CN109615081A - A kind of Model forecast system and method - Google Patents
A kind of Model forecast system and method Download PDFInfo
- Publication number
- CN109615081A CN109615081A CN201811121906.3A CN201811121906A CN109615081A CN 109615081 A CN109615081 A CN 109615081A CN 201811121906 A CN201811121906 A CN 201811121906A CN 109615081 A CN109615081 A CN 109615081A
- Authority
- CN
- China
- Prior art keywords
- model
- service
- module
- prediction
- model service
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013507 mapping Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 2
- 238000003860 storage Methods 0.000 description 23
- 230000006870 function Effects 0.000 description 19
- 238000012545 processing Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 14
- 241000207961 Sesamum Species 0.000 description 8
- 235000003434 Sesamum indicum Nutrition 0.000 description 8
- 230000006872 improvement Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000002955 isolation Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 239000000047 product Substances 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000000750 progressive effect Effects 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 229910021389 graphene Inorganic materials 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000011017 operating method Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45562—Creating, deleting, cloning virtual machine instances
Abstract
This specification provides a kind of Model forecast system and method, and different model service resource modules is isolated, so that the resources such as different CPU, memory of model prediction service will not be shared, reduces the interference between different model prediction services.It is independent to dispose different model service resource modules, it may be implemented individually to service the management such as progress capacity reducing, dilatation to some model prediction, reduce the wasting of resources for merging deployment.The management of the operating status of model prediction service then may be implemented in prediction model service module, improves stability, flexibility and the service ability of prediction service.
Description
Technical field
This specification belongs to field of computer technology more particularly to a kind of Model forecast system and method.
Background technique
Nowadays, machine learning is more and more fiery, and traditional rule model is more substituted in the model of machine learning output
Business uses on line, and the ability that model is given a mark in real time needs a unified framework to realize.Model prediction service can be with
It realizes online prediction in real time, and to guarantee to service the stabilization of itself, basic online service ability is provided.
In the prior art, on-line prediction service is deployed in above an application server generally by packing, multiple clothes
Business occupies same point of resource, may influence each other between service, for example a service committed memory is too high, will affect accordingly
Other services.Also, it is more troublesome to service dilatation, when different service dilatation, needs to expand together, leads to unnecessary resource wave
Take, and make online service ability inflexible, it has not been convenient to manage.
Summary of the invention
This specification is designed to provide a kind of Model forecast system and method, realizes the spirit of on-time model prediction service
Management living and O&M.
One side this specification embodiment provides a kind of Model forecast system, comprising:
Model service resource module, for providing model prediction service, between different model service resource modules mutually
Isolation；
Prediction model service module, for safeguarding the model service resource module and the model service resource module
Mapping relations between deployment facility, and when detecting that model service resource module changes, update the model clothes
Be engaged in the operating status of resource module, the model service resource module change include: model service resource module it is newly-increased,
The unlatching and stopping for the model prediction service that deletion, modification and model service resource module provide.
Further, in another embodiment of the system, different model service resource modules is arranged in different
In container.
Further, in another embodiment of the system, different model service resource modules is arranged in different
In virtual machine.
Further, in another embodiment of the system, the model service resource mould is constructed using deployment template
Block, the deployment template include: the corresponding running environment of different programming languages.
Further, in another embodiment of the system, the model service resource module support java, c++,
At least one of python programming language.
Further, in another embodiment of the system, the model service resource module supports Hyper text transfer
At least one of agreement, remote procedure call protocol.
Further, in another embodiment of the system, the prediction model service module is specifically used for: according to institute
The registration information for stating model service resource module obtains the disposes facility information of the model service resource module, and will be described
Mapping relations between model service resource module and the deployment facility of the model service resource module are stored.
Further, in another embodiment of the system, the system also includes clients, are used for the prediction
Model service module sends service-seeking request, the corresponding deployment facility of service needed for inquiring, according to query result, calling model
Service resource module.
Further, in another embodiment of the system, the system also includes services to manage module, for institute
State model service resource module and carry out life cycle management, the life cycle management include: dilatation, capacity reducing, in update extremely
Few one kind.
Further, in another embodiment of the system, the service control module is specifically used for: according to the mould
The load information of type service resource module carries out dilatation or capacity reducing to the model service resource module.
On the other hand, present description provides a kind of Model forecast systems, comprising:
Model service engine, it is mutually isolated between different model service engines for providing model prediction service；
Service discovery module, for constructing between the model service engine and the deployment facility of the model service engine
Mapping relations, and update the operating status of the model service engine, the model service engine carried out abnormality detection；
Control platform is serviced, for carrying out life cycle management, the life cycle management to the model service engine
It include: at least one of dilatation, capacity reducing, upper offline, service update
Client, for passing through model service engine described in the service discovery module polls and the model service engine
Deployment facility between mapping relations, and the model service engine is called to carry out model prediction service.
In another aspect, present description provides a kind of model prediction methods, comprising:
The solicited message of calling model prediction service is initiated by client；
The client obtains corresponding in the solicited message according to the solicited message, query service model list
The deployment facility of model service resource module is stored with model service resource module and the model in the service model list
Mapping relations between the deployment facility of service resource module, it is mutually isolated between different model service resource modules；
The client calls the model service resource module to carry out model prediction service according to query result, and returns
Return prediction result.
The Model forecast system and method that this specification provides, different model service resource modules is isolated, is made
Obtaining the resources such as different CPU, the memory of model prediction service will not share, and reduce dry between different model prediction services
It disturbs.It is independent to dispose different model service resource modules, it may be implemented individually to service progress capacity reducing, dilatation to some model prediction
Deng management, the wasting of resources for merging deployment is reduced.The accurate of model prediction service then may be implemented in prediction model service module
The unified management of calling and the operating status of model prediction service improves stability, flexibility and the service of prediction service
Ability.
Detailed description of the invention
In order to illustrate more clearly of this specification embodiment or technical solution in the prior art, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
The some embodiments recorded in this specification, for those of ordinary skill in the art, in not making the creative labor property
Under the premise of, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is Model forecast system module diagram in this specification one embodiment；
Fig. 2 is the structural schematic diagram of Model forecast system in another embodiment of this specification；
Fig. 3 is the functional module schematic diagram of Model forecast system in this specification one embodiment；
Fig. 4 is the flow diagram of model prediction method in this specification embodiment.
Specific embodiment
In order to make those skilled in the art more fully understand the technical solution in this specification, below in conjunction with this explanation
Attached drawing in book embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described
Embodiment be only this specification a part of the embodiment, instead of all the embodiments.The embodiment of base in this manual,
Every other embodiment obtained by those of ordinary skill in the art without making creative efforts, all should belong to
The range of this specification protection.
Model can indicate for describe can the pipe world mathematical model, the model in this specification embodiment can indicate
Machine term model, machine learning model can obtain by the learning training to data.Prediction service can indicate to provide
The service of line real-time calling can be used for predicting the data of specified time, such as: predict that the trading volume of specified time, prediction are specified
The credit rating etc. of user can carry out service prediction by way of model is given a mark, predict this theory of the concrete application form of service
Bright book embodiment is not especially limited.
A kind of Model forecast system is provided in this specification one embodiment, and multiple moulds can be provided in a system
Type service resource module, mutually isolated, i.e., the CPU of each model service resource module between different model service resource modules
The resources such as (Central Processing Unit, central processing unit), memory are without shared.Each model service resource
Module can be understood as a model, may be used to provide prediction service.By the way that different model service resource modules is carried out
Isolation may be implemented to service the shared resource quantity of single dimension distribution by model prediction, and each model prediction services independent portion
Administration, the management such as dilatation, the capacity reducing that can individually be serviced, avoids wasting of resources when integral deployment, reduces different
Interference between model prediction service.
Model forecast system in this specification embodiment may include distributed system, software (application), module, mould
Block, server, client etc. simultaneously combine the necessary device for implementing hardware.It is used below, term " unit " or " mould
The combination of the software and/or hardware of predetermined function may be implemented in block ".Although system described in this specification embodiment is preferably
It is realized with software, but the realization of the combination of hardware or software and hardware is also that may and be contemplated.
Specifically, Fig. 1 is Model forecast system module diagram in this specification one embodiment, as shown in Figure 1, this theory
The Model forecast system provided in bright book one embodiment may include: model service resource module 102, prediction model service mould
Block 104, in which:
Model service resource module 102, may be used to provide model prediction service, different model service resource modules it
Between it is mutually isolated；
Prediction model service module 104 can be used for safeguarding that the model service resource module and the model service provide
Mapping relations between the deployment facility of source module, and when detecting that model service resource module changes, update institute
The operating status of model service resource module is stated, it includes: model service resource mould that the model service resource module, which changes,
The unlatching and stopping for the model prediction service that the newly-increased and deletion of block and model service resource module provide.
Model service resource module 102 can be understood as prediction service model, may be used to provide model prediction service such as:
Model marking predicts that the sesame credit score of user, credit rating, fraud are transferred accounts.Model is pre- in this specification one embodiment
Model prediction service in examining system can be micro services, and micro services can indicate a kind of Software Architecture Design thought, micro services
It is mutually isolated between service in accordance with single responsibility principle, pass through unified API (Application Programming
Interface, application programming interface) gateway externally services.It can in a Model forecast system in this specification embodiment
To include multiple model service resource modules, each model service resource module can provide a model prediction service, each
It is mutually isolated between model service resource module, i.e., the resources such as CPU, memory between different model service resource modules not into
Row is shared, can be provided when constructing model service resource module with CPU, memory of pre-configured model service resource module etc.
Source.
In this specification one embodiment, can by the way of vessel isolation by different model service resource modules into
Row isolation, such as: different model service resource modules can be arranged in different container Docker, Docker can be indicated
The application container engine of one open source can will be applied and will be bundled in a transplantable container, each container Docker can
To carry a model prediction service.
In another embodiment of this specification, can using virtualization by different model service resource module carry out every
From such as: different model service resource modules can be arranged in different virtual machines, so that different model prediction services
It can be mutually indepedent.
Certainly, according to actual needs can also using other modes by different model service resource modules carry out every
From as long as may be implemented to interfere between different model prediction services.
In addition, multiple model prediction modules in a Model forecast system can be different, certainly, according to practical need
It wants, the identical model prediction module of specified quantity also can be set, it can there is the model prediction module of specified quantity to provide
The same model prediction service.
Prediction model service module, prediction model can also be set in Model forecast system in this specification one embodiment
Service module can be used for safeguard service (service of issuing into of prediction service model provides away) and ISP's (prediction clothes
Be engaged in model deployment physical equipment) mapping relations, may include building or more new demand servicing and service supplier mapping relations.
The deployment that can store each model service resource module and the model service resource module in prediction model service module is set
Mapping relations between standby, such as: it can store port numbers, the IP address deployment facility of model service resource module A.Predict mould
Type service module can also update the operation of model service resource module when detecting that model service resource module changes
State, operating status may include that whether model service resource module is registered, whether model service resource module is being predicted
Service, the service start times of model service resource module and end time, model service resource module are whether normal operation etc..
I.e. whether prediction model service module can newly-increased with detection model service resource module or be deleted, such as: detection model service money
Whether the service content that source module provides changes, if so, can then update the operating status of model service resource module.
Such as: if there is new service registration, a model service resource mould can be newly increased in service model list
Block, and add the disposes facility information of the model service resource module.Alternatively, the service content of some model service resource module
It changes, such as: being predicted by original sesame credit, become fraud prediction, then prediction model service module can modify this
The service content of model service resource module.Alternatively, some model service resource module starts to provide model prediction service, then may be used
The operating status of the model service resource module to be revised as servicing.
Compared to existing model prediction, the fortune of each model service resource module is updated by prediction model service module
Row state realizes each model service unified management, facilitates the management and calling of each service.
In this specification one embodiment, after model service resource module is provided with, mould can be serviced in prediction model
It is registered on block, according to the registration information of model service resource module, prediction model service module is available to arrive each mould
The disposes facility informations such as the deployment port numbers of type service resource module, IP address.After the completion of service registration, the service can be indicated
It can externally issue, it can externally provide service function, user, which can choose, calls the service.Prediction model service module can
To store the mapping relations between model service resource module and the deployment facility of model service resource module, such as: logical
Service model list is crossed to close the mapping between each model service resource module and the deployment facility of model service resource module
System is stored, and searching and managing is facilitated.Different model service resource modules is mutually indepedent, by service registration discovery mechanism,
Operating status, the disposes facility information etc. of model service resource module are managed, support the dynamic sensing of ISP,
When needing model prediction to service, the prediction service model that can be directly needed by the inquiry of prediction model service module is realized
The flexible calling of model.
The Model forecast system that this specification embodiment provides, different model service resource modules is isolated, is made
Obtaining the resources such as different CPU, the memory of model prediction service will not share, and reduce dry between different model prediction services
It disturbs.It is independent to dispose different model service resource modules, it may be implemented individually to service progress capacity reducing, dilatation to some model prediction
Deng management, the wasting of resources for merging deployment is reduced.The accurate of model prediction service then may be implemented in prediction model service module
The unified management of calling and the operating status of model prediction service improves stability, flexibility and the service of prediction service
Ability.
On the basis of the above embodiments, the model can be constructed using deployment template in this specification one embodiment
Service resource module, the deployment template include: the corresponding running environment of different programming languages.It predicts to service in deployment model
When, deployment template can be preset, may include the environment relied on when the operation of same speech like sound, software letter in deployment template
Breath, such as programming language java need jdk (Software Development Kit that jdk is java language) running environment, programming language c++
Need gcc (a kind of compiler external member) running environment etc..It can unify the difference between abstract isomery language by deployment template,
Realize the fast and flexible deployment of prediction service.
On the basis of the above embodiments, model service resource module described in this specification one embodiment can be supported
At least one of java, c++, python programming language.According to actual needs, model service resource module can also support it
His computer programming language selects different programming languages according to different application scenarios, Model forecast system can be improved
Applicability.
On the basis of the above embodiments, in this specification one embodiment, the model service resource module is supported super
Text transfer protocol, that is, HTTP (Hyper Text Transfer Protocol)), remote procedure call protocol RPC (Remote
At least one of Procedure Call).According to actual needs, model service resource module can also support other associations
View, model service resource module support various protocols, the applicability of Model forecast system can be improved.
Fig. 2 is the structural schematic diagram of Model forecast system in another embodiment of this specification, as shown in Fig. 2, this explanation
Model forecast system can also include client in book one embodiment, and service discovery module, that is, prediction model services mould in Fig. 2
Block services model service resource module A, B of A, service B, that is, different in figure, multiple models can also be arranged according to actual needs
Service resource module.Model call request can be initiated by client using the user of model prediction, client can be to clothes
The discovery module, that is, prediction model service module of being engaged in sends service-seeking request, the model service resource module serviced needed for inquiring with
Mapping relations between the deployment facility of the model service resource module, according to the deployment of the model service resource module inquired
Equipment is such as: IP address, port numbers, and calling model service resource module carries out on-time model prediction service.Client can also be tieed up
Service routing is protected, offline in dynamic sensing service, user does not need to be concerned about ISP i.e. model deployment facility, passes through visitor
Family end realizes the flexible calling of model prediction service.
On the basis of the above embodiments, in this specification one embodiment, Model forecast system can also include service
Module is managed, for carrying out life cycle management to the model service resource module, the life cycle management includes: to expand
At least one of appearance, capacity reducing, update.As shown in Fig. 2, the control platform in service control module, that is, Fig. 2, control platform can be with
Operation management model service provides model service life cycle management, such as: can expand each model service resource module
It is appearance, capacity reducing, upper offline (upper offline to indicate that starting to provide prediction service or prediction service completes, and also may indicate that and has registered
At, can provide prediction service or de-registration stop provide prediction service etc.), model modification (i.e. update model service resource
Service content or method of service of module etc.) etc..Eaily each model service can be provided by service control module
Source module is individually managed, such as: dilatation, capacity reducing, update are carried out for one of model service resource module.
It, can also be according to the load information of each model service resource module, to described in this specification one embodiment
Model service resource module carries out automatic dilatation or capacity reducing, such as: if the load of some model service resource module increases suddenly,
Automatic dilatation can be carried out to the model service resource module, realize the automatic management of each prediction service management.
Fig. 3 is the functional module schematic diagram of Model forecast system in this specification one embodiment, as shown in figure 3, this
Model forecast system in specification one embodiment may include: model service engine, service discovery module, service to manage and put down
Platform, client.
In the specific implementation process, single model prediction Service Instance can be deployed in an independent Docker to hold
In device, as used Docker cluster in Fig. 3, different services is arranged in different clusters, cluster A and cluster B are corresponding
2 different model services can be respectively arranged in model-container.Micro services engine, that is, model service engine in Fig. 3.Make
With container service between isolation, each service provides independent function, unrelated between each other.This specification embodiment
In deployment model service, deployment template is provided, deployment template may include the environment relied on when the operation of same speech like sound, software
Information, such as: java needs jdk running environment, and c++ needs gcc running environment etc..As shown in figure 3, cluster A and cluster B is corresponding
The programming language of model service, running environment, the agreement of support can not be identical, oss (Object Storage in Fig. 3
Service the object storage service of Ali's cloud) can be indicated, the model server in Fig. 3 can indicate model service,
Model-engine can indicate modeling engine, as shown in figure 3, caffe (Convolutional Architecture for
Fast Feature Embedding) it can indicate that a kind of deep learning frame, tensorflow can indicate a kind of artificial intelligence
Energy learning system, user can also according to actual needs, customized specific model service engine.It is taken out by the way that deployment template is unified
As the difference between isomery language, the function of supporting a variety of isomery language model marking may be implemented.Service discovery module can be with
Support the dynamic sensing ability of service such as: upper offline, whether abnormal, the similar zookeepr (distributed program of aware services
Coordination service) service discovery function, service discovery module support remote room service dynamic push, delay usually 50ms with
Under.
Management for the whole life cycle of prediction service, can use unified control platform, can be easier
Dynamic dilatation, capacity reducing are done to model prediction service, it might even be possible to realize according to the loading condition of each model service itself
Accomplish automatic dilatation and capacity reducing.As shown in figure 3, control platform, that is, micro services control platform can also carry out workflow management, mirror
As the management such as offline, server resets, service update in management, Template Manager, service.In addition, can be in Model forecast system
It is arranged sdk-client (Software Development Kit-client), sdk can indicate Software Development Kit,
Client can indicate client, and meta can indicate metadata in figure, and model meta can indicate model element.Client can
With the routing logic (supplier of model name and service) of transparence model prediction service, and the heart is established with service discovery machine
Chaser device can be predicted the machine quantity of service with dynamic sensing "current" model, facilitate System Fault Tolerance, provide load balance access plan
Slightly, and user is facilitated to access.
The Model forecast system that this specification embodiment provides, different model prediction services is isolated, Ke Yidan
Solely single model prediction service is updated, dilatation, capacity reducing, facilitates O&M and management.Also, a service model only provides
One prediction service, responsibility is single, and data-handling efficiency is high, and allocation of duties is clear, can also by service registration discovery mechanism,
The dynamic sensing for supporting ISP, that is, model service deployment facility, facilitates calling.
Various embodiments are described in a progressive manner for above system in this specification, identical between each embodiment
Similar part may refer to each other, and each embodiment focuses on the differences from other embodiments.Correlation
Place illustrates referring to the part of embodiment of the method.
Fig. 4 is the flow diagram of model prediction method in this specification embodiment, as shown in figure 4, this specification is implemented
Example provide model prediction method may include:
Step 402, the solicited message that calling model prediction service is initiated by client.
If user needs to service using model prediction, calling model can be initiated to Model forecast system by client
Predict the solicited message of service, such as: in the scape of sesame branch, the app (Application, application program) of sesame point being called to pass through
Client initiates the solicited message of calling model prediction service.
According to the solicited message, query service model list obtains the solicited message for step 404, the client
In corresponding model service resource module deployment facility, be stored in the service model list model service resource module with
Mapping relations between the deployment facility of the model service resource module, between different model service resource modules mutually every
From.
After client receives the solicited message of calling model prediction service of user's initiation, client can be to model
Prediction model service module in forecasting system sends service-seeking and requests, acquisition service model list, in service model list
May include the corresponding disposes facility information of different model services such as: IP address, port numbers.Client can be arranged in service
The deployment facility that the prediction in user request information services corresponding model service resource module is obtained in table, such as: if user asks
It asks and carries out the prediction of sesame credit, then can inquire sesame credit and predict that corresponding model service resource module is model A, service
The available disposes facility information to model A is such as in model list: IP address, port numbers.
Step 406, the client call the model service resource module to carry out model prediction clothes according to query result
Business, and return to prediction result.
According to query result, client can choose calling model service resource module, that is, call which service model, can
With according to regulative strategy such as: random call, Weight call, call distal end prediction micro services.Wherein, different model service
Mutually isolated between resource module, specific partition method can refer to the record of above-described embodiment, and details are not described herein again.If having
Multiple models can provide required model prediction service, can be according to regulative strategy such as: random call, Weight call,
One of model service resource module is called to carry out model prediction service, and can be result (such as: sesame branch scape, mould
Type prediction result is the sesame score value of user) return to user.
In addition, in this specification embodiment dynamic dilatation, capacity reducing, update etc. can also be carried out to each model service
Management.
The model prediction method that this specification embodiment provides, using Model forecast system, implementation model prediction is serviced
It flexibly calls, completes model on-line prediction service function, each model service is mutually isolated, facilitates O&M and management.
It should be noted that method described above can also include other embodiment party according to the description of system embodiment
Formula.Concrete implementation mode is referred to the description of related system embodiment, does not repeat one by one herein.
Model forecast system in this specification embodiment can be model prediction data processing equipment, comprising: at least one
A processor and memory for storage processor executable instruction, the processor are realized above-mentioned when executing described instruction
The model prediction function of the Model forecast system of embodiment,
The storage medium may include the physical unit for storing information, usually by after information digitalization again with benefit
The media of the modes such as electricity consumption, magnetic or optics are stored.It may include: that letter is stored in the way of electric energy that the storage medium, which has,
The device of breath such as, various memory, such as RAM, ROM；The device of information is stored in the way of magnetic energy such as, hard disk, floppy disk, magnetic
Band, core memory, magnetic bubble memory, USB flash disk；Using optical mode storage information device such as, CD or DVD.Certainly, there are also it
Readable storage medium storing program for executing of his mode, such as quantum memory, graphene memory etc..
It should be noted that processing equipment described above can also include other implement according to the description of system embodiment
Mode.Concrete implementation mode is referred to the description of related system embodiment, does not repeat one by one herein.
The Model forecast system that this specification provides can be the determination system of individual abnormal data, can also apply
In a variety of Data Analysis Services systems.The system can be individual server, also may include having used this specification
One or more the methods or the server cluster of one or more embodiment device, system (including distributed system),
Software (application), practical operation device, logic gates device, quantum computer etc. simultaneously combine the necessary terminal for implementing hardware
Device.The determination system of the abnormal data may include that at least one processor and storage computer executable instructions are deposited
The step of reservoir, the processor realizes method described in above-mentioned any one or multiple embodiments when executing described instruction.
It is above-mentioned that this specification specific embodiment is described.Other embodiments are in the scope of the appended claims
It is interior.In some cases, the movement recorded in detail in the claims or step can be come according to the sequence being different from embodiment
It executes and desired result still may be implemented.In addition, process depicted in the drawing not necessarily require show it is specific suitable
Sequence or consecutive order are just able to achieve desired result.In some embodiments, multitasking and parallel processing be also can
With or may be advantageous.
Method described in above-described embodiment that this specification provides or system can realize that business is patrolled by computer program
It collects and records on a storage medium, the storage medium can be read and be executed with computer, realize this specification embodiment institute
The effect of description scheme.
The above-mentioned Model forecast system or method that this specification embodiment provides can be executed by processor in a computer
Corresponding program instruction realizes, such as using the c++ language of windows operating system the end PC is realized, Linux system is realized,
Or other are for example realized using android, iOS system programming language in intelligent terminal, and based on quantum computer
Handle logic realization etc..
It should be noted that specification system described above, computer storage medium are retouched according to related method embodiment
Stating can also include other embodiments, and concrete implementation mode is referred to the description of corresponding method embodiment, herein not
It repeats one by one.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for hardware+
For program class embodiment, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to side
The part of method embodiment illustrates.
This specification embodiment is not limited to meet industry communication standard, standard computer data processing sum number
According to situation described in storage rule or this specification one or more embodiment.The right way of conduct is made in certain professional standards or use by oneself
In formula or the practice processes of embodiment description embodiment modified slightly also may be implemented above-described embodiment it is identical, it is equivalent or
The implementation result being anticipated that after close or deformation.Using these modifications or deformed data acquisition, storage, judgement, processing side
The embodiment of the acquisitions such as formula still may belong within the scope of the optional embodiment of this specification embodiment.
In the 1990s, the improvement of a technology can be distinguished clearly be on hardware improvement (for example,
Improvement to circuit structures such as diode, transistor, switches) or software on improvement (improvement for method flow).So
And with the development of technology, the improvement of current many method flows can be considered as directly improving for hardware circuit.
Designer nearly all obtains corresponding hardware circuit by the way that improved method flow to be programmed into hardware circuit.Cause
This, it cannot be said that the improvement of a method flow cannot be realized with hardware entities module.For example, programmable logic device
(Programmable Logic Device, PLD) (such as field programmable gate array (Field Programmable Gate
Array, FPGA)) it is exactly such a integrated circuit, logic function determines device programming by user.By designer
Voluntarily programming comes a digital display circuit " integrated " on a piece of PLD, designs and makes without asking chip maker
Dedicated IC chip.Moreover, nowadays, substitution manually makes IC chip, this programming is also used instead mostly " is patrolled
Volume compiler (logic compiler) " software realizes that software compiler used is similar when it writes with program development,
And the source code before compiling also write by handy specific programming language, this is referred to as hardware description language
(Hardware Description Language, HDL), and HDL is also not only a kind of, but there are many kind, such as ABEL
(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description
Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL
(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby
Hardware Description Language) etc., VHDL (Very-High-Speed is most generally used at present
Integrated Circuit Hardware Description Language) and Verilog.Those skilled in the art also answer
This understands, it is only necessary to method flow slightly programming in logic and is programmed into integrated circuit with above-mentioned several hardware description languages,
The hardware circuit for realizing the logical method process can be readily available.
Controller can be implemented in any suitable manner, for example, controller can take such as microprocessor or processing
The computer for the computer readable program code (such as software or firmware) that device and storage can be executed by (micro-) processor can
Read medium, logic gate, switch, specific integrated circuit (Application Specific Integrated Circuit,
ASIC), the form of programmable logic controller (PLC) and insertion microcontroller, the example of controller includes but is not limited to following microcontroller
Device: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320 are deposited
Memory controller is also implemented as a part of the control logic of memory.It is also known in the art that in addition to
Pure computer readable program code mode is realized other than controller, can be made completely by the way that method and step is carried out programming in logic
Controller is obtained to come in fact in the form of logic gate, switch, specific integrated circuit, programmable logic controller (PLC) and insertion microcontroller etc.
Existing identical function.Therefore this controller is considered a kind of hardware component, and to including for realizing various in it
The device of function can also be considered as the structure in hardware component.Or even, it can will be regarded for realizing the device of various functions
For either the software module of implementation method can be the structure in hardware component again.
System, device, module or the unit that above-described embodiment illustrates can specifically realize by computer chip or entity,
Or it is realized by the product with certain function.It is a kind of typically to realize that equipment is computer.Specifically, computer for example may be used
Think personal computer, laptop computer, vehicle-mounted human-computer interaction device, cellular phone, camera phone, smart phone, individual
Digital assistants, media player, navigation equipment, electronic mail equipment, game console, tablet computer, wearable device or
The combination of any equipment in these equipment of person.
Although this specification one or more embodiment provides the method operating procedure as described in embodiment or flow chart,
It but may include more or less operating procedure based on conventional or without creativeness means.The step of being enumerated in embodiment
Sequence is only one of numerous step execution sequence mode, does not represent and unique executes sequence.Device in practice or
When end product executes, can be executed according to embodiment or the execution of method shown in the drawings sequence or parallel (such as it is parallel
The environment of processor or multiple threads, even distributed data processing environment).The terms "include", "comprise" or its
Any other variant is intended to non-exclusive inclusion so that include the process, methods of a series of elements, product or
Equipment not only includes those elements, but also including other elements that are not explicitly listed, or further include for this process,
Method, product or the intrinsic element of equipment.In the absence of more restrictions, being not precluded is including the element
There is also other identical or equivalent elements in process, method, product or equipment.The first, the second equal words are used to indicate name
Claim, and does not indicate any particular order.
For convenience of description, it is divided into various modules when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each module can be realized in the same or multiple software and or hardware when specification one or more, it can also be with
The module for realizing same function is realized by the combination of multiple submodule or subelement etc..Installation practice described above is only
It is only illustrative, for example, in addition the division of the unit, only a kind of logical function partition can have in actual implementation
Division mode, such as multiple units or module can be combined or can be integrated into another system or some features can be with
Ignore, or does not execute.Another point, shown or discussed mutual coupling, direct-coupling or communication connection can be logical
Some interfaces are crossed, the indirect coupling or communication connection of device or unit can be electrical property, mechanical or other forms.
The present invention be referring to according to the method for the embodiment of the present invention, the process of device (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage, graphene stores or other
Magnetic storage device or any other non-transmission medium, can be used for storage can be accessed by a computing device information.According to herein
In define, computer-readable medium does not include temporary computer readable media (transitory media), such as the data of modulation
Signal and carrier wave.
It will be understood by those skilled in the art that this specification one or more embodiment can provide as method, system or calculating
Machine program product.Therefore, this specification one or more embodiment can be used complete hardware embodiment, complete software embodiment or
The form of embodiment combining software and hardware aspects.Moreover, this specification one or more embodiment can be used at one or
It is multiple wherein include computer usable program code computer-usable storage medium (including but not limited to magnetic disk storage,
CD-ROM, optical memory etc.) on the form of computer program product implemented.
This specification one or more embodiment can computer executable instructions it is general on
It hereinafter describes, such as program module.Generally, program module includes executing particular task or realization particular abstract data type
Routine, programs, objects, module, data structure etc..This this specification one can also be practiced in a distributed computing environment
Or multiple embodiments, in these distributed computing environments, by being held by the connected remote processing devices of communication network
Row task.In a distributed computing environment, program module can be located at the local and remote computer including storage equipment
In storage medium.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ",
The description of " specific example " or " some examples " etc. means specific features described in conjunction with this embodiment or example, structure, material
Or feature is contained at least one embodiment or example of this specification.In the present specification, to the signal of above-mentioned term
Property statement be necessarily directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described
It may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other, this
The technical staff in field can be by the spy of different embodiments or examples described in this specification and different embodiments or examples
Sign is combined.
The foregoing is merely the embodiments of this specification one or more embodiment, are not limited to book explanation
Book one or more embodiment.To those skilled in the art, this specification one or more embodiment can have various
Change and variation.All any modification, equivalent replacement, improvement and so within the spirit and principle of this specification, should all wrap
It is contained within scope of the claims.
Claims (12)
1. a kind of Model forecast system, comprising:
Model service resource module, it is mutually isolated between different model service resource modules for providing model prediction service；
Prediction model service module, for safeguarding the deployment of the model service resource module Yu the model service resource module
Mapping relations between equipment, and when detecting that model service resource module changes, update the model service money
The operating status of source module, it includes: increasing newly, deleting for model service resource module that the model service resource module, which changes,
Remove, modify and model service resource module provide model prediction service unlatching and stopping.
2. the system as claimed in claim 1, different model service resource modules is arranged in different containers.
3. the system as claimed in claim 1, different model service resource modules is arranged in different virtual machines.
4. the system as claimed in claim 1 constructs the model service resource module, the deployment template using deployment template
It include: the corresponding running environment of different programming languages.
5. the system as claimed in claim 1, the model service resource module supports at least one in java, c++, python
Kind programming language.
6. the system as claimed in claim 1, the model service resource module supports hypertext transfer protocol, remote process tune
With at least one of agreement.
7. the system as claimed in claim 1, the prediction model service module is specifically used for: according to the model service resource
The registration information of module, obtains the disposes facility information of the model service resource module, and by the model service resource mould
Mapping relations between block and the deployment facility of the model service resource module are stored.
8. the system as claimed in claim 1, the system also includes clients, for sending out to the prediction model service module
Business inquiry request is taken, the corresponding deployment facility of service needed for inquiring, according to query result, calling model service resource module.
9. the system as claimed in claim 1, the system also includes services to manage module, for the model service resource
Module carries out life cycle management, and the life cycle management includes: at least one of dilatation, capacity reducing, update.
10. system as claimed in claim 9, the service control module is specifically used for: according to the model service resource mould
The load information of block carries out dilatation or capacity reducing to the model service resource module.
11. a kind of Model forecast system, comprising:
Model service engine, it is mutually isolated between different model service engines for providing model prediction service；
Service discovery module, for constructing reflecting between the model service engine and the deployment facility of the model service engine
Relationship is penetrated, and updates the operating status of the model service engine, the model service engine is carried out abnormality detection；
Control platform is serviced, for carrying out life cycle management to the model service engine, the life cycle management includes:
At least one of dilatation, capacity reducing, upper offline, service update
Client, for the portion by model service engine described in the service discovery module polls and the model service engine
The mapping relations between equipment are affixed one's name to, and the model service engine is called to carry out model prediction service.
12. a kind of model prediction method, comprising:
The solicited message of calling model prediction service is initiated by client；
The client obtains corresponding model in the solicited message according to the solicited message, query service model list
The deployment facility of service resource module is stored with model service resource module and the model service in the service model list
Mapping relations between the deployment facility of resource module, it is mutually isolated between different model service resource modules；
The client calls the model service resource module to carry out model prediction service according to query result, and returns pre-
Survey result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811121906.3A CN109615081A (en) | 2018-09-26 | 2018-09-26 | A kind of Model forecast system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811121906.3A CN109615081A (en) | 2018-09-26 | 2018-09-26 | A kind of Model forecast system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109615081A true CN109615081A (en) | 2019-04-12 |
Family
ID=66002790
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811121906.3A Pending CN109615081A (en) | 2018-09-26 | 2018-09-26 | A kind of Model forecast system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109615081A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414187A (en) * | 2019-07-03 | 2019-11-05 | 北京百度网讯科技有限公司 | Model safety delivers the system and method for automation |
| CN110555550A (en) * | 2019-08-22 | 2019-12-10 | 阿里巴巴集团控股有限公司 | Online prediction service deployment method, device and equipment |
| CN111340232A (en) * | 2020-02-17 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | Online prediction service deployment method and device, electronic equipment and storage medium |
| CN111414265A (en) * | 2020-03-23 | 2020-07-14 | 海信电子科技（深圳）有限公司 | Service framework and method for calling system resources |
| CN111431748A (en) * | 2020-03-20 | 2020-07-17 | 支付宝(杭州)信息技术有限公司 | Method, system and device for automatically operating and maintaining cluster |
| CN111461332A (en) * | 2020-03-24 | 2020-07-28 | 北京五八信息技术有限公司 | Deep learning model online reasoning method and device, electronic equipment and storage medium |
| CN111753006A (en) * | 2020-06-10 | 2020-10-09 | 北京智源人工智能研究院 | Prediction system and method based on federal learning |
| CN111752556A (en) * | 2020-06-28 | 2020-10-09 | 自然资源部信息中心 | Data prediction system and method based on double-layer model structure |
| CN112416575A (en) * | 2020-11-02 | 2021-02-26 | 中关村科学城城市大脑股份有限公司 | Algorithm model scheduling system and method for urban brain AI calculation |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030236691A1 (en) * | 2002-06-21 | 2003-12-25 | Fabio Casatl | Business processes |
| CN104008428A (en) * | 2014-05-19 | 2014-08-27 | 上海交通大学 | Product service demand forecasting and resource optimization configuration method |
| US20140330745A1 (en) * | 2013-05-01 | 2014-11-06 | International Business Machines Corporation | Analytic solution integration |
| CN105005575A (en) * | 2015-03-05 | 2015-10-28 | 张良均 | Quick developing interface method for enterprise intelligent prediction |
| CN105786681A (en) * | 2016-02-18 | 2016-07-20 | 中国科学院计算技术研究所 | Server performance evaluating and server updating method for data center |
| CN106600058A (en) * | 2016-12-13 | 2017-04-26 | 北京理工大学 | Prediction method for combinations of cloud manufacturing service quality of service (QoS) |
| US20170124487A1 (en) * | 2015-03-20 | 2017-05-04 | Salesforce.Com, Inc. | Systems, methods, and apparatuses for implementing machine learning model training and deployment with a rollback mechanism |
| CN107247592A (en) * | 2017-06-09 | 2017-10-13 | 携程旅游网络技术（上海）有限公司 | Tackle the model management system and method under multi-service scene |
| CN108366082A (en) * | 2017-01-26 | 2018-08-03 | 华为技术有限公司 | Expansion method and flash chamber |
-
2018
- 2018-09-26 CN CN201811121906.3A patent/CN109615081A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030236691A1 (en) * | 2002-06-21 | 2003-12-25 | Fabio Casatl | Business processes |
| US20140330745A1 (en) * | 2013-05-01 | 2014-11-06 | International Business Machines Corporation | Analytic solution integration |
| CN104008428A (en) * | 2014-05-19 | 2014-08-27 | 上海交通大学 | Product service demand forecasting and resource optimization configuration method |
| CN105005575A (en) * | 2015-03-05 | 2015-10-28 | 张良均 | Quick developing interface method for enterprise intelligent prediction |
| US20170124487A1 (en) * | 2015-03-20 | 2017-05-04 | Salesforce.Com, Inc. | Systems, methods, and apparatuses for implementing machine learning model training and deployment with a rollback mechanism |
| CN105786681A (en) * | 2016-02-18 | 2016-07-20 | 中国科学院计算技术研究所 | Server performance evaluating and server updating method for data center |
| CN106600058A (en) * | 2016-12-13 | 2017-04-26 | 北京理工大学 | Prediction method for combinations of cloud manufacturing service quality of service (QoS) |
| CN108366082A (en) * | 2017-01-26 | 2018-08-03 | 华为技术有限公司 | Expansion method and flash chamber |
| CN107247592A (en) * | 2017-06-09 | 2017-10-13 | 携程旅游网络技术（上海）有限公司 | Tackle the model management system and method under multi-service scene |
Non-Patent Citations (1)
| Title |
|---|
| 朱率率 等: "动态自适应网络存储***的设计与实现", 《计算机应用研究》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414187B (en) * | 2019-07-03 | 2021-09-17 | 北京百度网讯科技有限公司 | System and method for model safety delivery automation |
| CN110414187A (en) * | 2019-07-03 | 2019-11-05 | 北京百度网讯科技有限公司 | Model safety delivers the system and method for automation |
| CN110555550A (en) * | 2019-08-22 | 2019-12-10 | 阿里巴巴集团控股有限公司 | Online prediction service deployment method, device and equipment |
| CN110555550B (en) * | 2019-08-22 | 2023-06-23 | 创新先进技术有限公司 | Online prediction service deployment method, device and equipment |
| CN111340232A (en) * | 2020-02-17 | 2020-06-26 | 支付宝(杭州)信息技术有限公司 | Online prediction service deployment method and device, electronic equipment and storage medium |
| CN111431748A (en) * | 2020-03-20 | 2020-07-17 | 支付宝(杭州)信息技术有限公司 | Method, system and device for automatically operating and maintaining cluster |
| CN111431748B (en) * | 2020-03-20 | 2022-09-30 | 支付宝(杭州)信息技术有限公司 | Method, system and device for automatically operating and maintaining cluster |
| CN111414265A (en) * | 2020-03-23 | 2020-07-14 | 海信电子科技（深圳）有限公司 | Service framework and method for calling system resources |
| CN111414265B (en) * | 2020-03-23 | 2024-03-19 | 海信电子科技（深圳）有限公司 | Service framework and method for calling system resources |
| CN111461332A (en) * | 2020-03-24 | 2020-07-28 | 北京五八信息技术有限公司 | Deep learning model online reasoning method and device, electronic equipment and storage medium |
| CN111753006A (en) * | 2020-06-10 | 2020-10-09 | 北京智源人工智能研究院 | Prediction system and method based on federal learning |
| CN111753006B (en) * | 2020-06-10 | 2021-03-16 | 北京智源人工智能研究院 | Prediction system and method based on federal learning |
| CN111752556B (en) * | 2020-06-28 | 2021-04-30 | 自然资源部信息中心 | Data prediction system and method based on double-layer model structure |
| CN111752556A (en) * | 2020-06-28 | 2020-10-09 | 自然资源部信息中心 | Data prediction system and method based on double-layer model structure |
| CN112416575A (en) * | 2020-11-02 | 2021-02-26 | 中关村科学城城市大脑股份有限公司 | Algorithm model scheduling system and method for urban brain AI calculation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109615081A (en) | A kind of Model forecast system and method | |
| CN109597661A (en) | A kind of service function configuration method and device | |
| CN105453040B (en) | The method and system of data flow is handled in a distributed computing environment | |
| CN104657194B (en) | Method and system for calculating the influence acted in network | |
| US20120284730A1 (en) | System to provide computing services | |
| CN109615495A (en) | A kind of account checking method of data, device, equipment and system | |
| CN110096528A (en) | The method, apparatus and system of formation sequence in a kind of distributed system | |
| US20190050245A1 (en) | Scheduling framework for tightly coupled jobs | |
| US11188348B2 (en) | Hybrid computing device selection analysis | |
| US11243936B2 (en) | Selectively requesting updated data values | |
| CN110244957A (en) | A kind of resource data updating method, apparatus and system | |
| CN108683692A (en) | A kind of service request processing method and device | |
| US20210109932A1 (en) | Selecting an optimal combination of systems for query processing | |
| CN109241026A (en) | The method, apparatus and system of data management | |
| CN109936642A (en) | The method, apparatus and system of machine ID are generated in a kind of distributed system | |
| CN108345644A (en) | A kind of method and device of data processing | |
| CN110502415A (en) | One kind burying a setting method, device and equipment | |
| CN109947643A (en) | A kind of configuration method, device and the equipment of the experimental program based on A/B test | |
| KR20190088996A (en) | Facility to extend exclusive hold of cache lines in dedicated cache | |
| CN109584431A (en) | A kind of data processing method of priority queue, apparatus and system | |
| US10803246B2 (en) | Decomposing composite product reviews | |
| US11030015B2 (en) | Hardware and software resource optimization | |
| CN110032750A (en) | A kind of model construction, data life period prediction technique, device and equipment | |
| US11157406B2 (en) | Methods for providing data values using asynchronous operations and querying a plurality of servers | |
| CN106462446B (en) | Host is selected using hardware multithreading parameter for virtual machine |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201010 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20201010 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: Greater Cayman, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190412 |