CN107508698A - Software definition service reconfiguration method based on perception of content and weighted graph during mist calculates - Google Patents
Software definition service reconfiguration method based on perception of content and weighted graph during mist calculates Download PDFInfo
- Publication number
- CN107508698A CN107508698A CN201710597297.8A CN201710597297A CN107508698A CN 107508698 A CN107508698 A CN 107508698A CN 201710597297 A CN201710597297 A CN 201710597297A CN 107508698 A CN107508698 A CN 107508698A
- Authority
- CN
- China
- Prior art keywords
- service
- content
- mist
- data
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
- H04L41/0833—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability for reduction of network energy consumption
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5041—Network service management, e.g. ensuring proper service fulfilment according to agreements characterised by the time relationship between creation and deployment of a service
- H04L41/5051—Service on demand, e.g. definition and deployment of services in real time
Abstract
A kind of software definition service reconfiguration method based on perception of content and weighted graph in being calculated the invention provides mist, operate in the mist node architecture of a stratification, mist node is made up of three levels, respectively service reconstruction of layer, content driven key-course, user-defined reconfigure layer, simultaneously, three kinds of corresponding memory spaces are marked off in mist node, that is Service Source storehouse, the policy library based on content, demand storehouse, respectively to store-service module, content operation mapping table, mist user pretreatment demand.New software in mist node proposed by the present invention defines service reconfiguration mechanism compared with the service in traditional mist node provides mechanism, time cost and calculation cost can be largely reduced, so as to reduce mist system entirety energy consumption.
Description
Technical field
Software definition service reconfiguration mechanism based on perception of content and weighted graph in being calculated the present invention relates to a kind of mist.
Background technology
As the intermediate layer of distal end cloud platform and a large amount of underlying devices, mist is calculated abilities such as contents processing, calculating, storages
Concentrate in the node that network edge has certain intelligence.In this mode, gathered from bottom large-scale sensor network
To initial data will not directly be uploaded to high in the clouds, but related pretreatment is first performed according to user's request by local mist node
Afterwards, then cloud is consigned to and upper-layer user does further storage and processing.At present, mist calculating is because it in low delay and can mitigate cloud
The advantage of computation burden etc. is considered as radio sensing network of future generation (WSNs) trend.
How the progressively application and scale deployment calculated with mist, efficiently utilize its limited meter in each mist node
Calculation ability and related resource, so as to realize that the energy-conservation in mist calculating aspect seems very necessary.In the application and deployment that mist calculates
Under scene, on the one hand, different mists, which calculate user, different demands, and each use to the data prediction strategy in mist node
The demand at family shows the characteristic changed over time;On the other hand, the pre- place of different content type (such as video, audio) data
Reason demand also has larger difference.The dynamic and difference of demand are pre-processed present on user and data content the two aspects
Property so that User Defined strategy and the pre-processing service based on content reassemble into the pass efficiently utilized for mist node resource
Key.However, the mist node architecture proposed at present has the characteristics of relative solidification and nature static, this demand can not be met.
In the research for WSNs power savings in the past, most of solutions surround Resource Scheduling Mechanism in cloud, with
Realize the Energy-aware under the scene.Research for the mist node architecture of optimization has also been concerned only with each section in mist meshed network
Point optimizes, can not reduced whole in the interaction of network level, the efficient reuse and restructuring for being not directed to resource in single mist node
The energy consumption of body mist computing system.
The content of the invention
For in the prior art the defects of, it is an object of the invention to provide a kind of mist calculate in be based on perception of content and cum rights
The software definition service reconfiguration method of figure, by being abstracted, encapsulating by mist node computing resource and modular implementation resource
Recycle and can recombinate, different data prediction strategies is used according to the difference of the content type perceived, meanwhile, to be upper
Layer mist user provides programmable interface to allow user to carry out self-defined control to the bottom data pretreatment in mist node.One side
Face, the mechanism realize the efficient layout of mist node resource with using while different demands are met;On the other hand, handle
The flexibility of strategy is also greatly improved with dynamic.
The technical solution adopted in the present invention is as follows:
A kind of software definition service reconfiguration method based on perception of content and weighted graph during mist calculates, operates in a level
In the mist node architecture of change, mist node is made up of three levels, respectively services reconstruction of layer, the key-course of content driven, user
Definition reconfigures layer, meanwhile, mark off three kinds of corresponding memory spaces, i.e. Service Source storehouse, the policy library based on content, demand
Storehouse, respectively to store-service module, content-operation mapping table, the pretreatment demand of mist user.
The mist node is collected from data packet head and in load and extracts important data attribute, so as to complete to content
Classification, it is and different then using content tab technology to a large amount of original numbers caused by radio sensing network based on content type
According to being identified, subsequently to perform different pretreatment strategies to the content-data of different labels.
Mist user is reconfigured the interface that layer provides upwards specific bottom pretreatment strategy is carried out using user-defined
Control, call the service module in service scheduling process and be consistent with itself actual demand, meanwhile, mist user passes through DLL
The flexible management to service module, including rewriting, change, deletion, increase to service module are completed, dynamic weight is serviced to improve
The flexibility of group, ensure that by the layout of modularization resource various pretreatment demand can be met, and then reduce and redeploy
Energy consumption caused by mist node.
When selection module is instantiated, a pretreatment strategy is by one or more operation according to certain logical order
Composition, same operation correspond to one or more service modules that the operation is realized with different calculation costs and algorithm again, often
The data of kind content type correspond to different pretreatment strategies, that is, the combination of different service modules be correspond to, by CBL technologies
Initial data after labeling will search corresponding processing strategy in the policy library stored in the key-course of content driven
And the service module that the strategy need to be transferred is completed, the pretreatment process then strategically pre-defined in storehouse is to required service
Module carries out layout with certain logical order, and transfers to the service reconstruction of layer execution module of mist node to instantiate, parameter transmission,
Finally exported, the data after processing or valuable result are consigned into upper strata mist user.
The pretreatment demand of data, non-upper layer mist user and its time-varying for different content type, it is not necessary to be every
Individual strategy all a set of single preprocessor of deployment configuration, it is only necessary to which output and user's practical business based on perception of content need
Will, one or more service modules are called, when changes in demand, service module are recycled and dynamic reorganization, you can
While the flexibly demand of reply differentiation, the utilization ratio of Service Source is improved, reduces mist node energy consumption.
Service reconstruction of layer include be used for call, instantiate required service module and control its execution service reconfiguration unit,
For the initial data that is collected from radio sensing network is carried out feature extraction and classification perception of content unit, for pair
Each service module is registered, additions and deletions, the module management unit of renewal, meanwhile, service the Service Source that sets in reconstruction of layer
Storehouse, including the data buffer area temporarily stored for carrying out initial data, and for depositing the service of abstract encapsulation resource
Module pond, a large amount of initial data being collected into from bottom radio sensing network will be temporarily placed in after mist node is uploaded to
Data buffer area etc. is pending, and the packet into pretreatment process will be transmitted to perception of content unit first, the unit from
Simultaneously analysis of key attribute is extracted in data packet head and in load, the classification results to content is exported, is then based on different contents
Type, initial data caused by radio sensing network is identified using content tab technology, so as to complete subsequently according to content
Into rational service reconstruct.
After the specific pretreatment strategy and required service module that the key-course for receiving content driven issues, reconstruction of layer is serviced
Corresponding module will be chosen from service module pond, it will be instantiated with the parameter specified in default or strategy, and according to
Fixed logical order carries out layout and restructuring to module, so that it completes specific data processing needs, is received when from bottom-layer network
Collect different types of packet or when upper strata mist user is updated to data processing policy, service reconstruction of layer can be in existing clothes
On the basis of module of being engaged in and layout, it, which is replaced, or recombinated can meet new pretreatment demand.
The key-course of content driven is including for the demand resolution unit analyzed user's request and for carrying out
Data prediction strategy generating and optimization policy making unit, meanwhile, set in the key-course of content driven based on content
Policy library, including for storing the service logic sequence list of preset data handling process, and for defining different content class
The content of type data and its required respective operations-operation mapping table.
Policy making unit will be uploaded to the packet of content tab technology identification, by the unit in content-operation
The processing operation needed for it is searched in mapping table, meanwhile, if upper strata mist user has special demand, these demands will be also issued to
Demand resolution unit, for controlling the execution of specific operation or the selection of module.
During strategy generating, including two sub-steps, it is the policy development based on weighted graph and strategy respectively
Optimization, wherein:Policy development refers to service reconstruction of layer module according to needed for the service strategy selection that the key-course of content driven issues
Initialized and reconfigured, so that it meets the needs of different mist users and contents processing, obtain being possible in certain class
The data prediction strategy performed in content;Policy optimization refers to the particular demands issued according to energy consumption and user to data prediction
Strategy optimizes, and chooses the module restructuring of relatively reasonable strategy instruction service reconstruction of layer.
Compared with prior art, the present invention has following beneficial effect:
The present invention solves mist and calculated under scene, is meeting the same of the differentiation process demand of mist user and different pieces of information content
When, realize the efficient utilization of resource in single mist node.The present invention by by mist node computing resource carry out it is abstract with encapsulate with
Allow service reconfiguration, i.e., when process demand changes, without reconfiguring and disposing total data handling process, and only need
Modular processing operation, which is rearranged, can realize resource reuse and policy update.Pass through the service of the software definition
Recombination mechanism promotes the energy saving optimizing of single mist node, further, reduces the energy consumption of whole mist system.
Brief description of the drawings
The detailed description made by reading with reference to the following drawings to non-limiting example, further feature of the invention,
Objects and advantages will become more apparent upon:
Fig. 1 is architecture of the present invention;
Fig. 2 reconstructs schematic diagram for present invention service;
Fig. 3 is that service strategy of the present invention is formulated and demand parses interaction models;
Fig. 4 provides time delay comparison diagram under mechanism for different services;
Fig. 5 is the complexity comparison diagram provided under mechanism of the present invention distinct device service.
Embodiment
With reference to specific embodiment, the present invention is described in detail.Following examples will be helpful to the technology of this area
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill to this area
For personnel, without departing from the inventive concept of the premise, some changes and improvements can also be made.These belong to the present invention
Protection domain.
Software definition service reconfiguration mechanism based on perception of content and weighted graph during mist provided by the present invention calculates, its base
Plinth framework is as shown in Figure 1.
The service reconfiguration mechanism that mist calculates lower software definition is operated in the mist node architecture of a stratification.Mist node leads to
Cross and provide cloud node edge to calculating to different content categorical data, storage, pretreatment etc. local for upper strata mist user
Service function.The service orchestration mechanism for this mist node that we are proposed and traditional mist node pre-processing service
Maximum difference is that carries out modularized encapsulation by the calculating Service Source of mist node, it is achieved thereby that the service weight of software definition
Group and resource reuse, and then resource utilization is improved, reach the purpose for the overall energy consumption for reducing mist system.
As shown in figure 1, it is made up of corresponding to this service reconfiguration mechanism mist node three levels, respectively service reconstruct
Layer, content driven key-course, user-defined reconfigure layer.Meanwhile also need to mark off in mist node and deposited corresponding to three kinds
Space, i.e. Service Source storehouse, the policy library based on content and demand storehouse are stored up, is reflected respectively to store-service module, content-operation
Firing table and the pretreatment demand of mist user.
Because mist node has the characteristic of perception of content, it can collect from data packet head and in load and extract important
Data attribute, so as to complete the classification to content, and the difference based on content type then utilizes CBL (content tab) technology
A large amount of initial data caused by WSNs (radio sensing network) network are identified, so as to subsequently to the content number of different labels
According to the different pretreatment strategy of execution.Mist user reconfigures interface that layer provides upwards to specific bottom using user-defined
Pretreatment strategy is controlled, and is called the module in service scheduling process and is consistent with itself actual demand, meanwhile, mist user is led to
Flexible management of the DLL completion to service module, including rewriting, change, deletion, increase to module are crossed, with the service of improving
The flexibility of dynamic reorganization, ensure that by the layout of modularization resource various pretreatment demand can be met, and then reduce weight
Energy consumption caused by new deployment mist node.
When selection module is instantiated, a pretreatment strategy is by one or more operation according to certain logical order
Composition, same operation correspond to one or more service modules that the operation is realized with different calculation costs and algorithm again.Often
The data of kind content type correspond to different pretreatment strategies, that is, correspond to the combination of different service modules.By CBL technologies
Initial data after labeling will search corresponding processing strategy in the policy library stored in the key-course of content driven
And the module that the strategy need to be transferred is completed, the pretreatment process then strategically pre-defined in storehouse is to required service module
Layout is carried out with certain logical order, and transfers to the service reconstruction of layer execution module of mist node to instantiate, parameter transmission, finally
Exported, the data after processing or valuable result are consigned into upper strata mist user.
In present mechanism, the pretreatment demand of data, non-upper layer mist user and its time-varying for different content type,
Need not be a set of single preprocessor of each tactful deployment configuration, it is only necessary to output and user based on perception of content
Actual business requirement, one or more service modules are called, when changes in demand, they are recycled and dynamic is heavy
Group, you can while the flexibly demand of reply differentiation, improve the utilization ratio of Service Source, reduce mist node energy consumption.
It is the specific design that Reconstruction Mechanism is serviced in mist node architecture proposed by the invention shown in Fig. 2.In the mechanism
Key component include be used for call, instantiate required service module and control its execution service reconfiguration unit, for from
The initial data collected in WSNs networks carries out the perception of content unit of feature extraction and classification, and for each service
Module is registered, additions and deletions, renewal module management unit.Meanwhile Service Source storehouse is set, the resources bank include being used for into
The data buffer area that row initial data temporarily stores, and for depositing the service module pond of abstract encapsulation resource.
The a large amount of initial data being collected into from bottom WSNs networks will be temporarily placed in number after mist node is uploaded to
It is pending according to buffer area etc..Packet into pretreatment process will be transmitted to perception of content unit first, and the unit is from number
According to simultaneously analysis of key attribute is extracted in packet header and in load, the classification results to content are exported, are then based on different content classes
Type, initial data caused by WSNs networks is identified using CBL technologies, subsequently to complete rational service according to content
Reconstruct.
On the other hand, after the specific pretreatment strategy and required service module that the key-course for receiving content driven issues,
Service reconfiguration unit will choose corresponding module from service module pond, and example is carried out to it with the parameter specified in default or strategy
Change, and layout and restructuring are carried out to module according to given logical order, so that it completes specific data processing needs.
Here, the operation for forming some processing strategy correspond to one or more service modules realized by algorithms of different,
So as to which the module progress data prediction for choosing different has different calculation costs and energy consumption.Service module and its corresponding behaviour
Contextual definition between work is:
Wherein, DPOi represents the set for the in kind difference service modules that i-th kind of data processing operation is included.
Each service module and the interactive relation between them are represented by:
Wherein,O represents the concrete operations that the algorithm encapsulated in the module performs to packet;In represents the service
A series of inputs of module;Out represents a series of outputs of the service module;Represent to be initialized to the module
Relevant parameter, these parameters by service reconfiguration unit with numeral or string format specify, the meaning involved by parameter includes
But data compression rate, filter condition, data package size threshold value etc. are not limited to, these parameters will be used with current network state and mist
The change dynamic of family demand adjusts.
When being collected into different types of packet from bottom-layer network or upper strata mist user is updated to data processing policy
When, service Reconstruction Mechanism can on the basis of existing service module and layout, it is replaced or recombinated can meet it is new
Pretreatment demand.
The controlling mechanism of present disclosure driving is to realize the core of software definition service reconfiguration in mist node, including with
In the demand resolution unit analyzed user's request, and for carrying out the strategy of data prediction strategy generating and optimization
Formulate unit.Meanwhile the policy library based on content is set, the policy library includes being used for the clothes for storing preset data handling process
Business logical order table, and for defining content-operation mapping table of different content categorical data and its required respective operations.Should
Mechanism and the interaction models of other units are as shown in Figure 3.
Policy making unit will be uploaded to the packet of CBL technology identifications, by the unit in content-operation mapping table
The middle processing operation searched needed for it.Meanwhile if upper strata mist user has special demand, these demands will also be issued to demand solution
Unit is analysed, for controlling the execution of specific operation or the selection of module.During strategy generating, there are two sub-steps, point
It is not the policy development based on weighted graph, and policy optimization.
1) policy development
Service reconstruction of layer module according to needed for being chosen service strategy that the key-course of content driven issues carry out initialization and
Reconfigure, so that it meets the needs of different mist users and contents processing.Service strategy is defined as:
DP={ mod, para, ord }, (3)
Wherein, mod represents required service module;Para represents the initial parameter used in preprocessing process;Ord represents each
The execution sequence of individual operation.
During policy development, the service module being related in strategy and the transforming relationship between them are made in strategy
A global weighted graph is abstracted as in order memberWhereinTo be several by S
The finite nonempty set of point composition is closed, and each node represents a kind of possible intermediateness of data in preprocessing process; ForA binary subclass, a line is corresponded in weighted graph, represents that some is special
Fixed service module.One connection ni and nj side eij shows that pending data to be made is transferred to its critical condition j from state i
The operation performed by module corresponding to the eij of side need to be called.Under normal circumstances, have at two the front and rear node (state) after relation it
Between multiple summits (service module) be present because same operation can by many algorithms realize module complete.
Policy making unit, can be according to its label in after the initial data Cg with feature extraction by analysis is received
All operations needed for such data are searched in appearance-operation mapping table, and are determined with reference to packet current state in tactful way
Starting point nl and terminating point nk;Next, the pretreatment strategy subgraph for being data configuration corresponding to the content type corresponding to it,
Follow service logic sequence list and the processing that should be included operation spanning subgraph E=(Vs,Es), include in the subgraph all
The strategy of content Cg processing pretreatment demands can be met.
2) policy optimization
Tactful subgraph E=(V corresponding to the initial data Cg for certain class content generated during policy developments,Es)
Afterwards, just obtained being possible to the data prediction strategy performed on Cg, each strategy includes different service modules, i.e.,
It correspond to different calculation costs and mist user's certain moduli block dispatching requirement.Next the spy that should be issued according to energy consumption and user
Determine demand to optimize this initial policy collection, choose the module restructuring of relatively reasonable strategy instruction service reconstruction of layer.
To reduce overall energy consumption, adjacency matrix is introduced during policy optimization to enter to the Different Strategies of strategy concentration
Row is weighed and compared.If G is the packet total amount that input receives pretreatment to mist node in some time window, remember therein each
Packet is Cg, wherein g=1,2 ..., G.Remember the number that H is required pretreatment operation, K is that can realize some with many algorithms
The number of modules of concrete operations, then hk represent k-th corresponding of service module of h-th of operation.Here it there will naturally be relation H=
S -1, wherein S represent the stateful number that its in data handling procedure is likely to be at.With E=(eij), to mark global place
Manage policy mapCorresponding adjacency matrix, wherein:
It is similar, define the submatrix of global policiesThe matrix is by G × K BooleanGroup
Into, wherein:
MatrixTo identify global policies figureIn side { eijWhether being selected into subgraph, i.e., h-th operation is
The no data C for certain types of contentgIt is required.
In order to assess the resource consumption total amount of each strategy in the set of strategies of previous step generation, weight matrix is defined In matrix and each elementFor the calculation cost of k-th of module corresponding to h-th of operation.Policy optimization
Purpose is on the premise of user's request is met, select the relatively minimal sequence of modules of a resource consumption.Here introduce
Parameter Res represents to the resource consumption total amount of all raw data packets in some time window, wherein:
In principle, policy making unit always selects the minimum pretreatment strategy of resource consumption total amount to carry out weight to module
Group, but in some cases, mist user can be by the user-defined external interface for reconfiguring layer offer according to practical business
Scene specifies some special demands, and such as pretreatment to certain class content must call some or certain several specific service modules
Deng.Such user-defined event is arranged to limit priority by policy making unit, on the premise of user's request is ensured, as far as possible
Optimize resource to use.
As shown in figure 4, described by the figure be when being recombinated to different number of service module, based on the present invention
Mist node serve mechanism provides the time delay cost of mechanism with other traditional services.
As shown in figure 5, described by the figure be when being recombinated to different number of service module, based on the present invention
The calculation cost of distinct device under mist node serve mechanism.
Experiment shows that the new software in mist node proposed by the present invention is defined in service reconfiguration mechanism and traditional mist node
Service provide mechanism compare, time cost and calculation cost can be largely reduced, so as to reduce mist system entirety energy consumption.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or change within the scope of the claims, this not shadow
Ring the substantive content of the present invention.In the case where not conflicting, the feature in embodiments herein and embodiment can any phase
Mutually combination.
Claims (10)
1. a kind of software definition service reconfiguration method based on perception of content and weighted graph during mist calculates, it is characterised in that described
Service reconfiguration method is operated in the mist node architecture of a stratification, and mist node is made up of three levels, respectively service weight
Structure layer, content driven key-course, it is user-defined reconfigure layer, meanwhile, three kinds of corresponding memory spaces are marked off in mist node,
That is Service Source storehouse, the policy library based on content, demand storehouse, respectively to store-service module, content-operation mapping table, mist
The pretreatment demand of user.
The software definition service reconfiguration method based on perception of content and weighted graph during 2. mist according to claim 1 calculates,
Characterized in that, the mist node is collected from data packet head and in load and extracts important data attribute, so as to complete pair
The classification of content, and it is different then using content tab technology to a large amount of former caused by radio sensing network based on content type
Beginning data are identified, subsequently to perform different pretreatment strategies to the content-data of different labels.
The software definition service reconfiguration method based on perception of content and weighted graph during 3. mist according to claim 1 calculates,
Characterized in that, mist user is reconfigured the interface that layer provides upwards specific bottom pretreatment strategy is entered using user-defined
Row control, call the service module in service scheduling process and be consistent with itself actual demand, meanwhile, mist user is connect by programming
Mouth completes the flexible management to service module, including rewriting, change, deletion, increase to service module, to improve service dynamic
The flexibility of restructuring, ensure that by the layout of modularization resource various pretreatment demand can be met, and then reduce portion again
Affix one's name to energy consumption caused by mist node.
The software definition service reconfiguration method based on perception of content and weighted graph during 4. mist according to claim 1 calculates,
Characterized in that, when selection module is instantiated, a pretreatment strategy is by one or more operation according to certain logic
Order forms, and same operation correspond to one or more service moulds that the operation is realized with different calculation costs and algorithm again
Block, the data of every kind of content type correspond to different pretreatment strategies, that is, correspond to the combination of different service modules, pass through
Initial data after CBL technical labels will search corresponding place in the policy library stored in the key-course of content driven
Reason is tactful and completes the service module that the strategy need to be transferred, and the pretreatment process then strategically pre-defined in storehouse is to institute
Need service module to carry out layout with certain logical order, and transfer to the service reconstruction of layer execution module of mist node to instantiate, join
Number transmits, and is finally exported, the data after processing or valuable result are consigned into upper strata mist user.
The software definition service reconfiguration method based on perception of content and weighted graph during 5. mist according to claim 1 calculates,
Characterized in that, the pretreatment demand of the data, non-upper layer mist user and its time-varying for different content type, it is not necessary to be
The a set of single preprocessor of each strategy deployment configuration, it is only necessary to output and user's practical business based on perception of content
Need, call one or more service modules, when changes in demand, service module is recycled and dynamic reorganization, i.e.,
It can improve the utilization ratio of Service Source while the flexibly demand of reply differentiation, reduce mist node energy consumption.
The software definition service reconfiguration method based on perception of content and weighted graph during 6. mist according to claim 1 calculates,
Characterized in that, service reconstruction of layer includes being used for the service reconfiguration list for calling, instantiate required service module and controlling it to perform
Member, the perception of content unit for carrying out feature extraction and classification to the initial data collected from radio sensing network, use
In each service module is registered, additions and deletions, renewal module management unit, meanwhile, service the service that sets in reconstruction of layer
Resources bank, including the data buffer area temporarily stored for carrying out initial data, and for depositing abstract encapsulation resource
Service module pond, a large amount of initial data being collected into from bottom radio sensing network will temporarily be put after mist node is uploaded to
Put pending in data buffer area etc., the packet into pretreatment process will be transmitted to perception of content unit first, the list
Member extraction and analysis of key attribute from data packet head and in load, export the classification results to content, are then based on different
Content type, initial data caused by radio sensing network is identified using content tab technology, so as to follow-up in
Hold and complete rational service reconstruct.
The software definition service reconfiguration method based on perception of content and weighted graph during 7. mist according to claim 6 calculates,
Characterized in that, after the specific pretreatment strategy and required service module that the key-course for receiving content driven issues, service weight
Structure layer will choose corresponding module from service module pond, and it is instantiated with the parameter specified in default or strategy, and presses
Layout and restructuring are carried out to module according to given logical order, so that it completes specific data processing needs, when from underlying network
Network is collected into different types of packet or when upper strata mist user is updated to data processing policy, and service reconstruction of layer can be existing
On the basis of having service module and layout, it, which is replaced, or recombinated can meet new pretreatment demand.
The software definition service reconfiguration method based on perception of content and weighted graph during 8. mist according to claim 1 calculates,
Characterized in that, the key-course of content driven includes the demand resolution unit for being analyzed user's request and is used for
Carry out data prediction strategy generating and optimization policy making unit, meanwhile, set in the key-course of content driven based on
The policy library of content, including for storing the service logic sequence list of preset data handling process, and for defining in difference
Hold categorical data and its content-operation mapping table of required respective operations.
The software definition service reconfiguration method based on perception of content and weighted graph during 9. mist according to claim 8 calculates,
Characterized in that, policy making unit will be uploaded to the packet of content tab technology identification, by the unit in content-behaviour
Make to search the processing operation needed for it in mapping table, meanwhile, if upper strata mist user has special demand, these demands will also issue
To demand resolution unit, for controlling the execution of specific operation or the selection of module.
The software definition service reconfiguration method based on perception of content and weighted graph during 10. mist according to claim 9 calculates,
Characterized in that, during strategy generating, including two sub-steps, it is the policy development based on weighted graph and plan respectively
Slightly optimize, wherein:Policy development refers to service reconstruction of layer mould according to needed for the service strategy selection that the key-course of content driven issues
Block is initialized and reconfigured, so that it meets the needs of different mist users and contents processing, obtains being possible at certain
The data prediction strategy performed in class content;Policy optimization refers to according to the particular demands that energy consumption and user issue to the pre- place of data
Reason strategy optimizes, and chooses the module restructuring of relatively reasonable strategy instruction service reconstruction of layer.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710597297.8A CN107508698B (en) | 2017-07-20 | 2017-07-20 | Software defined service reorganization method based on content perception and weighted graph in fog calculation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710597297.8A CN107508698B (en) | 2017-07-20 | 2017-07-20 | Software defined service reorganization method based on content perception and weighted graph in fog calculation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107508698A true CN107508698A (en) | 2017-12-22 |
CN107508698B CN107508698B (en) | 2020-07-24 |
Family
ID=60678914
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710597297.8A Active CN107508698B (en) | 2017-07-20 | 2017-07-20 | Software defined service reorganization method based on content perception and weighted graph in fog calculation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107508698B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3407194A2 (en) | 2018-07-19 | 2018-11-28 | Erle Robotics, S.L. | Method for the deployment of distributed fog computing and storage architectures in robotic modular components |
CN110473120A (en) * | 2018-05-10 | 2019-11-19 | 深圳富桂精密工业有限公司 | Micro services isomery regenerative system, method and storage medium based on industry internet |
CN110555138A (en) * | 2019-08-05 | 2019-12-10 | 慧镕电子***工程股份有限公司 | hybrid cloud storage method under cloud computing architecture |
CN111857886A (en) * | 2019-04-26 | 2020-10-30 | 张明明 | Software running method, system, computing equipment and storage medium |
CN114296947A (en) * | 2022-03-09 | 2022-04-08 | 四川大学 | Multi-calculation model management method for complex scene |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103688489A (en) * | 2012-12-03 | 2014-03-26 | 华为技术有限公司 | Method for strategy processing and network equipment |
US20160334276A1 (en) * | 2015-05-12 | 2016-11-17 | BioSensing Systems, LLC | Apparatuses and methods for bio-sensing using unmanned aerial vehicles |
CN106533806A (en) * | 2016-12-26 | 2017-03-22 | 上海交通大学 | Method for providing cross-layer quality of service (QoS) based on application awareness in multi-tenant software defined network (SDN) |
-
2017
- 2017-07-20 CN CN201710597297.8A patent/CN107508698B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103688489A (en) * | 2012-12-03 | 2014-03-26 | 华为技术有限公司 | Method for strategy processing and network equipment |
US20160334276A1 (en) * | 2015-05-12 | 2016-11-17 | BioSensing Systems, LLC | Apparatuses and methods for bio-sensing using unmanned aerial vehicles |
CN106533806A (en) * | 2016-12-26 | 2017-03-22 | 上海交通大学 | Method for providing cross-layer quality of service (QoS) based on application awareness in multi-tenant software defined network (SDN) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110473120A (en) * | 2018-05-10 | 2019-11-19 | 深圳富桂精密工业有限公司 | Micro services isomery regenerative system, method and storage medium based on industry internet |
EP3407194A2 (en) | 2018-07-19 | 2018-11-28 | Erle Robotics, S.L. | Method for the deployment of distributed fog computing and storage architectures in robotic modular components |
CN111857886A (en) * | 2019-04-26 | 2020-10-30 | 张明明 | Software running method, system, computing equipment and storage medium |
CN110555138A (en) * | 2019-08-05 | 2019-12-10 | 慧镕电子***工程股份有限公司 | hybrid cloud storage method under cloud computing architecture |
CN110555138B (en) * | 2019-08-05 | 2022-09-13 | 慧镕电子***工程股份有限公司 | Hybrid cloud storage method under cloud computing architecture |
CN114296947A (en) * | 2022-03-09 | 2022-04-08 | 四川大学 | Multi-calculation model management method for complex scene |
CN114296947B (en) * | 2022-03-09 | 2022-07-08 | 四川大学 | Multi-calculation model management method for complex scene |
Also Published As
Publication number | Publication date |
---|---|
CN107508698B (en) | 2020-07-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107508698A (en) | Software definition service reconfiguration method based on perception of content and weighted graph during mist calculates | |
CN110321999A (en) | Neural computing figure optimization method | |
CN107451653A (en) | Computational methods, device and the readable storage medium storing program for executing of deep neural network | |
CN107016175A (en) | It is applicable the Automation Design method, device and the optimization method of neural network processor | |
CN103825811A (en) | Implementation scheme for accessing sensor protocol self-identification IOT (internet of things) gateway | |
CN103473031B (en) | Collaborative concurrent type frog messaging bus, driving member composition model and component method for splitting | |
CN109636887A (en) | A kind of conversion of Two-dimensional electron technical drawing format and vector quantization interactive system | |
CN106919769A (en) | A kind of hierarchy type FPGA placement-and-routings method based on Hierarchy Method and empowerment hypergraph | |
CN108256636A (en) | A kind of convolutional neural networks algorithm design implementation method based on Heterogeneous Computing | |
CN110100255A (en) | Region is effective, reconfigurable, energy saving, the effective neural network substrate of speed | |
CN107633298A (en) | A kind of hardware structure of the recurrent neural network accelerator based on model compression | |
CN105760272B (en) | Monitoring backstage business customizing method and its system based on plug-in unit | |
CN110321064A (en) | Computing platform realization method and system for neural network | |
CN107704239A (en) | Configuration generating method, device and electronic equipment | |
CN109447276A (en) | A kind of machine learning method, system, equipment and application method | |
CN107122252A (en) | A kind of interactive method between systems and device | |
CN103488697B (en) | System and mobile terminal capable of automatically collecting and exchanging fragmented commercial information | |
CN104050193B (en) | Generate the method for message and realize the data handling system of this method | |
CN108985459A (en) | The method and apparatus of training pattern | |
CN107479866A (en) | The method that open terminal applies data and function are realized based on reconfiguration technique | |
CN115619005A (en) | Intelligent power utilization network resource optimal configuration method and system | |
CN103258255A (en) | Knowledge discovery method applicable to power grid management system | |
CN109376344A (en) | The generation method and terminal device of list | |
CN107885691A (en) | Serial data transmission method, distributed apparatus and main frame | |
CN107911804A (en) | A kind of wireless sensor network data fusion method |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |