CN104980517A - Snapshot-based cluster sensing system, method and device - Google Patents
Snapshot-based cluster sensing system, method and device Download PDFInfo
- Publication number
- CN104980517A CN104980517A CN201510366434.8A CN201510366434A CN104980517A CN 104980517 A CN104980517 A CN 104980517A CN 201510366434 A CN201510366434 A CN 201510366434A CN 104980517 A CN104980517 A CN 104980517A
- Authority
- CN
- China
- Prior art keywords
- subsystem
- reset
- retrieval
- estimate
- snapshot file
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/11—File system administration, e.g. details of archiving or snapshots
- G06F16/128—Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/14—Details of searching files based on file metadata
- G06F16/148—File search processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Abstract
The invention relates to a snapshot-based cluster sensing system, method and device. The system includes an independent retrieval subsystem, an independent estimation subsystem and an independent rearrangement subsystem which are respectively connected with a routing scheduling subsystem and a push service subsystem; the routing scheduling subsystem is used for obtaining the static snapshot files of the retrieval subsystem, the estimation subsystem and the rearrangement subsystem and updating the operation and maintenance configuration information of the subsystems according to the static snapshot files; the subsystems report operation state information; and the routing scheduling subsystem is used for obtaining the dynamic snapshot files of the subsystems and updating the algorithms of the subsystems according to the dynamic snapshot files. According to the above snapshot-based cluster sensing system, method and device, functions of the subsystems are isolated from each other and are independent from each other; the operation and maintenance configuration information and the operation state information of the subsystems are sensed through the dynamic snapshot files; and therefore, the whole system has the advantages of excellent expansibility, improved automation degree and decreased operation and maintenance cost.
Description
Technical field
The present invention relates to network application field, particularly relate to a kind of cluster sensory perceptual system, method and apparatus based on snapshot.
Background technology
Along with the development of network technology, increasing user is by Network Capture resource information.In order to meet the demand of user, push resource information to user by resource recommendation system.In resource recommendation system, complete once recommend service request to need to estimate through resource retrieval, resource primary election, resource, the step such as resource rearrangement, each step needs perform inquiry and calculate.The high concurrency request brought to support the user of magnanimity and complicated business demand, necessarily require system to have good autgmentability.But the subsystems in traditional resource recommendation system is associated by Static Configuration Files, the service that scalable appearance and config update can have influence on other subsystems is carried out to one of them subsystem, cause the O&M cost of system high, and interdepend between subsystem, add new subsystem and New function very complicated, the autgmentability of system is very poor.
Summary of the invention
Based on this, be necessary the high and problem of autgmentability difference of O&M cost for traditional resource recommendation system, a kind of cluster sensory perceptual system based on snapshot be provided, can O&M cost be reduced, and can autgmentability be improved.
In addition, there is a need to provide a kind of cluster cognitive method based on snapshot, can O&M cost be reduced, and can autgmentability be improved.
In addition, there is a need to provide a kind of cluster sensing device based on snapshot, can O&M cost be reduced, and can autgmentability be improved.
Based on a cluster sensory perceptual system for snapshot, comprise independently retrieval subsystem, estimate subsystem, reset subsystem, Push Service subsystem and routing scheduling subsystem; Described retrieval subsystem, estimate subsystem with rearrangement subsystem be connected with described routing scheduling subsystem respectively; Described retrieval subsystem, estimate subsystem, reset subsystem be connected with described Push Service subsystem respectively with routing scheduling subsystem;
Described Push Service subsystem is described retrieval subsystem, estimate subsystem and reset subsystem pushes algorithm required separately respectively;
Described retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file respectively and upgrade, described static snapshot file is according to retrieval subsystem, estimates subsystem and reset subsystem O&M configuration information separately and generate;
Described routing scheduling subsystem for obtaining retrieval subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Described retrieval subsystem, estimate subsystem and reset subsystem and report running state information respectively, to generate dynamic snapshot file according to retrieval subsystem, the running state information estimating subsystem and reset subsystem;
Described routing scheduling subsystem for obtaining described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
Based on a cluster cognitive method for snapshot, comprise the following steps:
Independently retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file, and upgrade according to static snapshot file, the static snapshot file of each subsystem generates according to the O&M configuration information of each subsystem;
Obtain described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Described retrieval subsystem, estimate subsystem and reset subsystem and obtain separately required algorithm from Push Service subsystem and load, report running state information again, to generate dynamic snapshot file according to described retrieval subsystem, the running state information estimating subsystem and reset subsystem;
Obtain described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
Based on a cluster cognitive method for snapshot, comprise the following steps:
Obtain independently retrieval subsystem, estimate subsystem and reset subsystem O&M configuration information separately;
Generate retrieval subsystem according to described O&M configuration information, estimate subsystem and reset subsystem static snapshot file separately;
Described static snapshot file sent to retrieval subsystem, estimates subsystem, reset subsystem and routing subsystem, making described retrieval subsystem, estimate subsystem, reset subsystem and upgrade according to respective static snapshot file, and described routing scheduling subsystem according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Receive described retrieval subsystem, estimate subsystem and reset the running state information that reports of subsystem, generate respective dynamic snapshot file according to described running state information;
By described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem and send to described routing scheduling subsystem, with make described routing scheduling subsystem according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
Based on a cluster sensing device for snapshot, comprising:
Static information update module, for by independently retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file, and upgrade according to static snapshot file, the static snapshot file of each subsystem generates according to the O&M configuration information of each subsystem, and obtain described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Reporting module, for by described retrieval subsystem, estimate subsystem and reset subsystem and obtain separately required algorithm from Push Service subsystem and load, report running state information again, to generate dynamic snapshot file according to described retrieval subsystem, the running state information estimating subsystem and reset subsystem;
Multidate information update module, for obtaining described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
Based on a cluster sensing device for snapshot, comprising:
Acquisition module, for obtain independently retrieval subsystem, estimate subsystem and reset subsystem O&M configuration information separately;
Generation module, for generating retrieval subsystem according to described O&M configuration information, estimating subsystem and reset subsystem static snapshot file separately;
Sending module, for described static snapshot file being sent to retrieval subsystem, estimating subsystem, resetting subsystem and routing subsystem, making described retrieval subsystem, estimate subsystem, reset subsystem and upgrade according to respective static snapshot file, and described routing scheduling subsystem according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Receiver module, for receiving described retrieval subsystem, estimates subsystem and resets the running state information that subsystem reports;
Described generation module is also for generating retrieval subsystem according to described running state information, estimating subsystem and reset subsystem dynamic snapshot file separately;
Described sending module also for by described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem and send to described routing scheduling subsystem, with make described routing scheduling subsystem according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
The above-mentioned cluster sensory perceptual system based on snapshot, method and apparatus, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost.
Accompanying drawing explanation
Fig. 1 is the internal structure schematic diagram of server in an embodiment;
Fig. 2 is the structured flowchart based on the cluster sensory perceptual system of snapshot in an embodiment;
Fig. 3 is the structured flowchart based on the cluster sensory perceptual system of snapshot in another embodiment;
Fig. 4 is the sequential chart that subsystem is estimated in the perception of routing scheduling subsystem;
Fig. 5 is the flow chart based on the cluster cognitive method of snapshot in an embodiment;
Fig. 6 is the flow chart based on the cluster cognitive method of snapshot in another embodiment;
Fig. 7 is the structured flowchart based on the cluster sensing device of snapshot in an embodiment;
Fig. 8 is the structured flowchart based on the cluster sensing device of snapshot in another embodiment;
Fig. 9 is the structured flowchart based on the cluster sensing device of snapshot in another embodiment.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Fig. 1 is the internal structure schematic diagram of server in an embodiment.As shown in Figure 1, this server comprises the processor, storage medium, internal memory and the network interface that are connected by system bus.Wherein, the storage medium of this server stores operating system, database and the cluster sensing device based on snapshot, O&M configuration information, running state information, static snapshot file and dynamic snapshot file etc. are stored in database, should based on the cluster sensing device of snapshot for realizing a kind of cluster cognitive method based on snapshot being applicable to server.The processor of this server is used for providing calculating and control ability, supports the operation of whole server.The operation saving as the cluster sensing device based on snapshot in storage medium in this server provides environment.The network interface of this server is used for being connected by network with the terminal of outside according to this communicating, the business recommended request of such as receiving terminal transmission and return service related data etc. to terminal.Server can realize with the server cluster that independently server or multiple server form.It will be appreciated by those skilled in the art that, structure shown in Fig. 1, it is only the block diagram of the part-structure relevant to the application's scheme, do not form the restriction to the server that the application's scheme is applied thereon, concrete server can comprise than parts more or less shown in figure, or combine some parts, or there is different parts layouts.
Fig. 2 is the structured flowchart based on the cluster sensory perceptual system of snapshot in an embodiment.As shown in Figure 2, a kind of cluster sensory perceptual system based on snapshot, comprise independently retrieval subsystem 210, estimate subsystem 220, reset subsystem 230, Push Service subsystem 240 and routing scheduling subsystem 250.Retrieval subsystem 210, estimate subsystem 220 with rearrangement subsystem 230 be connected with routing scheduling subsystem 250 respectively; Retrieval subsystem 210, estimate subsystem 220, reset subsystem 230 be connected with Push Service subsystem 240 respectively with routing scheduling subsystem 250.
Retrieval subsystem 210, i.e. retrival+scoring subsystem, can carry out retrieve resources according to the directed information comprised in business recommended request, and carry out Preliminary screening and scoring process etc. to resource.Resource can be advertising resource, News Resources or article resource etc.This directed information can comprise the conditional informations such as crowd, sex, age, region, time.Crowd refer to resource for crowd's type, such as child, the young and the middle aged, the elderly etc.Sex can comprise man and female.Age can be divided into Different age group, as 0 to 3 year old, and 6 to 12 years old etc.Region can be divided into the zones of different in country variant, area or same country, such as Chinese Guangdong province, Hainan Province etc.Time refers to the time that resource is thrown in, be divided into ground floor time and second layer time, the ground floor time can be divided into Monday to week seven, and the second layer time can be divided into intraday multiple time period, as by one day to be per half an hour a timing node, be divided into 48 timing nodes.The algorithm that retrieval subsystem 210 is carrying out adopting Push Service subsystem 240 to provide in retrieving is retrieved.
Estimate subsystem 220, i.e. Pctr subsystem, for the basic representation data according to user's historical behavior data and user, and the scene information of resource-niche builds a forecast model, estimates out the prediction clicking rate of resource according to historical behavior data and forecast model.Resource-niche refers to resource location information, such as advertisement position information.The basic representation data of user refers to the personal information of user, can comprise name, sex, age, region etc.Estimate the algorithm that subsystem 220 adopts Push Service subsystem 240 to provide in the course of the work and carry out work.
Reset subsystem 230, i.e. Reranking subsystem, for screening resource according to the stereotactic conditions preset, and the prediction clicking rate of resource being revised, according to revised prediction clicking rate, resource being resequenced.Particularly, stereotactic conditions can set as required, such as, can set the conditions such as crowd, age, sex, region.Reset subsystem 230 to revise prediction clicking rate according to active user behavioral data.The algorithm that rearrangement subsystem 230 adopts Push Service subsystem 240 to provide at work carries out work.
Push server subsystem 240, i.e. PushServer subsystem, for for retrieval subsystem 210, estimate subsystem 220 and reset subsystem 230 pushes respectively separately needed for algorithm.In addition, also model service can be pushed.
Routing scheduling subsystem 250, i.e. Router subsystem, for dispatching retrieval subsystem 210 successively according to business recommended request, estimate subsystem 220, reset subsystem 230 and carry out work, and collect result.
Retrieval subsystem 210, estimate subsystem 220 and reset subsystem 230 and obtain respective static snapshot file respectively and upgrade, this static snapshot file is according to retrieval subsystem 210, estimates subsystem 220 and reset the respective O&M configuration information of subsystem 230 and generate.
Particularly, O&M configuration information can comprise the algorithm information etc. that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load.Static snapshot file is used for the O&M configuration information of descriptor system, the O&M configuration information of subsystem comprises the algorithm information that the number of machines of subsystem configures, machine IP (InternetProtocol, procotol) address and port, every platform machine will load.The network address of machine IP address and machine.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Routing scheduling subsystem 250 for obtaining retrieval subsystem 210, estimate subsystem 220 and reset the static snapshot file of subsystem 230, according to this static snapshot file update retrieval subsystem 210, estimate subsystem 220 and reset the O&M configuration information of subsystem 230.
Particularly, upgrade O&M configuration information and refer to the machine that may increase or delete subsystem, new interpolation algorithm so (i.e. algorithm dynamic base) or deletion algorithm dynamic base etc.
Retrieval subsystem 210, estimate subsystem 220 and reset subsystem 230 and report running state information respectively, to generate dynamic snapshot file according to retrieval subsystem 210, the running state information estimating subsystem 220 and reset subsystem 230.
Particularly, the running state information algorithm information that comprises the network address of machine when subsystem runs and port, machine state and loaded.Machine state can comprise the parameter etc. when machine runs.Dynamic snapshot file is used for information during descriptor system cloud gray model, the algorithm information that information during operation comprises the network address of machine when subsystem runs and port, machine state and loaded.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Routing scheduling subsystem 250 for obtaining this retrieval subsystem 210, estimate subsystem 220 and reset the dynamic snapshot file of subsystem 230, according to this retrieval subsystem 210 of this dynamic snapshot file update, estimate subsystem 220 and reset the algorithm information of subsystem 230.In addition, routing scheduling subsystem 250 also for according to dynamic snapshot file update retrieval subsystem 210, estimate subsystem 220 and reset the server etc. of subsystem 230.
The above-mentioned cluster sensory perceptual system based on snapshot, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost.
Fig. 3 is the structured flowchart based on the cluster sensory perceptual system of snapshot in another embodiment.As shown in Figure 3, should based on cluster sensory perceptual system of snapshot comprise independently retrieval subsystem 210, estimate subsystem 220, reset subsystem 230, Push Service subsystem 240, routing scheduling subsystem 250, configuration management server 260 and distributed coordination server 270.Retrieval subsystem 210, estimate subsystem 220 with rearrangement subsystem 230 be connected with routing scheduling subsystem 250 respectively; Retrieval subsystem 210, estimate subsystem 220, reset subsystem 230 be connected with Push Service subsystem 240 respectively with routing scheduling subsystem 250.Configuration management server 260 is connected with distributed coordination server 270, distributed coordination server 270 respectively with retrieval subsystem 210, estimate subsystem 220, reset subsystem 230 and be connected with routing scheduling subsystem 250.
Configuration management server 260 is for obtaining retrieval subsystem 210, estimating the O&M configuration information of subsystem 220 and rearrangement subsystem 230, and according to this retrieval subsystem 210, the static snapshot file estimating the O&M configuration information generation correspondence of subsystem 220 and rearrangement subsystem 230, and receive this retrieval subsystem 210, estimate subsystem 220 and reset the running state information that reports of subsystem 230, and generate corresponding dynamic snapshot file according to this retrieval subsystem 210, the running state information estimating subsystem 220 and reset subsystem 230.
Distributed coordination server 270 for recording this retrieval subsystem 210, estimating static snapshot file and the dynamic snapshot file of subsystem 220 and rearrangement subsystem 230, and stores this retrieval subsystem 210, estimates O&M configuration information and the running state information of subsystem 220 and rearrangement subsystem 230.Static snapshot file and dynamic snapshot file is stored in distributed coordination server 270.
Distributed coordination server 270 obtains retrieval subsystem 210, estimate subsystem 220 and reset the running state information that subsystem 230 reports, and by retrieval subsystem 210, estimate subsystem 220 and reset the running state information that subsystem 230 reports and send to configuration management server 260.
In addition, experiment subsystem 280 is also comprised, for carrying out experiment detection and Algorithm for Training etc.
In other embodiments, configuration management server 260 and distributed coordination server 270 can be a server.Configuration management server 260 can record this retrieval subsystem 210, estimate subsystem 220 and reset static snapshot file and the dynamic snapshot file of subsystem 230, and stores this retrieval subsystem 210, estimates subsystem 220 and reset O&M configuration information and the running state information of subsystem 230.
The above-mentioned cluster sensory perceptual system based on snapshot, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost, static snapshot file is generated according to O&M configuration information by configuration management server, and generate dynamic snapshot file according to running state information, facilitate Push Service subsystem and routing scheduling subsystem to obtain static snapshot file and dynamic snapshot file, perception retrieval subsystem, estimate subsystem and reset the configuration variation situation of subsystem.
In one embodiment, routing scheduling subsystem 250 is also for obtaining business recommended request, dispatch this retrieval subsystem 210 successively according to this business recommended request, estimate subsystem 220 and reset subsystem 230 carry out work, this retrieval subsystem 210, estimate subsystem 220 and reset subsystem 230 load in the course of the work this Push Service subsystem 240 push corresponding algorithm.This routing scheduling subsystem 250 is also for obtaining this retrieval subsystem 210, estimating the static snapshot file of subsystem 220 and rearrangement subsystem 230, according to this retrieval subsystem 210 of this static snapshot file update, the O&M configuration information estimating subsystem 220 and rearrangement subsystem 230, and obtain this retrieval subsystem 210, estimate subsystem 220 and reset the dynamic snapshot file of subsystem 230, according to this retrieval subsystem 210 of this dynamic snapshot file update, estimate subsystem 220 and reset the algorithm of subsystem 230.
Particularly, routing scheduling subsystem 250, Push Service subsystem 240 can perceive retrieval subsystem 210 according to dynamic snapshot file, estimate the machine distribution of subsystem 220 and rearrangement subsystem 230 and algorithm load condition.
In order to further illustrate the course of work of the cluster sensory perceptual system based on snapshot, to estimate subsystem 220 and routing scheduling subsystem 250 is described the process that subsystem 220 is estimated in routing scheduling subsystem 250 perception.Fig. 4 is the sequential chart that subsystem 220 is estimated in routing scheduling subsystem 250 perception.As shown in Figure 4, the detailed process that subsystem 220 is estimated in routing scheduling subsystem 250 perception comprises:
(1) configuration management server 260 obtains the O&M configuration information estimating subsystem 220, and arranging O&M configuration information is that each subsystem generates static snapshot file, and by static snapshot file write distributed coordination server 270.
(2) estimate subsystem 220 and pull new static snapshot file from distributed coordination server 270, and upgrade according to static snapshot file.
(3) routing scheduling subsystem 250 pulls up-to-date static snapshot file from distributed coordination server 270, estimates the O&M configuration information of subsystem 220 according to static snapshot file update.
(4) estimating subsystem 220 reports running state information to arrive distributed coordination server 270.
(5) distributed coordination server 270 sends to configuration management server 260 by estimating the running state information that subsystem 220 reports.
(6) configuration management server 260 timing arranges the running state information estimating subsystem 220 and generates dynamic snapshot file, and by dynamic snapshot file write distributed coordination server 270.
(7) routing scheduling subsystem 250 gets the dynamic snapshot file estimating subsystem 220 from distributed coordination server 270, estimates the algorithm etc. of subsystem 220 according to dynamic snapshot file update.
The above-mentioned cluster sensory perceptual system based on snapshot, routing scheduling subsystem 250 and Push Service subsystem 240 perception retrieval subsystem 210, to reset the processes such as subsystem 230 the same, do not repeat them here.
Fig. 5 is the flow chart based on the cluster cognitive method of snapshot in an embodiment.As shown in Figure 5, a kind of cluster cognitive method based on snapshot, comprises the following steps:
Step 502, independently retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file, and upgrade according to static snapshot file, the static snapshot file of each subsystem generates according to the O&M configuration information of each subsystem.
Particularly, retrieval subsystem 210 can carry out retrieve resources according to the directed information comprised in business recommended request, and carries out Preliminary screening and scoring process etc. to resource.Resource can be advertising resource, News Resources or article resource etc.This directed information can comprise the conditional informations such as crowd, sex, age, region, time.Crowd refer to resource for crowd's type, such as child, the young and the middle aged, the elderly etc.Sex can comprise man and female.Age can be divided into Different age group, as 0 to 3 year old, and 6 to 12 years old etc.Region can be divided into the zones of different in country variant, area or same country, such as Chinese Guangdong province, Hainan Province etc.Time refers to the time that resource is thrown in, be divided into ground floor time and second layer time, the ground floor time can be divided into Monday to week seven, and the second layer time can be divided into intraday multiple time period, as by one day to be per half an hour a timing node, be divided into 48 timing nodes.The algorithm that retrieval subsystem 210 is carrying out adopting Push Service subsystem 240 to provide in retrieving is retrieved.
Estimate subsystem 220 for the basic representation data according to user's historical behavior data and user, and the scene information of resource-niche builds a forecast model, estimate out the prediction clicking rate of resource according to historical behavior data and forecast model.Resource-niche refers to resource location information, such as advertisement position information.The basic representation data of user refers to the personal information of user, can comprise name, sex, age, region etc.Estimate the algorithm that subsystem 220 adopts Push Service subsystem 240 to provide in the course of the work and carry out work.
Reset subsystem 230 for screening resource according to the stereotactic conditions preset, and the prediction clicking rate of resource is revised, according to revised prediction clicking rate, resource is resequenced.Particularly, stereotactic conditions can set as required, such as, can set the conditions such as crowd, age, sex, region.Reset subsystem 230 to revise prediction clicking rate according to active user behavioral data.The algorithm that rearrangement subsystem 230 adopts Push Service subsystem 240 to provide at work carries out work.
Retrieval subsystem 210, estimate subsystem 220, reset subsystem 230 according to the O&M configuration information of respective static snapshot file update self.
O&M configuration information can comprise the algorithm information etc. that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load.Static snapshot file is used for the O&M configuration information of descriptor system, the algorithm information that O&M configuration information comprises the number of machines of subsystem configures, machine IP address and port, every platform machine will load of subsystem.The network address of machine IP address and machine.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Step 504, obtains this retrieval subsystem by routing scheduling subsystem, estimates the static snapshot file of subsystem and rearrangement subsystem, according to this this retrieval subsystem of static snapshot file update, the O&M configuration information estimating subsystem and rearrangement subsystem.
Particularly, upgrade O&M configuration information and refer to the machine that may increase or delete subsystem, new interpolation algorithm so (i.e. algorithm dynamic base) or deletion algorithm dynamic base etc.
Step 506, retrieval subsystem, estimate subsystem and reset subsystem and obtain separately required algorithm from Push Service subsystem and load, report running state information again, to generate dynamic snapshot file according to this retrieval subsystem, the running state information estimating subsystem and reset subsystem.
Particularly, the running state information algorithm information that comprises the network address of machine when subsystem runs and port, machine state and loaded.Machine state can comprise the parameter etc. when machine runs.Dynamic snapshot file is used for information during descriptor system cloud gray model, the algorithm information that information during operation comprises the network address of machine when subsystem runs and port, machine state and loaded.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Step 508, obtains this retrieval subsystem by routing scheduling subsystem, estimates the dynamic snapshot file of subsystem and rearrangement subsystem, according to this this retrieval subsystem of dynamic snapshot file update, the algorithm estimating subsystem and rearrangement subsystem.
The above-mentioned cluster cognitive method based on snapshot, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost.
In one embodiment, when the above-mentioned cluster cognitive method based on snapshot is applied to business recommended, further comprising the steps of: to obtain business recommended request; Dispatch retrieval subsystem successively according to this business recommended request, estimate subsystem and reset subsystem carry out work, wherein, this retrieval subsystem, estimate subsystem and reset subsystem load in the course of the work Push Service subsystem push corresponding algorithm; Obtain this retrieval subsystem, estimate the static snapshot file of subsystem and rearrangement subsystem, according to this this retrieval subsystem of static snapshot file update, the O&M configuration information estimating subsystem and rearrangement subsystem, and obtain this retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to this this retrieval subsystem of dynamic snapshot file update, estimate subsystem and reset the algorithm of subsystem.
Particularly, routing scheduling subsystem 250, Push Service subsystem 240 can perceive retrieval subsystem 210 according to dynamic snapshot file, estimate the machine distribution of subsystem 220 and rearrangement subsystem 230 and algorithm load condition.
Fig. 6 is the flow chart based on the cluster cognitive method of snapshot in another embodiment.The cluster cognitive method based on snapshot in Fig. 6, is described with configuration management server angle.As shown in Figure 6, a kind of cluster cognitive method based on snapshot, comprises the following steps:
Step 602, obtain independently retrieval subsystem, estimate subsystem and reset subsystem O&M configuration information separately.
Particularly, O&M configuration information can comprise the algorithm information etc. that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load.
Step 604, generates retrieval subsystem according to this O&M configuration information, estimates subsystem and reset subsystem static snapshot file separately.
Particularly, static snapshot file is used for the O&M configuration information of descriptor system, the algorithm information that O&M configuration information comprises the number of machines of subsystem configures, machine IP address and port, every platform machine will load of subsystem.The network address of machine IP address and machine.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Step 606, this static snapshot file sent to retrieval subsystem, estimates subsystem, reset subsystem and routing subsystem, making this retrieval subsystem, estimate subsystem, reset subsystem and upgrade according to respective static snapshot file, and by routing scheduling subsystem according to this this retrieval subsystem of static snapshot file update, estimate subsystem and reset the O&M configuration information of subsystem.
Particularly, upgrade O&M configuration information and refer to the machine that may increase or delete subsystem, new interpolation algorithm so (i.e. algorithm dynamic base) or deletion algorithm dynamic base etc.
Step 608, receives this retrieval subsystem, estimates subsystem and reset the running state information that reports of subsystem, generate respective dynamic snapshot file according to this running state information.
Particularly, the running state information algorithm information that comprises the network address of machine when subsystem runs and port, machine state and loaded.Machine state can comprise the parameter etc. when machine runs.Dynamic snapshot file is used for information during descriptor system cloud gray model, the algorithm information that information during operation comprises the network address of machine when subsystem runs and port, machine state and loaded.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Step 610, by this retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem and send to this routing scheduling subsystem, to make this routing scheduling subsystem according to this this retrieval subsystem of dynamic snapshot file update, to estimate subsystem and reset the algorithm of subsystem.
The above-mentioned cluster cognitive method based on snapshot, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost.
Further, in one embodiment, the above-mentioned cluster cognitive method based on snapshot also comprises: record this retrieval subsystem, estimate subsystem and reset the static snapshot file of subsystem and dynamic snapshot file, and stores this retrieval subsystem, estimates subsystem and reset O&M configuration information and the running state information of subsystem.By record O&M configuration information, running state information, static snapshot file and dynamic snapshot file, conveniently pulled by each subsystem.
Fig. 7 is the structured flowchart based on the cluster sensing device of snapshot in an embodiment.As shown in Figure 7, a kind of cluster sensing device based on snapshot correspond in Fig. 5 based on snapshot cluster cognitive method constructed by virtual functions module, comprise static information update module 710, reporting module 720 and multidate information update module 730.Wherein:
Static information update module 710 for by independently retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file, and upgrade according to static snapshot file, the static snapshot file of each subsystem generates according to the O&M configuration information of each subsystem, and obtain this retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to this this retrieval subsystem of static snapshot file update, estimate subsystem and reset the O&M configuration information of subsystem.
Particularly, retrieval subsystem can carry out retrieve resources according to the directed information comprised in business recommended request, and carries out Preliminary screening and scoring process etc. to resource.Resource can be advertising resource, News Resources or article resource etc.This directed information can comprise the conditional informations such as crowd, sex, age, region, time.Crowd refer to resource for crowd's type, such as child, the young and the middle aged, the elderly etc.Sex can comprise man and female.Age can be divided into Different age group, as 0 to 3 year old, and 6 to 12 years old etc.Region can be divided into the zones of different in country variant, area or same country, such as Chinese Guangdong province, Hainan Province etc.Time refers to the time that resource is thrown in, be divided into ground floor time and second layer time, the ground floor time can be divided into Monday to week seven, and the second layer time can be divided into intraday multiple time period, as by one day to be per half an hour a timing node, be divided into 48 timing nodes.The algorithm that retrieval subsystem is carrying out adopting Push Service subsystem to provide in retrieving is retrieved.
Estimate subsystem for the basic representation data according to user's historical behavior data and user, and the scene information of resource-niche builds a forecast model, estimate out the prediction clicking rate of resource according to historical behavior data and forecast model.Resource-niche refers to resource location information, such as advertisement position information.The basic representation data of user refers to the personal information of user, can comprise name, sex, age, region etc.Estimate the algorithm that subsystem adopts Push Service subsystem to provide in the course of the work and carry out work.
Reset subsystem to be used for screening resource according to the stereotactic conditions preset, and the prediction clicking rate of resource is revised, according to revised prediction clicking rate, resource is resequenced.Particularly, stereotactic conditions can set as required, such as, can set the conditions such as crowd, age, sex, region.Reset subsystem to revise prediction clicking rate according to active user behavioral data.The algorithm that rearrangement subsystem adopts Push Service subsystem to provide at work carries out work.
Retrieval subsystem, estimate subsystem, reset subsystem according to the O&M configuration information of respective static snapshot file update self.
O&M configuration information can comprise the algorithm information etc. that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load.Static snapshot file is used for the O&M configuration information of descriptor system, the algorithm information that O&M configuration information comprises the number of machines of subsystem configures, machine IP address and port, every platform machine will load of subsystem.The network address of machine IP address and machine.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Upgrade O&M configuration information and refer to the machine that may increase or delete subsystem, new interpolation algorithm so (i.e. algorithm dynamic base) or deletion algorithm dynamic base etc.
Reporting module 720 for by this retrieval subsystem, estimate subsystem and reset subsystem and obtain separately required algorithm from Push Service subsystem and load, report running state information again, to generate dynamic snapshot file according to this retrieval subsystem, the running state information estimating subsystem and reset subsystem.
Particularly, the running state information algorithm information that comprises the network address of machine when subsystem runs and port, machine state and loaded.Machine state can comprise the parameter etc. when machine runs.Dynamic snapshot file is used for information during descriptor system cloud gray model, the algorithm information that information during operation comprises the network address of machine when subsystem runs and port, machine state and loaded.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Multidate information update module 730 for obtaining this retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to this this retrieval subsystem of dynamic snapshot file update, estimate subsystem and reset the algorithm of subsystem.
The above-mentioned cluster sensing device based on snapshot, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost.
Fig. 8 is the structured flowchart based on the cluster sensing device of snapshot in another embodiment.As shown in Figure 8, a kind of cluster sensing device based on snapshot runs on configuration management server, for in Fig. 6 based on the virtual functions module constructed by the cluster cognitive method of snapshot, comprise acquisition module 810, generation module 820, sending module 830 and receiver module 840.Wherein:
Acquisition module 810 for obtain independently retrieval subsystem, estimate subsystem and reset subsystem O&M configuration information separately.
Particularly, O&M configuration information can comprise the algorithm information etc. that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load.
Generation module 820 is for generating retrieval subsystem according to this O&M configuration information, estimating subsystem and reset subsystem static snapshot file separately.
Particularly, static snapshot file is used for the O&M configuration information of descriptor system, the algorithm information that O&M configuration information comprises the number of machines of subsystem configures, machine IP address and port, every platform machine will load of subsystem.The network address of machine IP address and machine.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Sending module 830 is for sending to retrieval subsystem by this static snapshot file, estimating subsystem, resetting subsystem and routing subsystem, making this retrieval subsystem, estimate subsystem, reset subsystem and upgrade according to respective static snapshot file, and this routing scheduling subsystem according to this this retrieval subsystem of static snapshot file update, estimate subsystem and reset the O&M configuration information of subsystem.
Particularly, upgrade O&M configuration information and refer to the machine that may increase or delete subsystem, new interpolation algorithm so (i.e. algorithm dynamic base) or deletion algorithm dynamic base etc.
Receiver module 840 for receiving this retrieval subsystem, estimate subsystem and reset the running state information that subsystem reports.
Generation module 820 is for generating respective dynamic snapshot file according to this running state information.
Particularly, the running state information algorithm information that comprises the network address of machine when subsystem runs and port, machine state and loaded.Machine state can comprise the parameter etc. when machine runs.Dynamic snapshot file is used for information during descriptor system cloud gray model, the algorithm information that information during operation comprises the network address of machine when subsystem runs and port, machine state and loaded.Algorithm information can comprise algorithm versions, algorithm details etc.Algorithm versions such as can be V1.0, V2.0 etc.Algorithm details can be used for describing the effect of algorithm and function etc.
Sending module 830 also for by this retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem and send to this routing scheduling subsystem, to make this routing scheduling subsystem according to this this retrieval subsystem of dynamic snapshot file update, to estimate subsystem and reset the algorithm of subsystem.
The above-mentioned cluster sensing device based on snapshot, function isolation between subsystems, do not interdepend, Push Service subsystem can provide corresponding algorithm Push Service according to the static snapshot file of each subsystem and dynamic snapshot file, need not change other subsystems, routing scheduling subsystem is by the O&M configuration information of each subsystem of dynamic snapshot file perception and running state information, whole system is made to have splendid autgmentability, routing scheduling subsystem United Dispatching, route service for whole system provides a kind of general and easy implementation, improve the automaticity of system, reduce O&M cost.
Fig. 9 is the structured flowchart based on the cluster sensing device of snapshot in another embodiment.As shown in Figure 9, a kind of cluster sensing device based on snapshot runs on configuration management server, except comprising acquisition module 810, generation module 820, sending module 830 and receiver module 840, also comprises memory module 850.Wherein:
Memory module 850 is for recording this retrieval subsystem, estimating static snapshot file and the dynamic snapshot file of subsystem and rearrangement subsystem, and store this retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem and running state information, this O&M configuration information comprises the algorithm information that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load; The algorithm information that this running state information comprises the network address of machine when subsystem runs and port, machine state and loaded.
By record O&M configuration information, running state information, static snapshot file and dynamic snapshot file, conveniently pulled by each subsystem.
One of ordinary skill in the art will appreciate that all or part of flow process realized in above-described embodiment method, that the hardware that can carry out instruction relevant by computer program has come, described program can be stored in a non-volatile computer read/write memory medium, this program, when performing, can comprise the flow process of the embodiment as above-mentioned each side method.Wherein, described storage medium can be magnetic disc, CD, read-only store-memory body (Read-OnlyMemory, ROM) etc.
The above embodiment only have expressed several execution mode of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection range of patent of the present invention should be as the criterion with claims.
Claims (15)
1., based on a cluster sensory perceptual system for snapshot, it is characterized in that, comprise independently retrieval subsystem, estimate subsystem, reset subsystem, Push Service subsystem and routing scheduling subsystem; Described retrieval subsystem, estimate subsystem with rearrangement subsystem be connected with described routing scheduling subsystem respectively; Described retrieval subsystem, estimate subsystem, reset subsystem be connected with described Push Service subsystem respectively with routing scheduling subsystem;
Described Push Service subsystem is described retrieval subsystem, estimate subsystem and reset subsystem pushes algorithm required separately respectively;
Described retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file respectively and upgrade, described static snapshot file is according to retrieval subsystem, estimates subsystem and reset subsystem O&M configuration information separately and generate;
Described routing scheduling subsystem for obtaining retrieval subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Described retrieval subsystem, estimate subsystem and reset subsystem and report running state information respectively, to generate dynamic snapshot file according to retrieval subsystem, the running state information estimating subsystem and reset subsystem;
Described routing scheduling subsystem for obtaining described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
2. system according to claim 1, it is characterized in that, described system also comprises configuration management server, described configuration management server is for obtaining described retrieval subsystem, estimate the O&M configuration information of subsystem and rearrangement subsystem, and according to described retrieval subsystem, the O&M configuration information estimating subsystem and rearrangement subsystem generates corresponding static snapshot file, and receive described retrieval subsystem, estimate the running state information that subsystem reports with rearrangement subsystem, and according to described retrieval subsystem, the running state information estimating subsystem and rearrangement subsystem generates corresponding dynamic snapshot file.
3. system according to claim 2, it is characterized in that, described system also comprises distributed coordination server, described distributed coordination server for recording described retrieval subsystem, estimating static snapshot file and the dynamic snapshot file of subsystem and rearrangement subsystem, and stores described retrieval subsystem, estimates O&M configuration information and the running state information of subsystem and rearrangement subsystem.
4. system according to claim 3, it is characterized in that, described routing scheduling subsystem is also for obtaining business recommended request, according to described business recommended request successively execution cost retrieval subsystem, estimate subsystem and reset subsystem carry out work, described retrieval subsystem, estimate subsystem and reset subsystem load in the course of the work described Push Service subsystem push corresponding algorithm; Described routing scheduling subsystem is also for obtaining described retrieval subsystem, estimating the static snapshot file of subsystem and rearrangement subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem, and obtain described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
5. system according to claim 1, is characterized in that, described O&M configuration information comprises the algorithm information that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load; The algorithm information that described running state information comprises the network address of machine when subsystem runs and port, machine state and loaded.
6., based on a cluster cognitive method for snapshot, comprise the following steps:
Independently retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file, and upgrade according to static snapshot file, the static snapshot file of each subsystem generates according to the O&M configuration information of each subsystem;
Obtain described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Described retrieval subsystem, estimate subsystem and reset subsystem and obtain separately required algorithm from Push Service subsystem and load, report running state information again, to generate dynamic snapshot file according to described retrieval subsystem, the running state information estimating subsystem and reset subsystem;
Obtain described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
7. method according to claim 6, is characterized in that, described O&M configuration information comprises the algorithm information that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load; The algorithm information that described running state information comprises the network address of machine when subsystem runs and port, machine state and loaded.
8. method according to claim 6, is characterized in that, described method also comprises:
Obtain business recommended request;
Dispatch retrieval subsystem successively according to described business recommended request, estimate subsystem and reset subsystem and carry out work, wherein, described retrieval subsystem, estimate subsystem and reset subsystem load in the course of the work Push Service subsystem push corresponding algorithm;
Obtain described retrieval subsystem, estimate the static snapshot file of subsystem and rearrangement subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem, and obtain described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
9., based on a cluster cognitive method for snapshot, comprise the following steps:
Obtain independently retrieval subsystem, estimate subsystem and reset subsystem O&M configuration information separately;
Generate retrieval subsystem according to described O&M configuration information, estimate subsystem and reset subsystem static snapshot file separately;
Described static snapshot file sent to retrieval subsystem, estimates subsystem, reset subsystem and routing subsystem, making described retrieval subsystem, estimate subsystem, reset subsystem and upgrade according to respective static snapshot file, and described routing scheduling subsystem according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Receive described retrieval subsystem, estimate subsystem and reset the running state information that reports of subsystem, generate respective dynamic snapshot file according to described running state information;
By described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem and send to described routing scheduling subsystem, with make described routing scheduling subsystem according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
10. method according to claim 9, is characterized in that, described method also comprises:
Record described retrieval subsystem, estimate subsystem and reset the static snapshot file of subsystem and dynamic snapshot file, and store described retrieval subsystem, estimate subsystem and reset O&M configuration information and the running state information of subsystem.
11. methods according to claim 9, is characterized in that, described O&M configuration information comprises the algorithm information that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load; The algorithm information that described running state information comprises the network address of machine when subsystem runs and port, machine state and loaded.
12. 1 kinds, based on the cluster sensing device of snapshot, is characterized in that, comprising:
Static information update module, for by independently retrieval subsystem, estimate subsystem and reset subsystem and obtain respective static snapshot file, and upgrade according to static snapshot file, the static snapshot file of each subsystem generates according to the O&M configuration information of each subsystem, and obtain described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the static snapshot file of subsystem, according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Reporting module, for by described retrieval subsystem, estimate subsystem and reset subsystem and obtain separately required algorithm from Push Service subsystem and load, report running state information again, to generate dynamic snapshot file according to described retrieval subsystem, the running state information estimating subsystem and reset subsystem;
Multidate information update module, for obtaining described retrieval subsystem by routing scheduling subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem, according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
13. devices according to claim 12, is characterized in that, described O&M configuration information comprises the algorithm information that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load; The algorithm information that described running state information comprises the network address of machine when subsystem runs and port, machine state and loaded.
14. 1 kinds, based on the cluster sensing device of snapshot, is characterized in that, comprising:
Acquisition module, for obtain independently retrieval subsystem, estimate subsystem and reset subsystem O&M configuration information separately;
Generation module, for generating retrieval subsystem according to described O&M configuration information, estimating subsystem and reset subsystem static snapshot file separately;
Sending module, for described static snapshot file being sent to retrieval subsystem, estimating subsystem, resetting subsystem and routing subsystem, making described retrieval subsystem, estimate subsystem, reset subsystem and upgrade according to respective static snapshot file, and described routing scheduling subsystem according to described static snapshot file update retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem;
Receiver module, for receiving described retrieval subsystem, estimates subsystem and resets the running state information that subsystem reports;
Described generation module is also for generating retrieval subsystem according to described running state information, estimating subsystem and reset subsystem dynamic snapshot file separately;
Described sending module also for by described retrieval subsystem, estimate subsystem and reset the dynamic snapshot file of subsystem and send to described routing scheduling subsystem, with make described routing scheduling subsystem according to described dynamic snapshot file update retrieval subsystem, estimate subsystem and reset the algorithm of subsystem.
15. devices according to claim 14, is characterized in that, described device also comprises:
Memory module, for recording described retrieval subsystem, estimating static snapshot file and the dynamic snapshot file of subsystem and rearrangement subsystem, and store described retrieval subsystem, estimate subsystem and reset the O&M configuration information of subsystem and running state information, described O&M configuration information comprises the algorithm information that the number of machines of subsystem configures, the network address of every platform machine and port, every platform machine will load; The algorithm information that described running state information comprises the network address of machine when subsystem runs and port, machine state and loaded.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510366434.8A CN104980517B (en) | 2015-06-26 | 2015-06-26 | Cluster sensory perceptual system based on snapshot, method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510366434.8A CN104980517B (en) | 2015-06-26 | 2015-06-26 | Cluster sensory perceptual system based on snapshot, method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104980517A true CN104980517A (en) | 2015-10-14 |
| CN104980517B CN104980517B (en) | 2018-04-06 |
Family
ID=54276610
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510366434.8A Active CN104980517B (en) | 2015-06-26 | 2015-06-26 | Cluster sensory perceptual system based on snapshot, method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104980517B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107515883A (en) * | 2016-06-17 | 2017-12-26 | 阿里巴巴集团控股有限公司 | A kind of method and apparatus of operation task |
| CN110022220A (en) * | 2018-01-08 | 2019-07-16 | 腾讯科技（深圳）有限公司 | Routing Activiation method and system in business card recognition |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6976063B1 (en) * | 2000-11-02 | 2005-12-13 | Microsoft Corporation | Method and system for dynamically configuring a server computer |
| CN201654777U (en) * | 2009-04-22 | 2010-11-24 | 同方知网(北京)技术有限公司 | Dynamic data acquisition device |
| CN102096564A (en) * | 2009-12-14 | 2011-06-15 | 北大方正集团有限公司 | Job processing method for printer controller and printer controller |
| CN102722413A (en) * | 2012-05-16 | 2012-10-10 | 上海兆民云计算科技有限公司 | Distributed resource scheduling method for desktop cloud cluster |
| CN102857371A (en) * | 2012-08-21 | 2013-01-02 | 曙光信息产业(北京)有限公司 | Dynamic allocation management method for cluster system |
| CN104615661A (en) * | 2015-01-05 | 2015-05-13 | 华为技术有限公司 | Service recommendation method, device and system oriented to cloud platform applications |
-
2015
- 2015-06-26 CN CN201510366434.8A patent/CN104980517B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6976063B1 (en) * | 2000-11-02 | 2005-12-13 | Microsoft Corporation | Method and system for dynamically configuring a server computer |
| CN201654777U (en) * | 2009-04-22 | 2010-11-24 | 同方知网(北京)技术有限公司 | Dynamic data acquisition device |
| CN102096564A (en) * | 2009-12-14 | 2011-06-15 | 北大方正集团有限公司 | Job processing method for printer controller and printer controller |
| CN102722413A (en) * | 2012-05-16 | 2012-10-10 | 上海兆民云计算科技有限公司 | Distributed resource scheduling method for desktop cloud cluster |
| CN102857371A (en) * | 2012-08-21 | 2013-01-02 | 曙光信息产业(北京)有限公司 | Dynamic allocation management method for cluster system |
| CN104615661A (en) * | 2015-01-05 | 2015-05-13 | 华为技术有限公司 | Service recommendation method, device and system oriented to cloud platform applications |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107515883A (en) * | 2016-06-17 | 2017-12-26 | 阿里巴巴集团控股有限公司 | A kind of method and apparatus of operation task |
| CN110022220A (en) * | 2018-01-08 | 2019-07-16 | 腾讯科技（深圳）有限公司 | Routing Activiation method and system in business card recognition |
| CN110022220B (en) * | 2018-01-08 | 2022-02-25 | 腾讯科技（深圳）有限公司 | Route activation method and system in business card recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104980517B (en) | 2018-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902849B (en) | User behavior prediction method and device, and behavior prediction model training method and device | |
| US11706645B2 (en) | Systems and methods for communications node upgrade and selection | |
| US9934290B1 (en) | Systems and methods for dynamic sharding of hierarchical data | |
| CN107077385B (en) | For reducing system, method and the storage medium of calculated examples starting time | |
| US20190228337A1 (en) | Method and apparatus for acquiring information | |
| CN104254865A (en) | Empirical expert determination and question routing system and method | |
| CN103069413A (en) | Customizing content displayed for a user based on user preferences of another user | |
| KR20130062442A (en) | Method and system for recommendation using style of collaborative filtering | |
| WO2018161940A1 (en) | Method and device for pushing media file, data storage medium, and electronic apparatus | |
| CN104769582A (en) | Real-time data management for a power grid | |
| CN104978324B (en) | Data processing method and device | |
| US10963920B2 (en) | Web page viewership prediction | |
| CN109788020A (en) | One kind is attended a banquet distribution method and relevant device | |
| CN103164525A (en) | Method and device for WEB application release | |
| CN102436506A (en) | Mass data processing method and device for network server | |
| CN104980517A (en) | Snapshot-based cluster sensing system, method and device | |
| CN110969466A (en) | Method and device for determining advertisement delivery area | |
| US11259169B2 (en) | Highly scalable home subscriber server | |
| CN111092755B (en) | Edge service migration simulation method based on resource occupation | |
| JP6601889B1 (en) | Information processing apparatus, information processing method, and information processing program | |
| JP5900487B2 (en) | Mobile object management system, mobile object management server, mobile object management method, and mobile object management program | |
| CN113971428A (en) | Data processing method, system, device, program product and storage medium | |
| US20200089714A1 (en) | Method and server for indexing web page in index | |
| EP3119111B1 (en) | Method, device, server, terminal, and system for mobile service information display | |
| EP4216073A1 (en) | Data management method, data management apparatus, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |