CN104714808B - Data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis - Google Patents
Data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis Download PDFInfo
- Publication number
- CN104714808B CN104714808B CN201510156076.8A CN201510156076A CN104714808B CN 104714808 B CN104714808 B CN 104714808B CN 201510156076 A CN201510156076 A CN 201510156076A CN 104714808 B CN104714808 B CN 104714808B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- app
- context
- task
- 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.)
- Expired - Fee Related
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides data in a kind of calamity emergency comprehensive analysis and computing resource virtualized management method, system is supported for all kinds of calamity emergency Response Decisions.The present invention will study and judge task call module abstracts for 3 agent, be App_Context, Model_Invoker and Data_Channel respectively；Abstract package is carried out to studying and judging model, data source, computing resource and data resource is formed；It is packaged to studying and judging task call interface, computing resource calling interface and data resource transfer interface；Interaction protocol is designed, task is studied and judged in 3 agent collaboration completions.The present invention is by studying and judging task, study and judge model and data source loose coupling in comprehensive analysis system, beneficial to the renewal and extension of the data resource species in system application process, the correctness for calculating studying and judging model implementation procedure is supported to verify, beneficial to error analysis and location of mistake, it is with good expansibility.
Description
Technical field
The present invention relates to the comprehensive analysis system in the class DSS (task system) responded towards calamity emergency
System, and in particular to data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis.
Background technology
In recent years, domestic and international frequent natural calamity shows, strengthens the construction that calamity emergency responds comprehensive analysis system,
Rescue and responding ability to improving burst natural calamity event, people's life's property loss is reduced to greatest extent has ten
Divide important realistic meaning.Such system it is main by module on duty, study and judge task management module, study and judge model computation module and grind
Sentence the composition such as result display module.Its core is to study and judge task management module and study and judge model computation module.Natural calamity
When event occurs, module on duty obtains alarm, is then sent to alarm and studies and judges task management module, studies and judges task management module
Called at different time points and study and judge model computation module, call correlation to study and judge model by studying and judging model computation module, to burst
(calculating) is studied and judged in Disaster Event expansion, obtains studying and judging report, and significant reference is provided for next step expansion disposing task.
By taking earthquake emergency disposal system as an example, it is comprehensive that earthquake emergency disposal system mainly includes earthquake situations monitoring subsystem, macroseism
Conjunction studies and judges subsystem and studies and judges result list circulation subsystem.Wherein, earthquake situations monitoring subsystem is responsible for repeating query monitoring information source
Earthquake situations are alerted, once receiving warning information, then start macroseism comprehensive analysis subsystem immediately and earthquake situations and the condition of a disaster are integrated
Study and judge, then by obtain study and judge result be sent to list circulation subsystem circulated, be final decision service.
The comprehensive analysis system of emergency response task system comprehensive analysis is deployed to accident needed for resource it is main by
Two classes are constituted：Computing resource and data resource.Wherein, computing resource is primarily referred to as studying and judging model, and one is studied and judged model correspondence one
Individual calculating process, the calculating process needs parameter and input, and the result of calculating process is to study and judge the output of model, that is, studies and judges knot
Really.Model is studied and judged typically to be proposed by industry specialists.In comprehensive analysis system, used it is substantial amounts of study and judge model, can up to count
Hundred, have and many study and judge model at present also in studying and improving the stage；Many studies and judges model and can be applied to different comprehensive analysis
System；Study and judge calculating implementation procedure and corresponding model parameter, the data source close coupling of input data of model so that study and judge mould
The verification of correctness that type calculates implementation procedure is extremely difficult.
Data resource supports that studying and judging model is calculated, and the data resource in comprehensive analysis system is general by a variety of data class
Type is constituted, including：Relational data, spatial data, text data and internet web page data etc., wherein relational data is main
For storing event attribute and studying and judging model parameter etc., spatial data is used for storing geo-spatial data and a part of model parameter
Data.
By taking earthquake emergency disposal system as an example, what earthquake emergency disposal system was related to, which study and judge model resource, is divided into two classes：Shake
Feelings model and the condition of a disaster model.Earthquake situations model is primarily referred to as to the earthquake model that some attributes are calculated in itself, including：It is strong
The models such as degree analysis, fracture belt analysis, the analysis of earthquake region topography and landform character, secondary disaster analysis；The condition of a disaster model refers to by ground
The model that population, building collapsing, economic loss that shake is caused etc. is calculated, including：The analysis of population injures and deaths, building collapsing
Analysis, density of population analysis, Analysis of Economic Loss etc..
Current such comprehensive analysis system in design it is general by application system, study and judge model and data resource three is tight
Coupling, is encapsulated as calculating component, task is studied and judged in execution, obtains studying and judging result, the shortcoming that so design is realized mainly has：
(1) if application system, studying and judging model and data resource three's close coupling, it is likely that hinder grinding for application system
Hair progress.
On the one hand, because calamity emergency rescues the complexity in field, correlation is studied and judged model also in the stage is studied and improved, needed
To be updated and extend during systematic difference, to meet newest business demand；On the other hand, ground even if determining
Sentence model, not exclusively, i.e., the data resource required for studying and judging model calculating process is also possible to：Current data resource may be discontented with
Foot studies and judges the parameter and input required for model calculating process, and studying and judging model can not still be applied in system, and thus having can
The research and development progress of application system can be hindered.
(2) if application system, studying and judging model and data source close coupling, it is unfavorable in system application process to studying and judging mould
Type is modified and extended.
In system study and judge model need modified according to the achievement in research in more recent application field, might have improve and
The model of studying and judging updated adds system.If studying and judging model close coupling in systems, be difficult to study and judge model be updated and
Extension；Secondly, study and judge model and can be applied to different application systems, if one is studied and judged model close coupling in an application system
In, it will do repeated work when it is applied in different systems.Accordingly, it would be desirable to which one unified is studied and judged model editing ring
Renewal and extension of the model in system application process are studied and judged in border, support, to support different application systems.
(3) in the system application stage, if application system, studying and judging model and data source close coupling, being unfavorable for should in system
The renewal and extension of data resource species during.
Data resource used in comprehensive analysis system generally comprises relational data, spatial data, text data, interconnection
Net web data etc., these data resources need to constantly update and extend in system application process, and make in application system
During, partial data, such as geo-spatial data, there is also the problems such as data aging.If application system, study and judge model and
Data resource close coupling, when updating and extending a class data resource, it has to change application system to adapt to data money
The addition in source.Therefore a kind of unified data resource management of design and integration mechanism and user interface are needed, data resource is supported
Renewal and extension.
(4) in the system application stage, if application system, studying and judging model and data resource close coupling, it is difficult to studying and judging mould
Type calculates implementation procedure correctness and verified, once there is mistake, is unfavorable for error analysis and positioning.
If by application system, studying and judging model and data resource close coupling, it is encapsulated as calculating component, externally shows as black
Box, in system operation, on the one hand can not be verified to the correctness for the calculating implementation procedure for studying and judging model, on the other hand,
Invisible due to calculating component inside state if mistake occurs in calculating process, mistake, which is analyzed and positioned, will become tired
It is difficult.Accordingly, it would be desirable to by application system, study and judge model and data resource loose coupling, they are subjected to abstract package, and design one kind
The calculating implementation procedure of model and the convergence transparent procedures of data resource are studied and judged in interaction protocol, support, to support checking to study and judge mould
Type calculates the correctness of implementation procedure.
The content of the invention
In order to solve the above problems, validity, scalability, the reusability of comprehensive analysis system are improved, the present invention is carried
Gone out data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis, task will be studied and judged, study and judge model and
Data source three's loose coupling, only includes necessary interactive relation：Model will be studied and judged to be packaged according to unified encapsulation specification, shape
Into the logical calculated resource of canonical representation；Unified encapsulation is carried out to data resource, unified data object is expressed as and is defined on
One group of unified operation on these data objects, i.e. logical data resource.Support to enter to studying and judging model in the system application stage
Row is updated and extended, and data resource class is updated and extended；Simultaneously as studying and judging task independently of studying and judging model sum
According to resource, thus make that the calculating process for entirely studying and judging task can be monitored, result of calculation can be assessed, abnormal behavior is controllable, once
There is mistake in the execution of calculating process, and mistake, which is positioned and analyzed, will also become convenient.
Data and computing resource virtualized management method, comprise the following steps in the calamity emergency comprehensive analysis of the present invention：
Step 1：Task is studied and judged into module abstracts for three part 3-agent, respectively context (App_
Context), model caller (Model_Invoker) and data channel (Data_Channel)；
The set of context record data source, business datum sequence and supplemental characteristic sequence and initial parameter
Context is sky under sequence, original state；
Model caller includes model loader, mapper and actuator；Model loader is used for the loading of model, to mould
Type carries out protocol testing and analytic modell analytical model；Mapper is used to set up mode input Variables Sequence and argument sequence arrives data source
Mapping, mode input variable and model parameter are instantiated；Actuator is used for the execution of model calculating process；
Data channel is intermediary of the application system to data source request data.Analyzed including request receiver, request of data
Device, sequencing unit and data hinge；Request receiver receives the request of data that request of data end is sent；Request of data analyzer
It is responsible for protocol testing and the parsing of request of data, request of data is then mapped as data source command；Sequencing unit is used to connect
Data source is connect, sends and orders to data source；The data that data source is returned are returned to request of data end by data hinge.
Step 2：Abstract package is carried out to studying and judging model, data source according to Resource Encapsulation specification, logical calculated resource is formed
With logical data resource；
Studying and judging the data structure of model includes：Study and judge model name, input variable sequence, argument sequence, output variable sequence
Row and model calculating process；
The data structure of data source includes：DSN, data source types, data source address and data acquisition system；Data
Set is made up of data, and the definition of each data is defined on data object including data name, one group of data object and one group
On operation；The definition of each data object includes data object title and the value of data object.
Step 3：According to Resource Encapsulation specification, calling interface is encapsulated, including studies and judges task call interface, computing resource and is called
Interface and data resource transfer interface；
Wherein, study and judge task call interface include study and judge task names, initial parameter list, call study and judge model name
And the data description of required by task data；Computing resource calling interface is to study and judge model name；Data resource calling interface bag
Data name, DSN, data manipulation needed for including, the predicate for determining data and returned data description.
Step 4：Interaction protocol is designed, task is studied and judged in 3-agent collaboration completions.
Data and computing resource virtualized management method in the calamity emergency comprehensive analysis of the present invention, its advantage and actively effect
Fruit is：
(1) use technical scheme, study and judge Model R ＆ D and update no longer be it is such towards calamity emergency response
Comprehensive analysis system exploitation bottleneck.Possess unified encapsulation specification due to studying and judging model, can not in the system research and development stage
It must wait until to study and judge after model is determined completely and develop again.System research and development personnel can first carry out other in addition to model is studied and judged
The research and development and test of operation flow, after studying and judging model and determining, will study and judge model dynamic integrity according to unified encapsulation specification and arrive
In.
(2) due to technical scheme provide one it is unified study and judge model editing environment, supporting should in system
It is updated and extends to studying and judging model during, thus can be according to the achievement in research in more recent application field in system
Model of studying and judging modify, by more it is advanced and improve study and judge model add system；Applied also for not in addition, studying and judging model
Same application system.
(3) present invention devises a kind of unified data resource management and integration mechanism, beneficial in system application process
Data resource category is updated and extended, when updating or extend a class data resource, it will not be necessary to which changing application system can
To adapt to the addition of the data resource.
(4) present invention devises one group of interaction protocol, supports to study and judge the calculating implementation procedure of model and the remittance of data resource
Collecting process is transparent, will study and judge task, studies and judges model, data resource three's loose coupling, supports to calculate implementation procedure to studying and judging model
Correctness verified, beneficial to error analysis and location of mistake.
Brief description of the drawings
Fig. 1 studies and judges the relation schematic diagram between the 3-agent of module abstracts for task in the present invention；
Fig. 2 studies and judges the schematic flow sheet that model whether there is for present invention inspection；
Fig. 3 checks the schematic flow sheet that data source whether there is for the present invention；
Fig. 4 is the schematic flow sheet that App_Context of the present invention is initialized；
Fig. 5 is to study and judge the schematic flow sheet of model initialization in the inventive method；
Fig. 6 studies and judges the schematic diagram of model calculating process to be performed in the inventive method；
Fig. 7 cooperates with the overall flow schematic diagram for completing to study and judge task for the present invention using 3-agent.
Embodiment
Below in conjunction with drawings and examples, the present invention is described in further detail.
Data and computing resource virtualized management method, are specifically included in the calamity emergency comprehensive analysis that the present invention is provided：
First, proposition is a kind of to study and judge tasks carrying framework based on 3-agent, and the task in comprehensive analysis system is studied and judged into module abstracts
For three parts, referred to as 3-agent is context (App_Context), model caller (Model_ respectively
Invoker) and data channel (Data_Channel), the interaction call relation between 3-agent is as shown in Figure 1；Second, should
Study and judge model and data source in anxious disposal task application encapsulate specification progress abstract package according to one group of virtual resources, respectively
Obtain logical calculated resource and logical data resource；3rd, interface encapsulation is carried out according to above-mentioned Resource Encapsulation specification, including study and judge
Task call interface, computing resource calling interface and data resource transfer interface；Finally, it is based between a kind of 3-agent of design
Interface specification good according to the rules is interacted between the interaction protocol of above-mentioned Resource Encapsulation specification, 3-agent, and collaboration completion is ground
Sentence task.
Below in conjunction with the accompanying drawings to the present invention each realize that step is described in detail.
Step 1:Task is studied and judged into module abstracts for 3-agent.
(1) context App_Context.
The set of context (App_Context) record data source, business datum sequence and supplemental characteristic sequence,
And the part of initial parameter sequence three.Under original state, context is sky.App_Context is defined as one three
Tuple, it is as follows：
App_Context=({ DB1,…,DBm},<(a1,cond1,db1),…,(an,condn,dbn)>,<init1,…,
initk>)
In above-mentioned expression：M, n, k are positive integer；
●{DB1,…,DBmData source set is represented, some of which is supplemental characteristic, and some are business datums；
●<(a1,cond1,db1),…,(an,condn,dbn)>Represent specific business datum sequence or supplemental characteristic sequence
Row, ai=ciRepresent supplemental characteristic, ciIt is probably a vector；ai=biRepresent business datum；condiIt is boolean's expression
Formula, for determining data query conditions, dbiRepresent aiThe database at place；
●<init1,…,initk>Initial parameter sequence is represented, for example<Longitude, latitude, earthquake magnitude, depth>.
Context is determined that then task and model end, the interaction of data source instantiate App_ by task
Context。
(2) model caller Model_Invoker.
The responsible inquiry of model caller (Model_Invoker) studies and judges model and whether there is, carries out agreement to studying and judging model
Check and perform and study and judge calculating process of model etc., including model loader, mapper and the part of actuator three, model caller
It is defined as a triple：
Model_Invoker=(Loader, Mapper, Executor)
In above-mentioned expression：
●Loader：Model loader, is responsible for the loading of model, and protocol testing is carried out to model, and mould is studied and judged in then parsing
Type；
●Mapper：Mapper, is responsible for setting up mode input Variables Sequence and argument sequence to the mapping of data source, by mould
Type input variable and model parameter instantiation；
●Executor：Actuator, is responsible for the execution of model calculating process.
(3) data channel Data_Channel.
Data channel (Data_Channel) is intermediary of the application system to data source Data_Source request datas, bag
Request receiver, request of data analyzer, sequencing unit and data hinge are included, Data_Channel is expressed as：
Data_Channel=(Req_Receiver, Req_Analyser, Cmd_Sender, Data_Hub)
Wherein：
●Req_Receiver：Request receiver, the request of data for receiving the transmission of request of data end；
●Req_Analyser：Request of data analyzer, is responsible for protocol testing, the parsing of request of data, then by data
Request is mapped as specific data source command；
●Cmd_Sender：Sequencing unit, for connecting data source, then sends to data source and orders；
●Data_Hub：Data hinge is represented, the data for being responsible for returning to data source return to request of data end.
Step 2：Abstract package is carried out to studying and judging model, data source according to Resource Encapsulation specification, logical calculated resource is formed
With logical data resource.
(1) model M odel encapsulation is studied and judged.
Studying and judging the data structure that model is abstracted into includes model name, input variable sequence, argument sequence, calculating process
With the part of output variable sequence 5.Model of studying and judging in comprehensive analysis system is packaged according to this 5 part, that is, obtains system
In logical calculated resource.
Study and judge model (Model) and be defined as a five-tuple：
Model=(name,<in1,…,inp>,<p1,…,pq>,proc,<out1,…,outr>)
In above-mentioned expression：
●name：The title of model is studied and judged in expression；
●<in1,…,inp>：Represent input variable sequence, ini∈<in1,…,inp>, i=1,2 ..., p, referred to as model
An input variable；P is positive integer；
●<p1,…,pq>：Represent argument sequence, pi∈<p1,…,pq>, i=1,2 ..., q, referred to as model a ginseng
Number；Q is positive integer；
●proc：The calculating process of model is represented, can be calculation formula or clear and definite mapping relations；
●<out1,…,outr>：Represent the output variable sequence of model, outi∈<out1,…,outr>, i=1,2 ...,
R, is referred to as an output variable for studying and judging model；R is positive integer.
(2) data source Data_Source encapsulation.
Data source (Data_Source) is combined into by volume of data collection, and data acquisition system is made up of one group of data again, data
Including data name and one group of data object and one group of operation being defined on data object, data object is by data object name
Claim, the value of data object is constituted.
The data structure that data resource is abstracted into includes：DSN, data source types, data source address and data
Set.Data source Data_Source is defined as a four-tuple：
Data_Source=(DS_Name, DS_Type, DS_Addr, Data_Set)
In above-mentioned expression：
●DS_Name:Represent DSN, one data source of unique mark；
●DS_Type：Data source types are represented, two class data sources --- relational data DB and spatial data are supported at present
SDB, i.e.,：DS-Type={ DB, SDB }；
●DS_Addr：Represent data source address；
●Data_Set：Represent the data acquisition system in data source.
Data acquisition system (Data_Set) is made up of one group of data, as follows：
Data_Set={ Data1,Data2,…Datan}
Each data (Data) are defined as a triple：
Data=(Data_Name, Data_Objects, Data_Operation)
In above-mentioned expression：
●Data_Name：Flag data title, one group of data of unique mark；
●Data_Objects：Represent the set being made up of one group of data object；
●Data_Operation：Represent one group of operation on data object Data_Objects：
Data_Operation={ GET, PUT, POST, DELETE }, wherein：
GET：Obtain (inquiry) data object operation；
POST：Data object operation is created, if there is the data object, is reported an error；
PUT：Object is created or updated the data, if data object is not present, data object is performed and creates operation；Such as
Fruit data object is present, then is updated work to it；
DELETE：Delete data object operation.
Data object includes data object title and the value of data object, and data object (Data_Object) is defined as one
Individual two tuple：
Data_Object=(Object_Name,<Object_Value1,…,Object_Valuen>)
In above-mentioned expression：
●Object_Name：Represent the title of data object；
●<Object_Value1,…,Object_Valuen>：Represent the sequence of the value of data object, the one of data object
Individual value Object_Valuei∈<Object_Value1,…,Object_Valuen>, i=1,2 ..., n, and Object_
Valuei∷=Data_Object | Object, an i.e. data object value either data object, or one not
Subdivisible object.
Step 3:According to Resource Encapsulation specification, calling interface is encapsulated, including studies and judges task call interface, computing resource and is called
Interface, data resource calling interface.
(1) task call interface is studied and judged.
Study and judge task task and be defined as a triple：
Task=(Name, Process, Form)
In above-mentioned expression：
●Name：Task names are studied and judged in expression, and unique mark one studies and judges task；
●Form：List is represented, by initial parameter sequence<init1,…,initk>Constitute, after tasks carrying terminates, will
To the model calculation insert list；
●Process：Flow is represented, what flow definition task to be called studies and judges model name and required by task data
Data description, you can be expressed as two tuples：
Process=(Model_Name, Data_Desc)
Then Task is represented by：
Task=(Name,<init1,…,initk>,Model_Name,Data_Desc)
Wherein, Data_Desc represents the data description of required by task data.
Study and judge task call interface and be described as above-mentioned Task data.
(2) computing resource calling interface.
Computing resource calling interface is model name Model_Name.
(3) data resource calling interface.
Data resource calling interface is that data describe Data_Desc, illustrates that agent ends propose request of data to data source
Interface specification.Request of data (Data_Req) is the request of data that Data_Channel is sent to Data_Source, Data_
Data description is first converted to request of data by Channel ends according to rule, then by the data manipulation in request of data according to rule
Then be converted to the data command of related data sources.Predicate (Predicate) is used to inquire about data in data source.Returned data is retouched
Stating (Ret_Data_Desc) then illustrates data source to interface specification during agent returned datas.
Data description (Data_Desc) is defined as a five-tuple：
Data_Desc=(Data_Name, DS_Name, Data_Operation, Predicates, Ret_Data_
Desc)
In above-mentioned expression：
●Data_Name：Data name needed for representing, value is as follows：
●DS_Name：Represent the DSN where the data；
●Data_Operation：Represent data manipulation, Data_Operation={ GET, PUT, POST, DELETE }；
●Predicates：The predicate for determining data is represented, inquiry (applicable) condition of data needed for representing；
●Ret_Data_Desc：Represent the description of returned data.
Request of data (Data_Req) be request of data end to data source request data when sent by request of data end,
A four-tuple can be expressed as：
Data_Req=(DS_Name, Data_Operation, Predicates, Ret_Data_Desc)
In above-mentioned expression：
●DS_Name：Represent the DSN of request；
●Data_Operation：The operation to data object is represented,
Data_Operation={ GET, PUT, POST, DELETE }；
●Predicates：Represent to be used for the predicate for determining target data objects in data source；
●Ret_Data_Desc：Represent the description of returned data.
Predicate (Predicate) is used for the condition that data object is determined in data source, is defined as：
● Predicate is an atom predicate expressions；
● set Predicate1,Predicate2Predicate expressions, then Predicate1∩Predicate2,
Predicate1∪Predicate2, Not Predicate are predicate expressions；
Atom predicate (Atom_Predicate) is a Boolean expression, is defined as：
Atom_Predicate=boolean (Object1,Object2)
Wherein：Boolean is specific a boolean operation, Object1,Object2It is boolean two operations pair
As；
For relevant database：
● boolean ∈=,>,<, >=,≤, /=
●Object1For the simple function defined in data name Data-Name or DB；
●Object2∈<init1,…,initk>, it is initial parameter.
For GIS data：
● boolean ∈=,>,<, >=,≤, /=∪ Equals, Disjoint, Intersects, Touches,
Crosses,Overlaps,Within,Contains}；
In above-mentioned expression：
=,>,<, >=,≤, /=represents be equal to, be more than, being less than, being more than or equal to, being less than or equal to, being not equal to fortune respectively
Calculate；
Equals is used to judge in searched figure layer with the presence or absence of the object being completely superposed with object search；
Disjoint be used to judge whether there is in searched figure layer with object search from object；
Intersects is used to judge to whether there is in searched figure layer to occur simultaneously not as all right of sky with object search
As；
Touches is used for the object for judging to touch with the presence or absence of its border and object search border in searched figure layer；
Crosses user judge in searched figure layer with the presence or absence of intersect with object search (line) all objects (line,
Face)；
Overlaps is used to judge to whether there is and the partly overlapping object of object search in searched figure layer；
Within is used to judge whether include object search in searched figure layer；
Contains is used to judge to be searched the object that object is included in searched figure layer；
●Object1For data name, or it is figure layer title, or is geometric object, or is the simple letter defined in SDB
Number；
●Object2∈<init1,…,initk>, it is initial parameter, or be geometric object；
Returned data description (Ret_Data_Desc) represents lattice of the data source to one end returned data of requested data
Formula is described, and is defined as：
Ret_Data_Desc=(Data_Name, Data_Objects)
In above-mentioned expression：
●Data_Name:Represent the title of returned data；
●Data_Objects：Represent the set of data objects returned；
Data_Objects={ Data_Object1,…,Data_Objectn, wherein Data_ObjectiRepresent data
The description of object,Data_Objecti∈ Data_Objects, initial setting up Data_Objecti.Object_Value=
null；
Step 4:Interaction protocol is designed, task is studied and judged in 3-agent collaboration completions.
As shown in figure 1, task, which studies and judges system, can be divided into application layer, Agent layer, intermediate layer and resource layer.Task task end is
Application layer, environmental context App_Context ends are Agent layer, model caller Model_Invoker ends and data channel
Data_Channel ends are located at intermediate layer, and resource layer includes computing resource and data resource, and computing resource, which refers to, studies and judges model, data
Resource refers to the data of numerous types of data.
3-agent collaborations complete to study and judge the overall flow of task as shown in fig. 7, the interaction protocol for completing to study and judge task includes
Five stages：In the stage 1, check that the model of studying and judging called whether there is, as shown in Figure 2；Stage 2, data resource needed for checking
It whether there is, as shown in Figure 3；Stage 3：Context is initialized, as shown in Figure 4；In the stage 4, Model instantiation is studied and judged,
As shown in Figure 5；Model calculating process is studied and judged in stage 5, execution, as shown in fig. 6, obtaining result of calculation.Each stage is carried out below
Illustrate.
Stage 1：Check that the model of studying and judging called whether there is, as shown in Figure 2.
Step 1.1：Task ends send model call request according to model name Model_Name to App_Context ends；
Model call request is to be judged whether to study and judge model accordingly according to model name；
Step 1.2：App_Context ends propose that corresponding model is called according to model name to Model_Invoker ends
Request；
Step 1.3：Model_Invoker ends propose model call request according to model name to computing resource Model ends；
Step 1.4：Model ends check the model list that itself is maintained；
Step 1.5：If there is model name Model_Name model in model list, to Model_
Invoker ends return to model and there is mark, otherwise, return to model and mark is not present；
Step 1.6：Model_Invoker ends return to the present or absent mark of model to App_Context ends；
Step 1.7：App_Context ends return to the present or absent mark of model to Task ends.
Stage 2：Check that required data resource whether there is, as shown in Figure 3：
Step 2.1：Task ends according to the data of required by task data describe Data_Desc and just know argument sequence to
App_Context ends send data call request；Data resource needed for being checked by data call request whether there is；
Step 2.2：App_Context ends send data call request according to Data_Desc to Data_Channel ends；
Step 2.3：Data_Channel ends carry out protocol testing to Data_Desc, and parse Data_Desc, obtain institute
Need DSN DS_Name, data manipulation Data_Operation, predicate Predicate and the return number where data
According to description；
Data_Channel ends carry out protocol testing to Data_Desc, that is, check whether Data_Desc meets encapsulation rule
Model, if not meeting, provides miscue, then needs to be packaged according to Data_Desc definition again, if met, solve
Data_Desc is analysed, DSN DS_Name, data manipulation Data_Operation, predicate needed for obtaining where data
Predicate and returned data description；
Step 2.4：Data_Channel ends propose request of data according to DS_Name to data resource Data_Source ends；
Step 2.5：Data_Source ends check the data source list that itself is maintained；
Step 2.6：If there is the data source that dataSource link is referred to as DS_Name in data source list, Data_Source
Hold to exist to Data_Channel ends returned data and identify, otherwise mark is not present in returned data；
Step 2.7：Data_Channel ends are to the present or absent mark of App_Context ends returned data；
Step 2.8：App_Context ends are to the present or absent mark of Task ends returned data.
Stage 3：Context App_Context is initialized, as shown in Figure 4：
Step 3.1：Task ends send App_Context initialization requests according to Data_Desc to App_Context ends；
Step 3.2：App_Context ends propose request of data according to Data_Desc to Data_Channel ends；
Step 3.3：Data are described Data_Desc and are mapped as request of data Data_Req by Data_Channel ends, mapping
Rule is as follows：
Data_Desc.DS_Name→Data_Req.DS_Name；
Data_Desc.Data_Operation→Data_Req.Data_Operation；
Data_Desc.Predicate→Data_Req.Predicate；
Data_Desc.Ret_Data_Desc→Data_Req.Ret_Data_Desc；
Rule Expression, by Data_Desc DSN DS_Name, data manipulation Data_Operation, is called above
Word Predicate and returned data describe Ret_Data_Desc, are correspondingly arranged DSN for Data_Req, data
Operation, predicate and returned data description.
Step 3.4：Data_Req is mapped as specific data source command sequence by Data_Channel ends, and mapping ruler is such as
Under：
Rule Expression, judges it is DB or SDB, then to inhomogeneity according to DS_Req data type DS_Type above
Different pieces of information operation Data_Operation under type data, is converted into corresponding data command Cmd.Cmd is to data manipulation
Specific descriptions.
Step 3.5：Command sequence Cmd and returned data are described Ret_Data_Desc and are sent to by Data_Channel ends
Data_Source ends；
Step 3.6：Data_Source ends perform command sequence, and describe Ret_Data_Desc operations according to returned data
Data object；
Step 3.7：Data object Data_Objects is returned to Data_Channel ends by Data_Source ends；
Step 3.8：Obtained data object Data_Objects is returned to App_Context by Data_Channel ends
End；
Step 3.9：App_Context ends are initialized；
Step 3.10：App_Context ends return to initialization to Task ends and complete mark.
Stage 4：Model instantiation is studied and judged, as shown in Figure 5：
Step 4.1：Task ends propose Model instantiation request according to model name to App_Context ends；
Step 4.2：App_Context ends propose Model instantiation request according to model name to Model_Invoker ends；
Step 4.3：Model_Invoker ends propose Model instantiation request according to model name to Model ends；
Step 4.4：Model is studied and judged for Model_Name in Model ends to Model_Invoker ends return model name；
Step 4.5：Model_Invoker ends carry out protocol testing to the model of studying and judging of return, and parse this and study and judge model,
Obtain input variable sequence<in1,…,inp>, argument sequence<p1,…,pq>, model calculating process proc and output variable
Sequence<out1,…,outr>；
Model_Invoker ends carry out protocol testing to studying and judging model, that is, check and study and judge whether model meets encapsulation specification,
If not meeting, miscue is provided, then needs to be packaged according to the definition for studying and judging model again, if met, parsing is ground
Sentence model.Obtain input variable sequence<in1,…,inp>, argument sequence<p1,…,pq>, model calculating process proc and defeated
Go out Variables Sequence<out1,…,outr>；
Step 4.6：Model_Invoker ends are interacted with App_Context ends, by input variable sequence<in1,…,inp>、
Argument sequence<p1,…,pq>Mapped with business datum sequence and supplemental characteristic sequence；
Step 4.7：App_Context ends return to input variable sequence and argument sequence gives Model_Invoker ends；
Step 4.8：Input variable sequence and argument sequence pair that Model_Invoker ends are returned with App_Context ends
Study and judge model to be instantiated, Model instantiation success flag is returned into App_Context ends；
Step 4.9：Model instantiation success flag is returned to Task ends by App_Context ends.
Stage 5:Model calculating process is studied and judged in execution, obtains result of calculation, as shown in Figure 6：
Step 5.1：Task ends propose model computation requests；
Step 5.2：The model computation requests received are sent to Model_Invoker ends by App_Context ends；
Step 5.3：Model_Invoker ends perform model calculating process proc, and obtained result of calculation is put into model
Output variable sequence<out1,…,outr>Corresponding output variable in；
Step 5.4：Model_Invoker ends are by model output variable sequence<out1,…,outr>Return to App_
Context ends；
Step 5.5：App_Context ends are by model output variable sequence<out1,…,outr>Return to Task ends；
Step 5.6：Task ends are by output variable sequence<out1,…,outr>In output variable insert successively in list,
Task call is studied and judged to terminate.
Illustrated below by taking international macroseism application disposal system as an example.
In embodiment, international macroseism application disposal system needs to study and judge the death toll caused by earthquake, earthquake
Human mortality's decision model is calculated according to earthquake intensity value and population kilometer Grid square, is obtained centered on focus, one
Determine the magnitude of the human mortality and death toll in region.According to the earthquake human mortality decision model of reference paper [1] such as
Shown in lower：
Reference paper [1]:Kishor Jaiswal,David Wald.An empirical model for global
earthquake fatality estimation[J].Earthquake Spectra,2010,26(4):1017-1038。
The implication of each symbol is as follows in the formula：
●S：Earthquake intensity circle is represented, is an ellipse；
●s：Represent earthquake intensity circle S earthquake intensity value；S is located at s/ θ molecule；
●θ,β：Model parameter, by earthquake centre, point is uniquely determined, and the value of country variant is different；
●Φ[·]：Represent the accumulation probability of standardized normal distribution；
●P(S)：Population in earthquake intensity circle S；
●E(S)：Human mortality's number in earthquake intensity circle S.
The encapsulation specification provided according to the inventive method, to studying and judging task call interface, computing resource calling interface, data
Resource transfer interface, study and judge model resource and data resource is packaged.
Task end Task is packaged according to encapsulation specification, represented using XML, it is as follows：
Above in code：
Task nodes represent that the name attributes in task object, task nodes represent task names, are herein t1；
Form nodes represent report objects；
Inits nodes in form nodes represent the initial parameter sequence of disaster, and init nodes represent an initial parameter,
The name attributes of init nodes represent parameter name, and value attributes are represented in parameter value, upper example：
X represents that the longitude of point occurs for earthquake；
Y represents that the latitude of point occurs for earthquake；
Depth represents that the depth of point occurs for earthquake；
Mag represents the earthquake magnitude of earthquake；
Loc represents the location information of earthquake；
Time represents the time of origin of earthquake；
Results nodes represent the implementing result sequence of task, and the task has a result of calculation, name attribute specification meters
Calculate the title of result；
Models nodes represent that what the task to be called studies and judges Model sequence；
Model name is studied and judged in the title for studying and judging model to be called of model node definitions, this example for casualty_
usgs_pager。
Model will be studied and judged -- earthquake human mortality decision model is packaged according to encapsulation specification, is represented using XML, as follows
It is shown：
Model is studied and judged in the embodiment of the present invention, during model name is casualty_usgs_pager, input variable sequence
Comprising a variable S, included in argument sequence comprising a variable E (S) in three parameter θs, β and P (S), output variable sequence,
Model calculating process is described in produres.
Data description needed for earthquake human mortality's decision model is packaged according to rule, represented using XML, it is as follows
It is shown：
Next 3-agent and its interaction protocol are realized, as shown in Fig. 2~Fig. 6, including such as next stage and step：
Stage 1：Check that the model of studying and judging called whether there is, as shown in Fig. 2 Task ends pass through model name
Casualty_usgs_pager sends model call request to App_Context ends；App_Context ends pass through model name
Casualty_usgs_pager proposes model call request to Model_Invoker ends；Model_Invoker ends pass through model
Title proposes model call request to Model ends；Model ends check the model list that itself is maintained, and there is model name is
Casualty_usgs_pager model, returns to Model_Invoker ends and there is mark；Model_Invoker ends are to App_
Context ends return to model and there is mark；App_Context ends return to model to Task ends and there is mark.
Stage 2：Check that required data resource whether there is, as shown in figure 3, data needed for Task ends are calculated by model are retouched
State and just know argument sequence to App_Context end data call requests；App_Context ends by Data_Desc to
Data_Channel ends send data call request；Data_Channel ends carry out protocol testing to Data_Desc, and parse
Data_Desc；Data_Channel ends propose request of data by DSN ds1 to Data_Source ends；Data_
Source ends check the data source list that itself is maintained, and there is the data source that dataSource link is referred to as ds1, to Data_Channel ends
Return and there is mark.
Stage 3：Context App_Context is initialized, as shown in figure 4, Data_Channel ends are by Data_
Desc is mapped as request of data Data_Req；Data_Req is mapped as specific data source command sequence by Data_Channel ends
Row, it is as follows：
SELECT beta FROM PARAM_TABLE WHERE code=' China '；
SELECT theta FROM PARAM_TABLE WHERE code=' China '；
SELECT population from EG_asPopulation WHERE ZONAL(circle)；
The description Ret_Data_Desc of command sequence and returned data is sent to Data_ by Data_Channel ends
Source ends；Data_Source ends perform command sequence, and describe Ret_Data_Desc peration datas pair according to returned data
As data object Data_Objects then is returned into Data_Channel ends；Data_Channel ends are by obtained data
Object Data_Objects returns to App_Context ends, the initialization of App_Context ends；App_Context ends are to Task ends
Return to initialization and complete mark.
Stage 4：Model instantiation is studied and judged, as shown in figure 5, Task ends pass through model name casualty_usgs_pager
Call request is proposed to App_Context ends；App_Context ends by model name casualty_usgs_pager to
Model_Invoker ends propose model call request；Model_Invoker ends pass through model name casualty_usgs_
Pager proposes model call request to Model ends；It is Model_ that Model ends return to model name to Model_Invoker ends
Name model object；Model_Invoker ends carry out protocol testing to the model object of return, and parse the model object；
Model_Invoker ends are interacted with App_Context ends, by input variable sequence<S,s,P(S)>, argument sequence<θ,β>With industry
Business data sequence and supplemental characteristic sequence are mapped；The data that Model_Invoker ends are returned with App_Context ends are to mould
Type is instantiated.
Stage 5：Model calculating process is studied and judged in execution, obtains result of calculation, as shown in fig. 6, Model_Invoker ends are performed
Model calculating process, model output variable sequence is put into by obtained result of calculation<E(S)>Corresponding output variable in；
Model_Invoker ends are by model output variable sequence<E(S)>Return to App_Context ends；App_Context ends are by mould
Type output variable sequence<E(S)>Return to Task ends；Task ends are by output variable sequence<E(S)>In output variable fill out successively
Enter in list, study and judge task call and terminate.
Claims (8)
1. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis, it is characterised in that including as follows
Step：
Step 1：Task is studied and judged into module abstracts for three parts, referred to as 3-agent, 3-agent is respectively context
App_Context, model caller Model_Invoker and data channel Data_Channel；
The set of context record data source, business datum sequence and supplemental characteristic sequence and initial parameter sequence,
Context is sky under original state；
Model caller includes model loader, mapper and actuator；Model loader is used for the loading of model, and model is entered
Row protocol testing and analytic modell analytical model；Mapper is used to set up mode input Variables Sequence and argument sequence reflecting to data source
Penetrate, mode input variable and model parameter are instantiated；Actuator is used for the execution of model calculating process；
Data channel includes request receiver, request of data analyzer, sequencing unit and data hinge；Request receiver is received
The request of data that request of data end is sent；Request of data analyzer is responsible for protocol testing and the parsing of request of data, then by number
Data source command is mapped as according to request；Sequencing unit is used to connect data source, sends and orders to data source；Data hinge is by number
The data returned according to source return to request of data end；
Step 2：Abstract package is carried out to studying and judging model, data source, computing resource and data resource is formed；
Studying and judging the data structure of model includes：Model name, input variable sequence, argument sequence, output variable sequence and mould
Type calculating process；The data structure of data source includes：DSN, data source types, data source address and data acquisition system；Number
It is made up of according to set data, the definition of each data is defined on data pair including data name, one group of data object and one group
As upper operation；The definition of each data object includes data object title and the value of data object；
Step 3：Calling interface is encapsulated, including studies and judges task call interface, computing resource calling interface and data resource transfer and is connect
Mouthful；
Study and judge task call interface include study and judge task names, initial parameter list, call study and judge model name and task
The data description of required data；Computing resource calling interface is to study and judge model name；Data resource calling interface includes required
Data name, DSN, data manipulation, the predicate for determining data and returned data description；
Step 4：Interaction protocol is designed, task is studied and judged in 3-agent collaboration completions.
2. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 1,
Characterized in that, described request of data, its data structure includes：The DSN of request, the operation to data object, use
In it is determined that predicate and the returned data description of target data objects.
3. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 1,
Characterized in that, in described step 4, the interaction protocol that 3-agent collaborations complete to study and judge task includes five stages：Stage 1,
Check that the model of studying and judging called whether there is；In the stage 2, check that required data resource whether there is；Stage 3：In application environment
Hereafter initialize；In the stage 4, study and judge Model instantiation；Model calculating process is studied and judged in stage 5, execution, obtains result of calculation.
4. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 3,
Characterized in that, the described stage 1 is realized especially by following steps：
Step 1.1：Task end sends model call request according to model name to App_Context ends；
Step 1.2：App_Context ends propose model call request according to model name to Model_Invoker ends；
Step 1.3：Model_Invoker ends propose model call request according to model name to computing resource；
Step 1.4：Computing resource checks the model list that itself is maintained；
Step 1.5：If there is the model of the model name in model list, return to model to Model_Invoker ends and deposit
In mark, otherwise, return to model and mark is not present；
Step 1.6：Model_Invoker ends return to the present or absent mark of model to App_Context ends；
Step 1.7：App_Context ends return to the present or absent mark of model to task end.
5. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 3,
Characterized in that, the described stage 2 is realized especially by following steps：
Step 2.1：Task end describes Data_Desc according to the data of required by task data and just knows argument sequence to App_
Context ends send data call request；
Step 2.2：App_Context ends send data call request according to Data_Desc to Data_Channel ends；
Step 2.3：Data_Channel ends carry out protocol testing to Data_Desc, and parse Data_Desc, obtain data institute
DSN, data manipulation, predicate and returned data description；
Step 2.4：Data_Channel ends propose request of data according to DSN to data resource；
Step 2.5：Data resource checks the data source list that itself is maintained；
Step 2.6：If there is the data source of corresponding data source name in data source list, data resource is to Data_
There is mark in Channel ends returned data, otherwise mark is not present in returned data；
Step 2.7：Data_Channel ends are to the present or absent mark of App_Context ends returned data；
Step 2.8：App_Context ends are to the present or absent mark of task end returned data.
6. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 3,
Characterized in that, the described stage 3 is realized especially by following steps：
Step 3.1：Task end describes Data_Desc to App_Context ends according to the data of required by task data and sends App_
Context initialization requests；
Step 3.2：App_Context ends propose request of data according to Data_Desc to Data_Channel ends；
Step 3.3：Data are described Data_Desc and are mapped as request of data Data_Req by Data_Channel ends；
Step 3.4：Data_Req is mapped as specific data source command sequence by Data_Channel ends；
Step 3.5：Data source command sequence and returned data description are sent to data resource by Data_Channel ends；
Step 3.6：Data resource performs command sequence, and describes peration data object according to returned data；
Step 3.7：Data object is returned to Data_Channel ends by Data_Source ends；
Step 3.8：Obtained data object is returned to App_Context ends by Data_Channel ends；
Step 3.9：App_Context ends are initialized；
Step 3.10：App_Context ends return to initialization to task end and complete mark.
7. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 3,
Characterized in that, the described stage 4 is realized especially by following steps：
Step 4.1：Task end proposes Model instantiation request according to model name to App_Context ends；
Step 4.2：App_Context ends propose Model instantiation request according to model name to Model_Invoker ends；
Step 4.3：Model_Invoker ends propose Model instantiation request according to model name to computing resource；
Step 4.4：Computing resource returns to the model object of corresponding model title to Model_Invoker ends；
Step 4.5：Model_Invoker ends carry out protocol testing to the model object of return, and parse the model object, obtain
Input variable sequence, argument sequence, output variable sequence and model calculating process；
Step 4.6：Model_Invoker ends are interacted with App_Context ends, by input variable sequence and argument sequence, with industry
Business data sequence and supplemental characteristic sequence are mapped；
Step 4.7：App_Context ends return to input variable sequence and argument sequence gives Model_Invoker ends；
Step 4.8：The input variable sequence and argument sequence that Model_Invoker ends are returned with App_Context ends are to studying and judging
Model is instantiated, and Model instantiation success flag is returned into App_Context ends；
Step 4.9：Model instantiation success flag is returned to task end by App_Context ends.
8. data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis according to claim 3,
Characterized in that, the described stage 5 is realized especially by following steps：
Step 5.1：Task end proposes model computation requests；
Step 5.2：The model computation requests received are sent to Model_Invoker ends by App_Context ends；
Step 5.3：Model_Invoker ends perform model calculating process, and the output that obtained result of calculation is put into model becomes
Measure in the corresponding output variable of sequence；
Step 5.4：The output variable sequence of model is returned to App_Context ends by Model_Invoker ends；
Step 5.5：The output variable sequence of model is returned to task end by App_Context ends；
Step 5.6：Task end inserts the output variable in output variable sequence in list successively, studies and judges task call and terminates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510156076.8A CN104714808B (en) | 2015-04-03 | 2015-04-03 | Data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510156076.8A CN104714808B (en) | 2015-04-03 | 2015-04-03 | Data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104714808A CN104714808A (en) | 2015-06-17 |
| CN104714808B true CN104714808B (en) | 2017-09-19 |
Family
ID=53414185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510156076.8A Expired - Fee Related CN104714808B (en) | 2015-04-03 | 2015-04-03 | Data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104714808B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1548987A (en) * | 2003-05-16 | 2004-11-24 | 远传电信股份有限公司 | Geographical disaster monitoring and alarming system of application communication network |
| CN101477207A (en) * | 2009-01-20 | 2009-07-08 | 中国科学院水利部成都山地灾害与环境研究所 | Intelligent geological calamity synthetic monitoring system and multi-stage prediction analysis method |
| CN104463759A (en) * | 2014-11-28 | 2015-03-25 | 东莞中国科学院云计算产业技术创新与育成中心 | Decision-making and command all-in-one machine based on digital schema and operation method of all-in-one machine |
-
2015
- 2015-04-03 CN CN201510156076.8A patent/CN104714808B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1548987A (en) * | 2003-05-16 | 2004-11-24 | 远传电信股份有限公司 | Geographical disaster monitoring and alarming system of application communication network |
| CN101477207A (en) * | 2009-01-20 | 2009-07-08 | 中国科学院水利部成都山地灾害与环境研究所 | Intelligent geological calamity synthetic monitoring system and multi-stage prediction analysis method |
| CN104463759A (en) * | 2014-11-28 | 2015-03-25 | 东莞中国科学院云计算产业技术创新与育成中心 | Decision-making and command all-in-one machine based on digital schema and operation method of all-in-one machine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104714808A (en) | 2015-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110989983B (en) | Zero-coding application software rapid construction system | |
| US8645906B2 (en) | Method for enforcing change policy based on project state | |
| CN108197261A (en) | A kind of wisdom traffic operating system | |
| CN110291517A (en) | Query language interoperability in chart database | |
| CN111176867B (en) | Data sharing exchange and open application platform | |
| US11303651B1 (en) | Security appliance to monitor networked computing environment | |
| US20080066050A1 (en) | Calculating defect density by file and source module | |
| CN108270640B (en) | Information interoperation consistency test method for intelligent marketing and distribution information integration system | |
| US20070073702A1 (en) | Systems and methods for mapping text | |
| US20090210748A1 (en) | Methods and systems to test airline information systems | |
| CN111191331A (en) | Transformer substation graph-model data quality verification device and method based on CIM and SVG | |
| Clark et al. | Event driven architecture modelling and simulation | |
| Giannasi et al. | Enhancing confidence in discrete event simulations | |
| Di Lucca et al. | Abstracting business level UML diagrams from web applications | |
| CN104714808B (en) | Data and computing resource virtualized management method in a kind of calamity emergency comprehensive analysis | |
| CN107748719A (en) | Test report generation method and device | |
| JP2007535013A (en) | Computer program design | |
| Liu | All in one: Design, verification, and implementation of SNOW-optimal read atomic transactions | |
| CN111143432A (en) | Data analysis early warning system and method for event processing result | |
| CN115357656A (en) | Information processing method and device based on big data and storage medium | |
| Pandey | Investigating community, reliability and usability of cfengine, chef and puppet | |
| Babac et al. | AgentTest: A specification language for agent-based system testing | |
| CN115168474A (en) | Internet of things center station system building method based on big data model | |
| CN108470087B (en) | Data bus of ramjet design simulation platform | |
| CN111245676B (en) | Communication protocol credibility verifying device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Li Qinyong Inventor after: Lv Jianghua Inventor after: Wei Yanyan Inventor after: Song Jiangong Inventor after: Du Jianhai Inventor after: Ma Shilong Inventor before: Li Qinyong Inventor before: Wei Yanyan Inventor before: Song Jiangong Inventor before: Du Jianhai Inventor before: Ma Shilong |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170919 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |