method according to a third embodiment of the present application.
Fig. 8 is a schematic diagram of a resource allocation device according to a fourth embodiment of the present application.
Fig. 9 is a flowchart of a resource allocation method according to a fifth embodiment of the present application.
Fig. 10 is a schematic diagram of a resource allocation device according to a sixth embodiment of the present application.
Fig. 11 is a flowchart of a resource allocation method according to a seventh embodiment of the present application.
Fig. 12 is a schematic diagram of a resource allocation device according to an eighth embodiment of the present application.
Fig. 13 is a schematic view of an electronic device according to a ninth embodiment of the present application.
Detailed Description
The application will be described in further detail with reference to the drawings and the detailed description, in order to make the objects, advantages and features of the application more apparent. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, the application may be embodied in many other forms than those described herein and similarly practiced by those skilled in the art without departing from the spirit or scope of the application, which is therefore not limited to the specific embodiments disclosed below.
It should be noted that in the description of the present application, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying any particular order or sequence. The specific meaning of the above terms in the present application will be understood in specific cases by those of ordinary skill in the art. Furthermore, in the description of the present application, the term "plurality" means two or more, unless otherwise indicated. The term "and/or" describes an association relationship of associated objects, meaning that there may be three relationships, e.g., a and/or B, which may represent: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship. The terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
Advertisements generally refer to commercial advertisements that a commodity operator or service provider incurs, introducing, directly or indirectly, through a medium and form, the commodity or service offered by itself. The form of the advertisement may include text, pictures, audio, video, etc., and accordingly, the playing of the advertisement may also include presentation of the text, pictures, playing of the audio, video, etc.
The online advertisement engine system is an online system for advertising on the Internet, and is responsible for producing advertisements seen by users in real time, and provides efficient and intelligent advertisement delivery and management services for advertisers and advertisement publishers. For example, a user opens an application for shopping on a mobile phone, and various commodity picture links presented on a home page of the application are all advertisements put on through an online advertisement engine system. Through the online advertisement engine system, an advertiser can select proper advertisement forms, position target audiences, set delivery time, budget and other information according to own requirements and targets, and manage and monitor advertisement delivery effects. For advertisement publishers, the online advertisement engine system can be used to manage and optimize ad spots, improving the quality and revenue of advertisements. The online advertisement engine system can select the most suitable advertisement for delivery through real-time data analysis and prediction, and provide accurate advertisement content and positioning, thereby attracting more users to click and participate.
With the continuous development of advertisement showing service, the flow scene accessed by the online advertisement engine system is more and more, and the advertisement putting strategy is more and more complex, so that the machine resource amount used by the online advertisement engine system is continuously increased, and the demand on calculation power is also continuously increased.
The traffic scenario refers to a traffic scenario capable of accepting traffic in the user service system, that is, a traffic scenario capable of being accessed by the user. The user service system may be an on-line to off-line system that provides services to the user, such as an on-line to off-line system for shopping, an on-line to off-line system for ordering take-away. The user service system presents service content to the user, typically in the form of an application or web site, with at least one business scenario providing service information to the user. In the application, the business particularly refers to a display advertisement business, and a business scene is a scene used for providing advertisement space in an application program or a website of a user service system. For example, in an application for shopping, pages that you like are shown in FIG. 1, where there are provided many links to merchandise that the user may be interested in, the links to merchandise being presented in text, pictures, video, etc. In fact, these commodity links are advertisements, and guessing that you like pages is one business scenario in the application that is used to provide ad slots. In addition, in the application program for shopping, besides the favorite business scenes, business scenes such as post-purchase recommendation, home page recommendation, point availability and the like can be provided, and the business scenes can provide advertisement positions so as to display commodity advertisements put in the online advertisement engine system, and a user can browse the business scenes so as to view the commodity advertisements displayed in the business scenes.
When a user opens a page of a service scene to browse, a page display request is carried out on the service scene once, and a flow is generated. Traffic received by a service scenario, i.e. the number of requests made by a user for the service scenario, in other words the number of requests received by the service scenario.
The application program of the user service system or different business scenes in the website are accessed into the online advertisement engine system, so that the online advertisement engine system can provide proper advertisements for the business scenes browsed by the user according to the requirements and setting information of advertisers and/or advertisement publishers and the historical behavior data of the user of the application program or the website.
The online advertisement engine system comprises at least one service module, and the service module is used for executing a corresponding service strategy, namely an advertisement putting strategy, so that a browsing user aiming at a current service scene recommends proper advertisements from massive candidate advertisements and presents the advertisements in a certain form. For example, the online advertisement engine system comprises a recall module, a ranking module and the like, wherein the recall module is used for executing a recall strategy, and the ranking module is used for executing a ranking strategy. Recall here refers to screening out a recommendable advertisement from a vast number of candidate advertisements. The ranking refers to ranking the recommended advertisements according to the personalized requirements of the users. When a user opens an application program for shopping, the home page of the application program presents a plurality of commodity picture links which are possibly interesting to the user, namely, a business scene recommended by the home page provides a plurality of commodity advertisements, the commodity advertisements are selected from a large number of candidate advertisements by executing a recall strategy through a recall module in an online advertisement engine system, the recommended advertisements selected by the recall module are arranged by executing a sort strategy through a sort module in the online advertisement engine system, and finally the arranged recommended advertisements are put into the business scene recommended by the home page for display.
The advertisement publisher, namely an advertisement service party providing advertisement promotion service, puts advertisements in the service scene of the application program or the website through the online advertisement engine system for users to browse, thereby helping advertisers, namely merchants, promote goods. After the advertisement is put, the merchant pays corresponding popularization fees to the advertisement service party, and accordingly, the advertisement service party obtains the benefits of displaying the advertisement service.
Typically, a service module in an online advertising engine system uses machine resources (also referred to as computing resources) to generate a cost of computing power when executing a service policy. In the computing field, a machine resource refers to a resource of a physical device such as a computer or a server. In the present application, the number of machine resources is a measure of the number of machine cores, or the number of CPU cores. The number of cores of a CPU refers to the number of cores of a Central Processing Unit (CPU), and each core (also referred to as a core) of the CPU can independently run instructions to perform parallel computation and multithreading. The number of the CPU cores has an important influence on the performance and the processing capacity of the computer, and the overall performance of the CPU is stronger as the number of the CPU cores is larger under the condition that the kernel frequency, the cache size and the like are the same. The machine resources used by the service module when executing the service policy may also be referred to as the number of machine cores or the number of CPU cores used by the service module when executing the service policy. The computational effort cost refers to the cost of the machine resources used to complete the computing task, and is generally related to the number of machine resources used and the computing time, the greater the number of machine resources used, the longer the computing time, and the higher the computational effort cost. In the present application, the service module performs the service policy, which is essentially performing high-performance calculation, and this calculation task often requires expensive dedicated hardware and consumes a large amount of energy, so the calculation cost is relatively high. It can be seen that in the presentation of advertising services, efficient management and utilization of machine resources is critical to improving service efficiency and revenue.
In practical applications, since the online advertisement engine system includes a plurality of service modules, there are a plurality of corresponding service policies, and each service policy may be further divided into a plurality of levels (also referred to as gears), where the levels are used to represent the status levels of the number of machine resources used when the service policies are executed. Thus, the number and time of machine resources used by a service module to execute a service policy for different service modules, or by a service module to execute a service policy of different levels, are not necessarily the same. Generally, the more complex the service policy, the more machine resources are used and the more computing power is used. For example, the ranking strategies executed by the ranking module may be classified into 10 ranks, and the ranking strategies of different ranks have different complexity due to the different ranking models, so that the ranking strategies with higher ranks are more complex, and therefore, the ranking module executes the ranking strategies with higher ranks with more machine resources. However, the more complicated the service policy is, the more advertisements are recommended to the user's needs, the more advertisements the user may browse, or the more probability of clicking to view and purchase is, the better the advertising effect is, and the higher the advertising service benefit is obtained, in other words, the higher the benefit brought by the machine resources, that is, the higher the use efficiency of the machine resources is.
Advertisement effectiveness refers to the impact and effectiveness of an advertisement in a target audience, which can be measured by the following data metrics: exposure, i.e., the number of times an advertisement is displayed; click rate, i.e., the ratio of the number of times an advertisement is clicked to the amount of exposure; conversion rate, i.e., the rate at which the user performs a desired action, such as purchasing a good, etc., after the advertisement is clicked; etc.
In addition, revenues for displaying advertising traffic are also related to traffic of traffic scenarios. Since the traffic received by different service scenarios within a certain period of time may be different, for example, in an application program for shopping within the same period of time, the traffic received by the service scenario recommended by the home page is generally greater than the traffic received by the service scenario recommended after shopping, that is, the number of users browsing the recommended page of the home page is generally greater than the number of users browsing the recommended page after shopping. Alternatively, the same business scenario may have different traffic rates received at different times, e.g., in an application for shopping, the top page recommended business scenario will typically receive a greater traffic rate during a promotional event than during no promotional event. Aiming at business scenes adapting to different flows, the same service module is used for executing the service strategies with the same grade to put advertisements, the finally obtained advertisement effects are not necessarily the same, and the obtained advertising business revenues are not necessarily the same.
Summarizing, the more complex the service policy, the higher the revenue from the traffic scenario. Under the condition of sufficient machine resources, advertisement delivery can be carried out on all business scenes by adopting the most complex service strategy so as to ensure that the advertising effect and the income of each business scene are best. Under the condition of insufficient machine resources, the advertisement publisher is more prone to advertisement delivery on business scenes with large flow, and adopts the most complex service strategy, but the advertisement effect and the benefits of other business scenes and the global benefits of all business scenes are affected, so that the problem of balance of the benefits rate among the business scenes exists, and the benefits rate of the business scenes = business scene benefits/machine resource cost.
At present, since the supply of machine resources is greatly affected by factors such as productivity and cost control, the shortage of resources can be a long-term and main challenge for an advertisement service platform, and the efficiency of converting machine resources into benefits can be an important index for measuring whether an online advertisement engine system is efficient. In such a background, it is becoming more and more important to improve the efficiency of use of machine resources by fully utilizing the machine resources currently existing. In addition, for the e-commerce platform, the service system often faces a sales promotion, and at that time, the flow accepted by each service scene in the application program or the website of the e-commerce platform can be increased by several times or even tens times, so that in order to ensure that the use efficiency of machine resources is improved under the condition of limited machine resources, a great deal of manpower is generally required to be input to carry out multiple pressure measurement on the online advertisement engine system before the sales promotion, and the advertisement delivery strategy is continuously adjusted. However, the user behavior and advertisement volume change usually occur along with the approaching of the promotion, so that the deviation of the prior pressure measurement data occurs, and the self-adaption capability is insufficient. In view of the foregoing, it is desirable to provide a method for machine resource allocation for business scenarios and service modules, so that machine resource utilization efficiency is maximized and automatic validation can be achieved on line.
In view of the above problems, machine resources can be allocated from the global view of multiple service scenarios according to the current machine resource status of the online advertisement engine system to achieve the goal of benefit, that is, under the limitation of the current machine resources, the goal of global service benefit maximization is to be achieved, how to allocate the machine resources among the service modules of the online advertisement engine system, how much traffic is accepted by each service scenario, and what level (also referred to as gear) of service policy connection is used for each service scenario.
Therefore, the application provides a resource allocation method, which aims at maximizing the benefits obtained by putting advertisements into all service scenes in a user service system by an online advertisement engine system, takes the target flow accepted by all the service scenes in the preset time and the machine resource data which can be used by the online advertisement engine system as constraint conditions, and obtains the machine resource data which is required to be allocated for all the service modules in the online advertisement engine system, the flow accepted by all the service scenes in the preset time and the service policy grade which can be used by all the service modules for all the service scenes.
Specifically, the advertisement publisher allocates machine resources to m (m is an integer greater than or equal to 1) service modules included in an online advertisement engine system accessing n (n is an integer greater than or equal to 1) service scenes by using the above resource allocation method, please refer to a flowchart shown in fig. 2, and the resource allocation process includes the following steps:
s201, the advertisement publisher inputs the target flow received by each business scene in the preset time and the total number of machine cores available to the online advertisement engine system through the information setting interface of the resource distribution system.
The information setting interface is shown in fig. 3, and includes a setting item for the total number of machine cores, and a setting item for the target flow (i.e., the maximum number of requests) received in a preset time for each service scenario, where the preset time is typically one second. The advertisement publisher can input the total quantity C of machine cores available to the online advertisement engine system and the target flow Q corresponding to the service scene i through the information setting interface i 。
S202, the resource allocation system executes a resource allocation strategy, namely, a linear programming model is utilized to obtain a resource allocation result according to the target flow received by each service scene in the preset time and the total number of machine cores available to the online advertisement engine system.
The linear programming model is a mathematical optimization method for solving the optimization problem under the linear constraint condition. Its goal is to find the maximum or minimum of a linear objective function under a set of linear constraints. The linear programming model is widely applied to the fields of supply chain optimization, resource allocation, production planning, investment combination optimization and the like.
In the application, aiming at the display advertisement service, a constructed linear programming model is as follows:
q i ≤Q i
rt i，j ≤RT i，j
in the above formula, i represents an ith service scene in n service scenes;
j represents a j-th service module in m service modules;
q i the number of requests qps in every second of the ith service scenario, namely, the traffic received in every second of the ith service scenario;
Q i representing the maximum request number in every second of the ith service scene, namely the target flow accepted by the ith service scene every second;
l i representing a service policy combination gear used for an ith service scenario, which contains gears of service policies respectively executed by a plurality of service modules, e.g. l i ＝<50，30，40>The gear of the service strategy executed by the first service module is 50 gears, the gear of the service strategy executed by the second service module is 30 gears, and the gear of the service strategy executed by the third service module is a combined gear formed by 40 gears aiming at the ith service scene;
rpm i Representing the benefit obtained by advertising the ith service scene receiving the single traffic, namely the unit traffic benefit corresponding to the ith service scene, the benefit and the service scene and l j Correlation;
core j representing the number of machine cores used by the jth service module;
c represents the total number of machine cores available to the online advertising engine system;
rt i，j representing the actual use time of the jth service module for the ith service scene;
RT i，j representing the maximum usable time of the jth service module for the ith service scenario.
In the linear programming model, a calculation function of gains obtained by putting advertisements into n business scenes by the online advertisement engine system in each second is taken as an objective function, namely, the sum of products of flow received by each business scene and corresponding unit flow gains is taken as the objective function.
The solving variables of the objective function include: (1) Traffic q received per second for ith traffic scenario i The method comprises the steps of carrying out a first treatment on the surface of the (2) Service policy combination gear l used for ith service scenario i The method comprises the steps of carrying out a first treatment on the surface of the (3) Number of cores of machine used by jth service module j 。
The constraint conditions include: (1) Traffic q received per second for ith traffic scenario i Target flow Q accepted per second of no more than ith service scene i The method comprises the steps of carrying out a first treatment on the surface of the (2) The sum of the machine core numbers used by the m service modules is not more than the total machine core number C which can be used by the online advertisement engine system; (3) The jth service module aims at the actual use time rt of the ith service scene i，j No more than the maximum usable time RT of the jth service module for the ith service scene i，j 。
Wherein, the target flow Q accepted by the ith service scene per second i And the total quantity C of machine cores which can be used by the online advertisement engine system is obtained by the advertisement publisher according to the actual situation setting, and the jth service module aims at the maximum usable time RT of the ith service scene i，j Is a conventional technologyParameters.
Since the unit flow benefit corresponding to a certain service scenario is related to the service policy combination gear used for the service scenario, that is, different service policy combination gears are used for a certain service scenario receiving the unit flow, the benefit corresponding to the service scenario finally obtained is not necessarily the same. Therefore, the unit flow gain corresponding to the service scene is determined after data measurement, namely, the relation among the gear of the service strategy, the number of used machine cores and the gain is measured in the drilling environment, so that the combined gear with higher cost performance is screened out, and then the unit flow gain corresponding to the service scene is determined according to the screened combined gear. The cost performance is defined as the ratio of the profit of the service scene in the combined gear to the number of machine cores used in the combined gear, and the higher cost performance indicates that the ratio of the profit of the service scene in the combined gear to the number of machine cores used in the combined gear is greater than a preset threshold.
Specifically, a complete online advertisement engine system is deployed on a server cluster to serve as an exercise environment, and under the condition that the same benefits are found through a heuristic iteration method in the exercise environment, a combined gear with the minimum number of machine cores is used, and then the combined gear with higher cost performance is screened out. Usually, the accuracy of the directly measured data is verified and adjusted, for example, the measured data in the drilling environment has a stable gap from the actual data in the on-line environment, and then the measured data is corrected to find the preferred combination gear. And determining the benefit corresponding to the service scene adapting to the single flow when the preferred combined gear is used as the unit flow benefit corresponding to the service scene.
Substituting unit flow gains corresponding to service scenes into an objective function, solving variables under constraint conditions, and obtaining the number of machine cores used by each service module, the flow received by each service scene in preset time and service strategy gears used by each service module for each service scene respectively when the gains obtained by putting advertisements to each service scene in a preset time in-line advertisement engine system are maximum.
S203, the resource distribution system outputs a resource distribution result, namely, the number of machine cores required to be distributed for each service module, the flow which can be accepted by each service scene in the preset time and the service policy gear which can be used by each service module for each service scene respectively when the profit obtained by advertising the advertisement to each service scene by the online advertisement engine system is maximum in the preset time.
And outputting the number of machine cores used by each service module, the flow respectively accepted by each service scene in preset time and the service policy gear respectively used by each service module for each service scene as a resource allocation result when the obtained benefit of advertising to each service scene by the online advertising engine system in preset time is maximum by solving the linear programming model, wherein the number of machine cores required to be allocated for each service module, the flow respectively accepted by each service scene in preset time and the service policy gear respectively usable by each service module for each service scene are respectively corresponding to each service module.
The resource allocation result can be displayed in the form of a resource allocation report, as shown in fig. 4, and the resource allocation report displayed by the resource allocation result display interface includes the number of machine cores corresponding to each service module, the receiving flow corresponding to each service scene and the service policy gear. The advertisement publisher can learn the results calculated by the resource allocation system through the resource allocation report.
In addition, the resource allocation results may be directly output to a corresponding interface of the online advertising engine system, thereby being effective online. The online advertisement engine system can schedule machine resources according to the number of machine cores required to be distributed for each service module so as to adjust the number of machine cores which can be used by each service module; performing flow control according to the flow which can be respectively accepted by each service scene in the preset time so as to adjust the maximum flow which can be respectively accepted by each service scene in the preset time; and setting gears according to service policy gears which can be used by each service module for each service scene respectively so as to adjust the service policy gears which are used by each service module for each service scene respectively. So far, the online advertisement engine system can deliver advertisements to each business scene according to the adjusted parameters.
The resource allocation method performs resource allocation with the aim of maximizing service benefit, can effectively improve the utilization efficiency of machine resources, ensures that the machine resources are allocated more finely and accurately, and realizes the optimization of computing power configuration between service scenes and service modules. The resource allocation result can be directly effective on line, and the effectiveness and the self-adaption capability are stronger.
The above is an introduction of the application scenario embodiment of the resource allocation method provided by the present application, and the resource allocation method, the resource allocation device, the electronic device and the computer readable storage medium related to the present application are described in detail below.
First embodiment
A first embodiment of the present application provides a resource allocation method, as shown in fig. 5, including the following steps:
s501, obtaining target flow accepted by a service scene in a preset time, wherein the service scene is a service scene for providing service information for a user in a user service system;
s502, computing resource data usable by a service information processing system is obtained, and the service information processing system is used for throwing service information which can be processed by a user into a service scene in the user service system;
s503, obtaining calculation resource data required to be distributed for a service module in the service information processing system when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset profit condition according to the target flow accepted by the service scene in the preset time and the calculation resource data usable by the service information processing system, wherein the service module is used for determining the service information to be put into for the user accessing the service scene.
A detailed description of the steps in the flowchart of fig. 5 is provided below.
S501, obtaining target flow accepted by a service scene in a preset time, wherein the service scene is a service scene for providing service information for a user in a user service system.
The user service system refers to an online-to-offline system for providing services for users, and service contents are generally presented to the users in the form of application programs or websites, and at least one service scene for providing service information for the users is provided in the user service system. The service scenario herein, that is, the service scenario described above, is, for the user, a service scenario that can be accessed by the user and provides service information for the user; the service scenario for providing information exhibition is for the information publisher, which can accept traffic. The service information may be advertisements, the service information in the service scene is typically presented in text, pictures, audio, video, etc., which can be clicked and jumped by the user to link to the item detail page. Accordingly, the information publisher, i.e., the advertisement publisher, and the information display, i.e., the advertisement space.
The service information in the service scenario is typically produced and delivered by a service information handling system, which may be an online advertisement engine system. The advertisement publisher, namely the advertisement service party providing the advertisement promotion service, puts advertisements into service scenes in the user service system through the online advertisement engine system, thereby helping advertisers, namely merchants, promote goods. After the advertisement is put, the merchant pays corresponding popularization fees to the advertisement service party. That is, the advertisement service party obtains the benefit of exhibiting the advertisement service by putting the advertisement to the service scene in the user service system through the online advertisement engine system.
When a user opens a page of a service scene to browse, a page display request is performed on the service scene, and a flow is generated. Traffic carried by a service scenario, i.e. the number of requests made by a user to the service scenario, in other words the number of requests carried by the service scenario.
And obtaining the target flow accepted by the service scene in the preset time, namely obtaining the maximum number of requests accepted by the service scene in each second. When the user service system is provided with a plurality of service scenes, the target flow respectively accepted by the service scenes in the preset time is required to be obtained. The target flow accepted by each service scene in the preset time can be set by the information publisher according to the actual demand.
S502, obtaining computing resource data usable by a service information processing system, wherein the service information processing system is used for throwing service information which can be processed by a user into a service scene in the user service system.
The service information processing system comprises at least one service module, and the service module can execute corresponding service strategies to determine service information to be put in for users accessing the service scene. The service policies are divided into a plurality of levels representing the status levels of the computing resource data used when the service policies are executed. That is, the service module executes the corresponding service policy to determine the service information to be delivered for the user accessing the service scenario, and the computing resource, that is, the machine resource is used when executing the service policy. The different levels of service policies are performed using different amounts of computing resource data, and in general, the more complex service policies are performed using more computing resource data, but the higher the revenue that the more complex service policies result. In the advertisement showing service, the more complicated the advertisement recommended by the service strategy is, the more advertisements the user can browse, or the more probability that the user clicks to view and purchase is, the better the advertisement effect is, and the higher the obtained advertising showing service benefit is.
The computing resource data available to the service information handling system, i.e., the total computing resource data available to all of the service modules in the service information handling system, is obtained. The computing resource data usable by the service information handling system may be set by the information publisher based on actual available data of the computing resource.
S503, obtaining calculation resource data required to be distributed for a service module in the service information processing system when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset profit condition according to the target flow accepted by the service scene in the preset time and the calculation resource data usable by the service information processing system, wherein the service module is used for determining the service information to be put into for the user accessing the service scene.
The step is to obtain machine resource data required to be allocated for a service module in a service information processing system by taking a target flow accepted by the service scene in a preset time and computing resource data usable by the service information processing system as constraint conditions with the aim of maximizing the benefit obtained by putting the service information into the service scene in the user service system by the service information processing system.
When the user service system is provided with a plurality of service scenes, and the service information processing system comprises a plurality of service modules, according to target flow respectively accepted by the service scenes in the preset time and calculation resource data which can be used by the service information processing system, obtaining calculation resource data which are required to be allocated for the service modules in the service information processing system when the benefit data obtained by putting service information into the service scenes in the user service system by the service information processing system in the preset time meet preset benefit conditions.
Before that, unit flow gain data corresponding to the service scene is required to be obtained, wherein the unit flow gain data is gain data obtained by putting service information into the service scene receiving unit flow in the preset time, and the unit flow is a measurement standard of flow and is generally single flow. When a plurality of service scenes are arranged in the user service system, obtaining unit flow income data corresponding to the service scenes respectively.
Because the grade of the service policy can affect the income of the service scene, namely, the grade of the service policy executed by the service module is different for the same service scene receiving unit flow, and finally the obtained income data is also different. Accordingly, the unit flow rate profit data corresponding to the service scenario is related to not only the service scenario but also the level of the service policy used.
When the service information processing system includes a plurality of service modules, for the same service scenario for receiving unit traffic, the service policies executed by the plurality of service modules have different combination levels, and the finally obtained revenue data may be the same, but the total amount of used computing resources is not necessarily the same. The combination level is a combination level formed by the levels of the service policies respectively executed by the service modules aiming at the same service scene. For example, a combination level l i ＝<50，30，40>The level of the service policy executed by the first service module is 50, the level of the service policy executed by the second service module is 30, and the level of the service policy executed by the third service module is 40 for the ith service scenario. The level here is the gear described above, and the combined level is the combined gear described above.
In consideration of the above, the unit flow revenue data corresponding to the service scenario is obtained according to the grade of the service policy executed when the service module determines the service information to be put for the user accessing the service scenario. When a plurality of service scenes are arranged in a user service system, and a service information processing system comprises a plurality of service modules, according to the service modules, the combination level of service strategies executed when service information to be put in is determined for users accessing the service scenes, and unit flow gain data corresponding to the service scenes respectively is obtained.
Specifically, the relation among the grade of the service strategy, the used machine resource data and the obtained benefits is measured in the drilling environment, so that the combination grade with higher cost performance is screened out, and then the unit flow benefits corresponding to the service scene are determined according to the screened combination grade. The cost performance is defined as the ratio of the profit of the service scene under the combination level to the calculation resource data used by the combination level, and the higher cost performance indicates that the ratio of the profit of the service scene under the combination level to the number of machine cores used by the combination level is greater than a preset threshold.
If the service modules are more and the service strategies are more in grades, the number of the combined grades is huge, the relation among the grades of the service strategies, the used machine resource data and the obtained benefits is measured one by one in the drilling environment, and the workload is huge. Therefore, through a heuristic iteration method, when the computing resource data used by the service modules for executing the service policies of the reference combination level meet the preset computing resource conditions, the service information to be put determined by the service modules is put into the reference revenue data obtained by the service scene for accepting the unit flow. That is, a heuristic iteration method is adopted, and under the condition that the same benefits are found in a huge amount of combination grades, the combination grade with the least calculation resource data is used, then the combination grade with higher cost performance is screened out, and meanwhile, the benefit data corresponding to the combination grade with higher cost performance is obtained. And then, determining the reference revenue data corresponding to the service scenes as unit flow revenue data corresponding to the service scenes.
Obtaining benefit data obtained by putting service information into a service scene in the user service system by the service information processing system in the preset time according to the flow accepted by the service scene in the preset time and the unit flow benefit data corresponding to the service scene. The product of the flow accepted by the service scene in the preset time and the unit flow gain data corresponding to the service scene is calculated, and the calculation result is gain data obtained by putting service information into the service scene in the user service system by the service information processing system in the preset time. When a plurality of service scenes are arranged in the user service system, obtaining the obtained income data of putting service information into the plurality of service scenes in the user service system by the service information processing system in the preset time according to the flow respectively accepted by the plurality of service scenes in the preset time and the unit flow income data respectively corresponding to the plurality of service scenes. And calculating the sum of products of the flow received by each business scene and the corresponding unit flow gain.
Specifically, a linear programming model is firstly constructed, the linear programming model comprises an objective function and constraint conditions, wherein the objective function is a calculation function of benefit data obtained by putting service information into a service scene in a user service system by the service information processing system within the preset time, solution variables of the objective function comprise flow accepted by the service scene within the preset time, a grade of a service strategy executed when the service module determines service information to be put for a user accessing the service scene and calculation resource data required to be allocated for the service module in the service information processing system, and the constraint conditions comprise the objective flow accepted by the service scene within the preset time, calculation resource data available to the service information processing system and time available to the service module for determining the service information to be put for the user accessing the service scene.
The linear programming model was constructed as follows:
q i ≤Q i
rt i，j ≤RT i，j
wherein i represents an ith service scene in n service scenes, and n is an integer greater than or equal to 1;
j represents a j-th service module in m service modules, and m is an integer greater than or equal to 1;
q i Representing the number of requests qps in the ith service scenario per second, i.e., the traffic accepted by the ith service scenario per second;
Q i representing the maximum request number in every second of the ith service scene, namely the target flow accepted by the ith service scene every second;
l i representing a service policy combination level used for an ith service scenarioA hierarchy including service policies respectively executed by the plurality of service modules;
rpm i representing the benefit obtained by delivering service information to the ith service scenario receiving unit flow, namely the unit flow benefit corresponding to the ith service scenario, wherein the benefit is equal to the service scenario and l i Correlation;
core j computing resource data representing use of the jth service module;
c represents computing resource data usable by the service information processing system;
rt i，j representing the actual use time of the jth service module for the ith service scene;
RT i，j representing the maximum usable time of the jth service module for the ith service scenario.
And obtaining calculation resource data required to be distributed for a service module in the service information processing system, reference flow accepted by the service scene in the preset time and reference grade of a service strategy executed when the service module determines service information to be put for a user accessing the service scene when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets a preset profit condition by utilizing a linear programming model.
And adjusting computing resource data usable by the service module, flow acceptable by the service scene in the preset time and grade of executable service strategy when the service module determines service information to be put for a user accessing the service scene according to computing resource data required to be allocated for the service module in the service information processing system, reference flow accepted by the service scene in the preset time and reference grade of the service strategy executed when the service module determines the service information to be put for the user accessing the service scene when the service information processing system obtains the obtained benefit data of putting the service information to the service scene in the preset time by solving the linear programming model and the obtained benefit data of the service information processing system in the user service system meets preset benefit conditions.
When a plurality of service scenes are arranged in a user service system, and a service information processing system comprises a plurality of service modules, obtaining that when revenue data obtained by the service information processing system for delivering service information to the plurality of service scenes in the user service system within the preset time meets preset revenue conditions, computing resource data respectively required to be distributed for the plurality of service modules in the service information processing system, reference flow respectively accepted by the plurality of service scenes within the preset time, and reference combination grades of service strategies executed when the plurality of service modules determine service information to be delivered for users accessing the service scenes.
According to the calculation resource data required to be distributed for a plurality of service modules in the service information processing system respectively when the profit data obtained by putting service information into a plurality of service scenes in the user service system within the preset time meets the preset profit condition, the reference flow respectively accepted by the plurality of service scenes within the preset time and the reference combination level of the service strategy executed by the plurality of service modules when determining the service information to be put for a user accessing the service scenes, the calculation resource data respectively usable by the plurality of service modules, the flow respectively accepted by the plurality of service scenes within the preset time and the combination level of the service strategy executable by the plurality of service modules when determining the service information to be put for the user accessing the service scenes are adjusted.
The specific adjustment method may be that data obtained by solving the linear programming model is directly sent to the service information processing system, and the service information processing system processes service information according to the received data and puts the service information into a service scene of the user service system as an operation parameter of the service information processing system.
According to the resource allocation method, the resource allocation is carried out with the aim of maximization of the benefit, the utilization efficiency of computing resources can be effectively improved, and the resource allocation is finer and more accurate. The service information processing system can adjust the operation parameters in real time according to the resource allocation result, and has stronger timeliness and self-adaption capability.
Second embodiment
A second embodiment of the present application provides a resource allocation apparatus, as shown in fig. 6. The apparatus corresponds to the resource allocation method provided in the first embodiment, and since the apparatus embodiment is similar to the method embodiment, the description is relatively simple, and please refer to the content of the first embodiment for relevant points.
The resource allocation apparatus 600 provided in this embodiment includes:
the first target flow obtaining unit 601 is configured to obtain a target flow received by a service scenario in a preset time, where the service scenario is a service scenario in a user service system for providing service information for a user;
a first computing resource data obtaining unit 602, configured to obtain computing resource data usable by a service information processing system, where the service information processing system is configured to put service information that is available for processing by a user into a service scenario in the user service system;
The first computing resource allocation unit 603 obtains computing resource data required to be allocated to a service module in the service information processing system when revenue data obtained by putting service information into the service scenario in the user service system in the preset time meets a preset revenue condition according to target traffic carried by the service scenario in the preset time and computing resource data usable by the service information processing system, where the service module is used for determining service information to be put for a user accessing the service scenario.
Optionally, the apparatus further comprises:
the first unit flow rate profit data obtaining unit is used for obtaining unit flow rate profit data corresponding to the service scene, the unit flow rate profit data is obtained by putting service information into the service scene receiving unit flow rate in the preset time by the service information processing system, and the unit flow rate is a measurement standard of the flow rate;
the first profit data obtaining unit is used for obtaining profit data obtained by putting service information into a service scene in the user service system by the service information processing system in the preset time according to the flow accepted by the service scene in the preset time and the unit flow profit data corresponding to the service scene.
Optionally, the obtaining unit flow benefit data corresponding to the service scenario includes: and obtaining unit flow gain data corresponding to the service scene according to the grade of the service strategy executed when the service module determines the service information to be put for the user accessing the service scene, wherein the grade is used for representing the condition grade of the computing resource data used when the service strategy is executed.
Optionally, the obtaining, according to the target flow received by the service scenario in the preset time and the computing resource data usable by the service information processing system, computing resource data to be allocated to a service module in the service information processing system when revenue data obtained by the service information processing system in the preset time for delivering service information to the service scenario in the user service system meets a preset revenue condition includes:
obtaining calculation resource data required to be distributed for a service module in the service information processing system when the benefit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset benefit condition by utilizing a linear programming model;
The linear programming model comprises an objective function and constraint conditions, wherein the objective function is a calculation function of revenue data obtained by putting service information into a service scene in the user service system by the service information processing system within the preset time, solution variables of the objective function comprise flow accepted by the service scene within the preset time, a grade of a service strategy executed by the service module when determining service information to be put for a user accessing the service scene, and calculation resource data required to be allocated for the service module in the service information processing system, and the constraint conditions comprise the objective flow accepted by the service scene within the preset time, calculation resource data available to the service information processing system and time available to the service module when determining the service information to be put for the user accessing the service scene.
Optionally, the apparatus further comprises:
the first reference information obtaining unit is used for obtaining the reference grade of the service policy executed when the service module determines the service information to be put for the user accessing the service scene according to the target flow received by the service scene in the preset time and the calculation resource data usable by the service information processing system when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset profit condition.
Optionally, the apparatus further comprises:
and the first parameter adjustment unit is used for adjusting the calculation resource data usable by the service module, the flow acceptable by the service scene in the preset time and the grade of the executable service strategy when the service module determines the service information to be put for the user accessing the service scene according to the calculation resource data required to be allocated for the service module in the service information processing system, the reference flow accepted by the service scene in the preset time and the reference grade of the service strategy executed when the service module determines the service information to be put for the user accessing the service scene when the service information processing system puts the service information to the service scene in the preset time and the obtained profit data of the service scene in the service information processing system meet the preset profit condition.
Optionally, the user service system has a plurality of service scenes, and the service information processing system comprises a plurality of service modules;
the obtaining the target flow accepted by the service scene in the preset time comprises the following steps: obtaining target flow respectively accepted by the plurality of service scenes in the preset time;
The obtaining, according to the target flow received by the service scenario in the preset time and the computing resource data usable by the service information processing system, computing resource data to be allocated to a service module in the service information processing system when revenue data obtained by the service information processing system delivering service information to the service scenario in the user service system in the preset time meets a preset revenue condition includes:
according to the target flow respectively accepted by the service scenes and the computing resource data usable by the service information processing system in the preset time, obtaining computing resource data respectively required to be distributed for a plurality of service modules in the service information processing system when the profit data obtained by putting service information into the service scenes in the user service system in the preset time meets the preset profit condition.
Optionally, the first unit flow rate gain data obtaining unit is further configured to obtain unit flow rate gain data corresponding to each of the plurality of service scenarios, where the unit flow rate gain data is gain data obtained by the service information processing system throwing service information into the service scenario receiving the unit flow rate in the preset time, and the unit flow rate is a measurement standard of the flow rate;
The first revenue data obtaining unit is further configured to obtain revenue data obtained by the service information processing system throwing service information into a plurality of service scenes in the user service system in the preset time according to the traffic respectively accepted by the plurality of service scenes in the preset time and unit traffic revenue data respectively corresponding to the plurality of service scenes.
Optionally, the obtaining unit flow revenue data corresponding to each of the plurality of service scenarios includes:
and obtaining unit flow gain data corresponding to the service scenes respectively according to the combination grades of the service policies executed by the service modules when the service information to be put is determined for the users accessing the service scenes, wherein the combination grades are the combination grades determined according to the grades of the service policies executed by the service modules respectively, and the grades are used for representing the condition grades of the computing resource data used when the service policies are executed.
Optionally, the obtaining the unit flow revenue data corresponding to the multiple service scenes respectively according to the combination level of the service policies executed when the multiple service modules determine the service information to be put for the user accessing the service scenes includes:
Obtaining reference revenue data obtained by putting the service information to be put determined by the plurality of service modules into the service scene receiving unit flow when the computing resource data used by the plurality of service modules for executing the service policies of the reference combination level meets the preset computing resource conditions through a heuristic iteration method;
and determining the reference revenue data corresponding to the service scenes as unit flow revenue data corresponding to the service scenes.
Optionally, the first reference information obtaining unit is further configured to obtain, according to the target flows respectively accepted by the multiple service scenes in the preset time and the computing resource data usable by the service information processing system, a reference combination level of a service policy executed when the service information processing system puts service information into the multiple service scenes in the user service system in the preset time and the obtained revenue data meets a preset revenue condition, where the reference flows respectively accepted by the multiple service scenes in the preset time and/or the multiple service modules determine service information to be put for a user accessing the service scenes.
Optionally, the first parameter adjusting unit is further configured to adjust, according to the revenue data obtained by the service information processing system when delivering service information to a plurality of service scenes in the user service system in the preset time, computing resource data to be respectively allocated to a plurality of service modules in the service information processing system, reference traffic respectively accepted by the plurality of service scenes in the preset time, and a reference combination level of a service policy executed when the plurality of service modules determine service information to be delivered for a user accessing the service scenes, computing resource data respectively usable by the plurality of service modules, traffic respectively accepted by the plurality of service scenes in the preset time, and a combination level of a service policy executable when the plurality of service modules determine service information to be delivered for the user accessing the service scenes.
Third embodiment
A third embodiment of the present application provides a resource allocation method, which is applied to a server and corresponds to the method of the first embodiment, so that the description is relatively simple, and please refer to the content of the first embodiment specifically.
The resource allocation method provided in this embodiment, as shown in fig. 7, includes the following steps:
s701, obtaining scene flow information which is sent by a client and used for representing target flow accepted by a service scene in preset time and system resource information which is used for representing computing resource data which can be used by a service information processing system, wherein the service scene is a service scene which provides service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system;
s702, obtaining calculation resource data required to be distributed for a service module in the service information processing system when the benefit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset benefit condition according to the scene flow information and the system resource information, wherein the service module is used for determining service information to be put into for a user accessing the service scene;
s703, returning resource allocation information for representing computing resource data required to be allocated for a service module in the service information processing system when the benefit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset benefit condition to the client.
In step S702, for details of computing resource data to be allocated to a service module in the service information processing system when the obtained revenue data of the service information processing system delivering service information to a service scenario in the user service system within the preset time satisfies a preset revenue condition, please refer to the first embodiment.
Fourth embodiment
A fourth embodiment of the present application provides a resource allocation apparatus, as shown in fig. 8. The apparatus corresponds to the resource allocation method provided in the third embodiment, and since the apparatus embodiment is similar to the method embodiment, the description is relatively simple, and please refer to the content of the third embodiment for relevant points.
The resource allocation apparatus 800 provided in this embodiment includes:
a first information obtaining unit 801, configured to obtain scenario flow information sent by a client and used to represent a target flow carried by a service scenario within a preset time, and system resource information used to represent computing resource data usable by a service information processing system, where the service scenario is a service scenario in a user service system that provides service information for a user, and the service information processing system is configured to put service information that is available for processing by the user into the service scenario in the user service system;
A second computing resource allocation unit 802, configured to obtain, according to the scenario traffic information and the system resource information, computing resource data that needs to be allocated for a service module in the service information processing system when revenue data obtained by the service information processing system in delivering service information to a service scenario in the user service system in the preset time meets a preset revenue condition, where the service module is configured to determine service information to be delivered for a user accessing the service scenario;
a first information sending unit 803, configured to send, to the client, resource allocation information for indicating computing resource data to be allocated to a service module in the service information processing system when revenue data obtained by the service information processing system delivering service information to a service scenario in the user service system within the preset time satisfies a preset revenue condition.
Fifth embodiment
A fifth embodiment of the present application provides a resource allocation method, which is applied to a client and corresponds to the method of the first embodiment, so that the description is relatively simple, and please refer to the content of the first embodiment specifically.
The resource allocation method provided in this embodiment, as shown in fig. 9, includes the following steps:
S901, displaying an information setting interface for setting target flow accepted by a service scene in preset time and computing resource data usable by a service information processing system, wherein the service scene is a service scene for providing service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system;
s902, in response to detection of target traffic received in a preset time by a service scene set through the information setting interface and computing resource data usable by a service information processing system, obtaining scene traffic information for representing the target traffic received in the preset time by the service scene and system resource information for representing the computing resource data usable by the service information processing system;
s903, the scene flow information and the system resource information are sent to a server;
s904, obtaining resource allocation information which is returned by the service end and used for representing calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition;
And S905, displaying calculation resource data required to be distributed for a service module in the service information processing system when the benefit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset benefit condition according to the resource distribution information.
And the server returns resource allocation information for indicating the calculation resource data required to be allocated for the service module in the service information processing system when the benefit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset benefit condition, and the generation process of the server is referred to in the third embodiment.
Sixth embodiment
A sixth embodiment of the present application provides a resource allocation apparatus, as shown in fig. 10. The apparatus corresponds to the resource allocation method provided in the fifth embodiment, and since the apparatus embodiment is similar to the method embodiment, the description is relatively simple, and for the relevant points, please refer to the content of the fifth embodiment.
The resource allocation apparatus 1000 provided in this embodiment includes:
a first display unit 1001, configured to display an information setting interface for setting a target flow rate carried by a service scenario in a preset time and computing resource data usable by a service information processing system, where the service scenario is a service scenario in a user service system that provides service information for a user, and the service information processing system is configured to put service information that is available for processing by the user into the service scenario in the user service system;
A second information obtaining unit 1002, configured to obtain, in response to detection of a target flow rate received by a service scenario set through the information setting interface within a preset time and computing resource data usable by a service information processing system, scenario flow rate information for representing the target flow rate received by the service scenario within the preset time, and system resource information for representing computing resource data usable by the service information processing system;
a second information sending unit 1003, configured to send the scene traffic information and the system resource information to a server;
a third information obtaining unit 1004, configured to obtain resource allocation information returned by the server and used to indicate that, when revenue data obtained by the service information processing system in the preset time by delivering service information to a service scenario in the user service system meets a preset revenue condition, the resource allocation information is calculated resource data that needs to be allocated for a service module in the service information processing system;
and a second display unit 1005, configured to display, according to the resource allocation information, computing resource data that is required to be allocated for a service module in the service information processing system when revenue data obtained by the service information processing system delivering service information to a service scenario in the user service system in the preset time satisfies a preset revenue condition.
Seventh embodiment
A seventh embodiment of the present application provides a resource allocation method, which corresponds to the method of the first embodiment, so that the description is relatively simple, and please refer to the content of the first embodiment specifically.
The resource allocation method provided in this embodiment, as shown in fig. 11, includes the following steps:
s1101, obtaining a target flow received by a target scene in a preset time;
s1102, obtaining computing resource data usable by an information processing system, wherein the information processing system is used for throwing information into the target scene;
s1103, obtaining calculation resource data required to be distributed for a processing module in the information processing system when the profit data obtained by the information processing system throwing information into the target scene in the preset time meets the preset profit condition according to the target flow received by the target scene in the preset time and the calculation resource data usable by the information processing system, wherein the processing module is used for determining target information for a user accessing the target scene.
The present embodiment aims to provide a resource allocation method, which is not limited to application in displaying advertisement services, but is substantially the same as the method of the first embodiment, so details are not repeated here.
Eighth embodiment
An eighth embodiment of the present application provides a resource allocation apparatus, as shown in fig. 12. The apparatus corresponds to the resource allocation method provided in the seventh embodiment, and since the apparatus embodiment is similar to the method embodiment, the description is relatively simple, and for the relevant points, please refer to the content of the seventh embodiment.
The resource allocation apparatus 1200 provided in this embodiment includes:
a second target flow obtaining unit 1201, configured to obtain a target flow received by a target scene in a preset time;
a second computing resource data obtaining unit 1202, configured to obtain computing resource data usable by an information processing system, where the information processing system is configured to put information into the target scene;
the third computing resource allocation unit 1203 is configured to obtain, according to a target flow received by the target scene in a preset time and computing resource data usable by the information processing system, computing resource data required to be allocated to a processing module in the information processing system when revenue data obtained by the information processing system delivering information to the target scene in the preset time meets a preset revenue condition, where the processing module is configured to determine target information for a user accessing the target scene.
Ninth embodiment
A ninth embodiment of the present application provides an electronic apparatus as shown in fig. 13. The electronic device includes: at least one processor 1301, at least one memory 1302, at least one communication interface 1303, and at least one communication bus 1304. In the alternative, processor 1301 may be a processor CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present application. The memory 1302 may comprise high-speed RAM memory or may further comprise non-volatile memory (non-volatile memory), such as at least one disk memory. The communication interface 1303 may be an interface of a communication module, such as an interface of a GSM module. The memory 1302 stores a program and data, and the processor 1301 calls the program stored in the memory 1302 to implement the above-described resource allocation method.
Tenth embodiment
A tenth embodiment of the present application provides a computer-readable storage medium having stored thereon one or more computer instructions that are executed by a processor to implement the above-described resource allocation method.
It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region, and provide corresponding operation entries for the user to select authorization or rejection.
Although in the above detailed description several modules or units for action execution are mentioned, such partitioning is not mandatory. Indeed, the features and functions of two or more modules or units described above may be embodied in one module or unit, in accordance with embodiments of the present application. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.
Furthermore, although the steps of the methods of the present application are depicted in the accompanying drawings in a particular order, this is not required to either imply that the steps must be performed in that particular order, or that all of the illustrated steps be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform, etc.
It should be noted that the embodiments of the present application can be realized in hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those of ordinary skill in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The device of the present application and its modules may be implemented by hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., as well as software executed by various types of processors, or by a combination of the above hardware circuitry and software, such as firmware.
The foregoing is merely illustrative of specific embodiments of the present application, and the scope of the application is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present application will be apparent to those skilled in the art within the scope of the present application.
Claims (21)
1. A method for resource allocation, comprising:
obtaining target flow accepted by a service scene in a preset time, wherein the service scene is a service scene for providing service information for a user in a user service system;
acquiring computing resource data usable by a service information processing system, wherein the service information processing system is used for throwing service information which can be processed by a user into a service scene in the user service system;
according to the target flow accepted by the service scene in the preset time and the calculation resource data usable by the service information processing system, obtaining calculation resource data required to be distributed for a service module in the service information processing system when the income data obtained by the service information processing system for throwing service information into the service scene in the user service system in the preset time meets the preset income condition, wherein the service module is used for determining service information to be thrown for a user accessing the service scene.
2. The resource allocation method according to claim 1, further comprising:
obtaining unit flow rate gain data corresponding to the service scene, wherein the unit flow rate gain data is gain data obtained by putting service information into the service scene receiving unit flow rate in the preset time by the service information processing system, and the unit flow rate is a flow rate measurement standard;
obtaining benefit data obtained by putting service information into a service scene in the user service system by the service information processing system in the preset time according to the flow accepted by the service scene in the preset time and the unit flow benefit data corresponding to the service scene.
3. The method for allocating resources according to claim 2, wherein the obtaining unit flow rate profit data corresponding to the service scenario includes:
and obtaining unit flow gain data corresponding to the service scene according to the grade of the service strategy executed when the service module determines the service information to be put for the user accessing the service scene, wherein the grade is used for representing the condition grade of the computing resource data used when the service strategy is executed.
4. The resource allocation method according to claim 1, wherein the obtaining, according to the target traffic received by the service scenario in the preset time and the computing resource data usable by the service information processing system, the computing resource data to be allocated to the service module in the service information processing system when the revenue data obtained by the service information processing system delivering service information to the service scenario in the user service system in the preset time satisfies a preset revenue condition includes:
obtaining calculation resource data required to be distributed for a service module in the service information processing system when the benefit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset benefit condition by utilizing a linear programming model;
the linear programming model comprises an objective function and constraint conditions, wherein the objective function is a calculation function of revenue data obtained by putting service information into a service scene in the user service system by the service information processing system within the preset time, solution variables of the objective function comprise flow accepted by the service scene within the preset time, a grade of a service strategy executed by the service module when determining service information to be put for a user accessing the service scene, and calculation resource data required to be allocated for the service module in the service information processing system, and the constraint conditions comprise the objective flow accepted by the service scene within the preset time, calculation resource data available to the service information processing system and time available to the service module when determining the service information to be put for the user accessing the service scene.
5. The resource allocation method according to claim 1, further comprising:
obtaining, according to a target flow received by the service scene in a preset time and calculation resource data usable by the service information processing system, a reference level of a service policy executed when the service scene receives a reference flow and/or the service module determines service information to be released for a user accessing the service scene when revenue data obtained by the service information processing system releasing service information to the service scene in the user service system in the preset time meets a preset revenue condition.
6. The resource allocation method according to claim 5, further comprising:
according to the calculation resource data required to be distributed for a service module in the service information processing system, the reference flow accepted by the service scene in the preset time and the reference grade of a service strategy executed when the service module determines the service information to be put for a user accessing the service scene when the service information processing system puts the service information into the service scene in the user service system in the preset time and the obtained profit data meet the preset profit condition, the calculation resource data usable by the service module, the flow accepted by the service scene in the preset time and the grade of the executable service strategy when the service module determines the service information to be put for the user accessing the service scene are adjusted.
7. The resource allocation method according to claim 1, wherein the user service system has a plurality of service scenarios therein, and the service information processing system includes a plurality of service modules;
the obtaining the target flow accepted by the service scene in the preset time comprises the following steps: obtaining target flow respectively accepted by the plurality of service scenes in the preset time;
the obtaining, according to the target flow received by the service scenario in the preset time and the computing resource data usable by the service information processing system, computing resource data to be allocated to a service module in the service information processing system when revenue data obtained by the service information processing system delivering service information to the service scenario in the user service system in the preset time meets a preset revenue condition includes:
according to the target flow respectively accepted by the service scenes and the computing resource data usable by the service information processing system in the preset time, obtaining computing resource data respectively required to be distributed for a plurality of service modules in the service information processing system when the profit data obtained by putting service information into the service scenes in the user service system in the preset time meets the preset profit condition.
8. The resource allocation method according to claim 7, further comprising:
obtaining unit flow gain data corresponding to the service scenes respectively, wherein the unit flow gain data is gain data obtained by putting service information into the service scene receiving the unit flow in the preset time by the service information processing system, and the unit flow is a flow measurement standard;
obtaining the income data obtained by putting service information into a plurality of service scenes in the user service system by the service information processing system in the preset time according to the flow respectively accepted by the plurality of service scenes in the preset time and the unit flow income data respectively corresponding to the plurality of service scenes.
9. The method for allocating resources according to claim 8, wherein the obtaining unit flow rate gain data corresponding to each of the plurality of service scenarios includes:
and obtaining unit flow gain data corresponding to the service scenes respectively according to the combination grades of the service policies executed by the service modules when the service information to be put is determined for the users accessing the service scenes, wherein the combination grades are the combination grades determined according to the grades of the service policies executed by the service modules respectively, and the grades are used for representing the condition grades of the computing resource data used when the service policies are executed.
10. The method for allocating resources according to claim 9, wherein the obtaining unit flow rate gain data corresponding to each of the plurality of service scenarios according to the combination level of service policies executed when the plurality of service modules determine service information to be put for users accessing the service scenarios includes:
obtaining reference revenue data obtained by putting the service information to be put determined by the plurality of service modules into the service scene receiving unit flow when the computing resource data used by the plurality of service modules for executing the service policies of the reference combination level meets the preset computing resource conditions through a heuristic iteration method;
and determining the reference revenue data corresponding to the service scenes as unit flow revenue data corresponding to the service scenes.
11. The resource allocation method according to claim 7, further comprising:
obtaining, according to target flows respectively accepted by the multiple service scenes in the preset time and computing resource data usable by the service information processing system, reference combination levels of service policies executed by the multiple service scenes when the profit data obtained by the service information processing system for delivering service information to the multiple service scenes in the user service system in the preset time meets preset profit conditions, wherein the reference flows respectively accepted by the multiple service scenes in the preset time and/or the reference combination levels of service policies executed by the multiple service modules when determining service information to be delivered for users accessing the service scenes.
12. The resource allocation method according to claim 11, further comprising:
according to the calculation resource data required to be distributed for a plurality of service modules in the service information processing system respectively when the profit data obtained by putting service information into a plurality of service scenes in the user service system within the preset time meets the preset profit condition, the reference flow respectively accepted by the plurality of service scenes within the preset time and the reference combination grade of the service strategy executed when the plurality of service modules determine the service information to be put for a user accessing the service scenes, the calculation resource data respectively usable by the plurality of service modules, the flow respectively accepted by the plurality of service scenes within the preset time and the grade of the service strategy respectively executable when the plurality of service modules determine the service information to be put for the user accessing the service scenes are adjusted.
13. A resource allocation apparatus, comprising:
the system comprises a first target flow obtaining unit, a second target flow obtaining unit and a control unit, wherein the first target flow obtaining unit is used for obtaining target flow accepted by a service scene in a preset time, and the service scene is a service scene for providing service information for a user in a user service system;
The first computing resource data obtaining unit is used for obtaining computing resource data usable by a service information processing system, and the service information processing system is used for throwing service information which can be processed by a user into a service scene in the user service system;
the first computing resource allocation unit is used for obtaining computing resource data required to be allocated for a service module in the service information processing system when the benefit data obtained by putting service information into the service scene in the user service system in the preset time meets a preset benefit condition according to the target flow accepted by the service scene in the preset time and the computing resource data usable by the service information processing system, wherein the service module is used for determining service information to be put into for a user accessing the service scene.
14. A method for resource allocation, comprising:
obtaining scene flow information which is sent by a client and used for representing target flow accepted by a service scene in preset time and system resource information which is used for representing computing resource data which can be used by a service information processing system, wherein the service scene is a service scene which provides service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system;
Obtaining calculation resource data required to be distributed for a service module in the service information processing system when the income data obtained by putting service information into a service scene in the user service system in the preset time meets a preset income condition according to the scene flow information and the system resource information, wherein the service module is used for determining service information to be put into for a user accessing the service scene;
and returning resource allocation information for indicating calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition to the client.
15. A resource allocation apparatus, comprising:
the system comprises a first information obtaining unit, a second information obtaining unit and a service information processing unit, wherein the first information obtaining unit is used for obtaining scene flow information which is sent by a client and used for representing target flow accepted by a service scene in preset time, and system resource information which is used for representing calculation resource data which can be used by the service information processing system, the service scene is a service scene which provides service information for a user in a user service system, and the service information processing system is used for throwing the service information which can be processed by the user into the service scene in the user service system;
The second computing resource allocation unit is used for obtaining computing resource data required to be allocated for a service module in the service information processing system when the benefit data obtained by putting the service information into the service scene in the user service system in the preset time meets a preset benefit condition according to the scene flow information and the system resource information, and the service module is used for determining service information to be put for a user accessing the service scene;
and the first information sending unit is used for sending resource allocation information for representing calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition to the client.
16. A method for resource allocation, comprising:
displaying an information setting interface for setting target flow accepted by a service scene in a preset time and computing resource data usable by a service information processing system, wherein the service scene is a service scene for providing service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system;
In response to detecting the target flow accepted by the service scene set by the information setting interface in the preset time and the computing resource data usable by the service information processing system, obtaining scene flow information for representing the target flow accepted by the service scene in the preset time and system resource information for representing the computing resource data usable by the service information processing system;
the scene flow information and the system resource information are sent to a server;
acquiring resource allocation information which is returned by the service end and used for representing calculation resource data required to be allocated for a service module in the service information processing system when the profit data obtained by putting service information into a service scene in the user service system in the preset time meets a preset profit condition;
and according to the resource allocation information, displaying the calculation resource data required to be allocated for the service module in the service information processing system when the profit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset profit condition.
17. A resource allocation apparatus, comprising:
the system comprises a first display unit, a second display unit and a service information processing system, wherein the first display unit is used for displaying an information setting interface for setting target flow accepted by a service scene in preset time and computing resource data usable by the service information processing system, the service scene is a service scene for providing service information for a user in a user service system, and the service information processing system is used for throwing service information which can be processed by the user into the service scene in the user service system;
a second information obtaining unit configured to obtain, in response to detection of a target flow rate received by a service scenario set through the information setting interface within a preset time and calculation resource data usable by a service information processing system, scenario flow rate information for representing the target flow rate received by the service scenario within the preset time, and system resource information for representing the calculation resource data usable by the service information processing system;
the second information sending unit is used for sending the scene flow information and the system resource information to a server;
a third information obtaining unit, configured to obtain resource allocation information returned by the server, where the resource allocation information is used to indicate computing resource data that needs to be allocated to a service module in the service information processing system when revenue data obtained by putting service information into a service scenario in the user service system in the preset time meets a preset revenue condition;
And the second display unit is used for displaying the calculation resource data required to be distributed for the service module in the service information processing system when the benefit data obtained by putting the service information into the service scene in the user service system in the preset time meets the preset benefit condition according to the resource distribution information.
18. A method for resource allocation, comprising:
obtaining the target flow received by a target scene in a preset time;
obtaining computing resource data usable by an information processing system, wherein the information processing system is used for throwing information into the target scene;
according to the target flow received by the target scene in the preset time and the calculation resource data usable by the information processing system, when the income data obtained by the information processing system for throwing information into the target scene in the preset time meets the preset income condition, calculating resource data required to be distributed for a processing module in the information processing system, wherein the processing module is used for determining target information for a user accessing the target scene.
19. A resource allocation apparatus, comprising:
The second target flow obtaining unit is used for obtaining the target flow received by the target scene in the preset time;
the second computing resource data obtaining unit is used for obtaining computing resource data usable by an information processing system, and the information processing system is used for throwing information into the target scene;
the third computing resource allocation unit is used for obtaining computing resource data required to be allocated to a processing module in the information processing system when the benefit data obtained by the information processing system throwing information into the target scene in the preset time meets the preset benefit condition according to the target flow received by the target scene in the preset time and the computing resource data usable by the information processing system, and the processing module is used for determining target information for a user accessing the target scene.
20. An electronic device comprising a processor and a memory;
the memory is used for storing programs and data, and the processor invokes the programs stored in the memory to execute the resource allocation method according to any one of claims 1 to 12, 14, 16, 18.
21. A storage medium storing a program and data, the program being executable by a processor for implementing the resource allocation method of any one of claims 1-12, 14, 16, 18.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310908333.3A CN117149405A (en) | 2023-07-21 | 2023-07-21 | Resource allocation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310908333.3A CN117149405A (en) | 2023-07-21 | 2023-07-21 | Resource allocation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117149405A true CN117149405A (en) | 2023-12-01 |
Family
ID=88904966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310908333.3A Pending CN117149405A (en) | 2023-07-21 | 2023-07-21 | Resource allocation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117149405A (en) |
-
2023
- 2023-07-21 CN CN202310908333.3A patent/CN117149405A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6254283B2 (en) | Method and system for creating a data-driven attribution model that assigns attribution achievements to multiple events | |
| KR101807755B1 (en) | Multiple attribution models with return on ad spend | |
| TWI603273B (en) | Method and device for placing information search | |
| US8732015B1 (en) | Social media pricing engine | |
| CN104281962B (en) | For the advertisement in on-line system and the united market of content | |
| US20120046996A1 (en) | Unified data management platform | |
| US20100324993A1 (en) | Promotional content presentation based on search query | |
| US8645199B1 (en) | Using application characteristics for ad pricing | |
| CN104035926B (en) | A kind of dispensing of internet information and system | |
| US20170061515A1 (en) | Systems and methods for setting allocations and prices for content in an online marketplace | |
| US20100049644A1 (en) | Content item slot scheduling | |
| US8775251B1 (en) | Allocating advertising budgets | |
| US20160210656A1 (en) | System for marketing touchpoint attribution bias correction | |
| CN106407210A (en) | Display method and device of business object | |
| US20140372202A1 (en) | Predicting performance of content items using loss functions | |
| CN108777701A (en) | A kind of method and device of determining receiver | |
| US20120260185A1 (en) | Path length selector | |
| US10304081B1 (en) | Yielding content recommendations based on serving by probabilistic grade proportions | |
| US20210382952A1 (en) | Web content organization and presentation techniques | |
| US20240112210A1 (en) | Self-learning valuation | |
| WO2013116105A1 (en) | Alterations of calculations in attribution modeling | |
| KR20200113370A (en) | How to Implement a Sales Forecasting System on the Broadcast of YouTube Creators for Sale of Goods | |
| US10275795B1 (en) | System and method for ranking, allocation and pricing of content recommendations | |
| CN117149405A (en) | Resource allocation method and device | |
| KR20120139217A (en) | System of advertisement and method for managing advertisement |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |