CN109523123A - A kind of the intelligent allocation method and server of distributing real time system - Google Patents
A kind of the intelligent allocation method and server of distributing real time system Download PDFInfo
- Publication number
- CN109523123A CN109523123A CN201811191526.7A CN201811191526A CN109523123A CN 109523123 A CN109523123 A CN 109523123A CN 201811191526 A CN201811191526 A CN 201811191526A CN 109523123 A CN109523123 A CN 109523123A
- Authority
- CN
- China
- Prior art keywords
- affairs
- node
- processing load
- transaction
- server
- 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
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 223
- 238000003062 neural network model Methods 0.000 claims description 53
- 238000012549 training Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 19
- 240000002853 Nelumbo nucifera Species 0.000 claims description 9
- 235000006508 Nelumbo nucifera Nutrition 0.000 claims description 9
- 235000006510 Nelumbo pentapetala Nutrition 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 7
- 210000004218 nerve net Anatomy 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims 1
- 230000000916 dilatatory effect Effects 0.000 abstract description 9
- 238000010801 machine learning Methods 0.000 abstract description 3
- 238000012502 risk assessment Methods 0.000 description 22
- 238000004891 communication Methods 0.000 description 11
- 230000008859 change Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 235000013399 edible fruits Nutrition 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000000712 assembly Effects 0.000 description 2
- 238000000429 assembly Methods 0.000 description 2
- 238000012550 audit Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000000638 stimulation Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005401 electroluminescence Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0633—Workflow analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Abstract
Present invention discloses the intelligent allocation methods and server of a kind of distributing real time system, comprising: the affairs for obtaining a certain node flow into information and affairs outflow information;Wherein, it includes flowing into the Transaction Information of all the first affairs of the node that affairs, which flow into information, and affairs outflow information includes flowing out the Transaction Information of all the second affairs of the node;Information is flowed into according to affairs and affairs flow out information, determines the present processing load of the node;Judge whether present processing load is greater than processing load threshold value;If so, the Transaction Information of at least one target transaction on node is gone to replacement node;Wherein, which is at least one affairs in all affairs removed except the second affairs in the first affairs on the node.Under the method, the processing load of node can be predicted by machine learning model, realizes the rational allocation to magnanimity affairs, and then improve tediously long dilatory issued transaction status.
Description
Technical field
The present invention relates to field of computer technology, in particular to the intelligent allocation method and clothes of a kind of distributing real time system
Business device.
Background technique
Current some issued transactions (such as insurance business processing) need the node by much handling, and each node can be with
Regard a server or system as, each node can form a process chain according to the circulation sequence of issued transaction, whole
A transaction data just circulates among the nodes.
By taking insurance business is handled as an example, insurance business processing needs to be accepted department, government department, legal department and wind
Danger assessment company's constant pitch point carries out the circulation of transaction data, i.e. insurance business may be needed in processing by accepting department
The primary condition of applicant corresponding for the insurance business is audited, and accepts department in the primary condition of audit the applicant
Shi Ruguo, which leaves a question open the insurance business can circulate to relevant government department, does further audit, and government department is judging
Applicant's condition of the insurance business there are in the case where certain risk, which can be circulated to risk assessment company into
Row risk assessment;Or it can be by the insurance business when the applicant's condition for judging the insurance business needs to carry out law assessment
It circulates to legal department, relevant law assessment is carried out to it by legal department.
However, usually will appear following situations when insurance business circulates between different nodes: the number of a large amount of insurance business
Some node (such as server of risk assessment company) is flowed into according in the same period, so that the processing load of the node increases suddenly
Greatly, block due to a large amount of insurance business data are timely handled because being unable to get in the node, the process flow of entire insurance business
It is tediously long dilatory.
Summary of the invention
In order to solve to block due to a large amount of insurance business data present in the relevant technologies are timely handled because being unable to get in
A certain node, the tediously long dilatory technical problem of the process flow of entire insurance business, the present invention provides a kind of distributed transactions
The intelligent allocation method and server of processing.
A kind of intelligent allocation method of distributing real time system, which comprises
The affairs that server obtains a certain node flow into the affairs outflow information of information and the node;Wherein, the thing
It includes flowing into the Transaction Information of all the first affairs of the node that business, which flows into information, and the affairs outflow information includes outflow institute
State the Transaction Information of all the second affairs of node;
The server flows into information according to the affairs and the affairs flow out information, determines the current of the node
Handle load;
The server judges whether the present processing load is greater than processing load threshold value;
If the present processing load be greater than the processing load threshold value, the server by the node at least
The Transaction Information of one target transaction produce to replacement node, the target transaction be the node on the first affairs in remove
At least one affairs in all affairs except second affairs.
As an alternative embodiment, the affairs flow into information and also wrap in first aspect of the embodiment of the present invention
It includes affairs and flows into log, the affairs flow into the Transaction Identifier number that log carries first affairs, the affairs outflow letter
Breath further includes affairs outflow log, and the affairs outflow log carries the Transaction Identifier number of second affairs, the service
Device flows into information according to the affairs and the affairs flow out information, determines the present processing load of the node, comprising:
Server Transaction Identifier number according to entrained by affairs inflow log and the affairs flow out log
Entrained Transaction Identifier number determines that the Current transaction of the node, the Current transaction are the first thing on the node
All affairs except second affairs are removed in business;
The server is using the Transaction Information of the Current transaction as the present processing load of the node.
As an alternative embodiment, the affairs flow into log and also take in first aspect of the embodiment of the present invention
Transaction types with first affairs, the affairs outflow log also carry the transaction types of second affairs, such as
Present processing load described in fruit be greater than the processing load threshold value, the server by the node at least one of target thing
The Transaction Information of business is produced to replacement node, comprising:
The server determines at least one target transaction from the Current transaction;
The server inquires preset transaction types and the node table of comparisons, to find out transaction types and the target thing
The identical replacement node of the transaction types of business;
The server produces the Transaction Information of at least one of described target transaction to the replacement node, so that described
Replace the processing for the Transaction Information that node is completed at least one of described target transaction.
As an alternative embodiment, in first aspect of the embodiment of the present invention, the server is by the node
On at least one of target transaction Transaction Information produce to replacement node after, the method also includes:
The server obtains the flow direction of the target transaction;
The server updates the flow direction record of the target transaction, the flow direction according to the flow direction of the target transaction
Record is for tracking the circulation path for counting the target transaction.
As an alternative embodiment, in first aspect of the embodiment of the present invention, the method also includes:
If the present processing load is not more than the processing load threshold value, the server will be described currently processed negative
Lotus is input to processing load forecasting model;
The server is born based on the output of the processing load forecasting model as a result, determining that the prediction of the node is handled
Lotus;
The server judges whether the prediction processing load is greater than the processing load threshold value;
If prediction processing load is greater than the processing load threshold value, the server executes described by the section
The Transaction Information of at least one target transaction on point is produced to replacement node.
As an alternative embodiment, in first aspect of the embodiment of the present invention, if the present processing load
No more than the processing load threshold value, the server by the present processing load be input to processing load forecasting model it
Before, the method also includes:
The server obtains the history processing load of the node;
History described in the server by utilizing handles the initial neural network model of weight training, and it is pre- to obtain the processing load
Survey model.
As an alternative embodiment, being gone through described in the server by utilizing in first aspect of the embodiment of the present invention
History handles the initial neural network model of weight training, obtains the processing load forecasting model, comprising:
The server obtains N number of continuous processing load as sample from history processing load and inputs initial mind
Through network model, so that the initial neural network model exports output load corresponding with the sample;The N is positive whole
Number;
The output load is determined as target processing load by the server;
Target processing load is compared by the server with actual treatment load, updates institute according to comparison result
The parameter of initial neural network model is stated, the actual treatment load is that the server is getting the target processing load
At the time of from the nodal test to processing load;
The server judges whether the loss function of the initial neural network model meets preset condition, the loss
Function is used to characterize the error between the output load and actual value of the initial neural network model;
If the loss function of the initial neural network model meets the preset condition, the server will be described first
The parameter current of beginning network model is determined as handling the parameter of load forecasting model, and obtains the processing according to the parameter and bear
Lotus prediction model.
A kind of server, the server include:
Module is obtained, the affairs for obtaining a certain node flow into information and the affairs of the node flow out information;Wherein,
It includes flowing into the Transaction Information of all the first affairs of the node that the affairs, which flow into information, and the affairs outflow information includes
Flow out the Transaction Information of all the second affairs of the node;
First determining module determines the section for flowing into information and affairs outflow information according to the affairs
The present processing load of point;
Judgment module, for judging whether the present processing load is greater than processing load threshold value;
Shift module, for judging that the present processing load is greater than the processing load threshold value in the judgment module
When, the Transaction Information of at least one target transaction on the node is produced to replacement node, the target transaction is described
At least one affairs in all affairs except second affairs are removed in the first affairs on node.
The third aspect of the embodiment of the present invention discloses a kind of computer readable storage medium, stores computer program, wherein
The computer program makes computer execute a kind of intelligence of distributing real time system disclosed in first aspect of the embodiment of the present invention
It can concocting method.
Fourth aspect of the embodiment of the present invention discloses a kind of electronic equipment, and the electronic equipment includes:
Processor;
Memory is stored with computer-readable instruction on the memory, and the computer-readable instruction is by the processing
When device executes, foregoing method is realized.
The technical solution that the embodiment of the present invention provides can include the following benefits:
The intelligent allocation method of distributing real time system provided by the present invention includes the following steps, obtains a certain node
Affairs flow into information and the affairs of the node flow out information;Wherein, it includes flowing into all the first things of node that affairs, which flow into information,
The Transaction Information of business, affairs outflow information include flowing out the Transaction Information of all the second affairs of node;It is flowed into and is believed according to affairs
Breath and affairs flow out information, determine the present processing load of egress;Judge whether present processing load is greater than processing load threshold
Value;If present processing load is greater than processing load threshold value, the Transaction Information of at least one target transaction on node is produced
To replacement node, which is to remove in all affairs except the second affairs at least in the first affairs on the node
One affairs.
Under the method, real-time monitoring is carried out to the processing load of each node, when the Transaction Information for monitoring a large amount of affairs
Block when some node, the Transaction Information that will exceed the target transaction of the processing load of the node is produced to replacement node, is made
The processing load obtained between node obtains reasonable equilibrium, and then improves tediously long dilatory issued transaction status.To sum up, distributed
The intelligent allocation of issued transaction is achieved.
It should be understood that the above general description and the following detailed description are merely exemplary, this can not be limited
Invention.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention
Example, and in specification together principle for explaining the present invention.
Fig. 1 is a kind of schematic diagram of device shown according to an exemplary embodiment;
Fig. 2 is a kind of process of the intelligent allocation method of distributing real time system shown according to an exemplary embodiment
Figure;
Fig. 3 is a kind of process of the intelligent allocation method of the distributing real time system shown according to another exemplary embodiment
Figure;
Fig. 4 is a kind of block diagram of server shown according to an exemplary embodiment;
Fig. 5 is the block diagram of another server shown according to an exemplary embodiment;
Fig. 6 is the block diagram of another server shown according to an exemplary embodiment.
Specific embodiment
Here will the description is performed on the exemplary embodiment in detail, the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistented with the present invention.On the contrary, they be only with it is such as appended
The example of device and method being described in detail in claims, some aspects of the invention are consistent.
Fig. 1 is a kind of schematic diagram of device shown according to an exemplary embodiment.Device 100 can be electronic equipment.
As shown in Figure 1, device 100 may include following one or more components: processing component 102, memory 104, power supply module
106, multimedia component 108, audio component 110, sensor module 114 and communication component 116.
The integrated operation of the usual control device 100 of processing component 102, such as with display, telephone call, data communication, phase
Machine operation and the associated operation of record operation etc..Processing component 102 may include one or more processors 118 to execute
Instruction, to complete all or part of the steps of following methods.In addition, processing component 102 may include one or more modules,
For convenient for the interaction between processing component 102 and other assemblies.For example, processing component 102 may include multi-media module, use
In to facilitate the interaction between multimedia component 108 and processing component 102.
Memory 104 is configured as storing various types of data to support the operation in device 100.These data are shown
Example includes the instruction of any application or method for operating on the device 100.Memory 104 can be by any kind of
Volatibility or non-volatile memory device or their combination are realized, such as static random access memory (Static Random
Access Memory, abbreviation SRAM), electrically erasable programmable read-only memory (Electrically Erasable
Programmable Read-Only Memory, abbreviation EEPROM), Erasable Programmable Read Only Memory EPROM (Erasable
Programmable Read Only Memory, abbreviation EPROM), programmable read only memory (Programmable Red-
Only Memory, abbreviation PROM), read-only memory (Read-Only Memory, abbreviation ROM), magnetic memory, flash
Device, disk or CD.It is also stored with one or more modules in memory 104, is configured to for the one or more module
It is executed by the one or more processors 118, to complete all or part of step in method as follows.
Power supply module 106 provides electric power for the various assemblies of device 100.Power supply module 106 may include power management system
System, one or more power supplys and other with for device 100 generate, manage, and distribute the associated component of electric power.
Multimedia component 108 includes the screen of one output interface of offer between described device 100 and user.One
In a little embodiments, screen may include liquid crystal display (Liquid Crystal Display, abbreviation LCD) and touch panel.
If screen includes touch panel, screen may be implemented as touch screen, to receive input signal from the user.Touch panel
Including one or more touch sensors to sense the gesture on touch, slide, and touch panel.The touch sensor can be with
The boundary of a touch or slide action is not only sensed, but also detects duration associated with the touch or slide operation and pressure
Power.Screen can also include display of organic electroluminescence (Organic Light Emitting Display, abbreviation OLED).
Audio component 110 is configured as output and/or input audio signal.For example, audio component 110 includes a Mike
Wind (Microphone, abbreviation MIC), when device 100 is in operation mode, such as call model, logging mode and speech recognition mould
When formula, microphone is configured as receiving external audio signal.The received audio signal can be further stored in memory
104 or via communication component 116 send.In some embodiments, audio component 110 further includes a loudspeaker, for exporting
Audio signal.
Sensor module 114 includes one or more sensors, and the state for providing various aspects for device 100 is commented
Estimate.For example, sensor module 114 can detecte the state that opens/closes of device 100, the relative positioning of component, sensor group
Part 114 can be with the position change of 100 1 components of detection device 100 or device and the temperature change of device 100.Some
In embodiment, which can also include Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 116 is configured to facilitate the communication of wired or wireless way between device 100 and other equipment.Device
100 can access the wireless network based on communication standard, such as WiFi (Wireless-Fidelity, Wireless Fidelity).Show at one
In example property embodiment, communication component 116 receives broadcast singal or broadcast from external broadcasting management system via broadcast channel
Relevant information.In one exemplary embodiment, the communication component 116 further includes near-field communication (Near Field
Communication, abbreviation NFC) module, for promote short range communication.For example, radio frequency identification can be based in NFC module
(Radio Frequency Identification, abbreviation RFID) technology, Infrared Data Association (Infrared Data
Association, abbreviation IrDA) technology, ultra wide band (Ultra Wideband, abbreviation UWB) technology, Bluetooth technology and other skills
Art is realized.
In the exemplary embodiment, device 100 can be by one or more application specific integrated circuit (Application
Specific Integrated Circuit, abbreviation ASIC), it is digital signal processor, digital signal processing appts, programmable
Logical device, field programmable gate array, controller, microcontroller, microprocessor or other electronic components are realized, for executing
Following methods.
Fig. 2 is a kind of process of the intelligent allocation method of distributing real time system shown according to an exemplary embodiment
Figure.As shown in Fig. 2, the method includes the following steps.
Step 210, the affairs that server obtains a certain node flow into information and the affairs outflow information of the node.
Wherein, it may include flowing into the Transaction Information of all the first affairs of the node, the affairs which, which flows into information,
Outflow information may include flowing out the Transaction Information of all the second affairs of the node;Optionally, the Transaction Information of affairs can be with
The Transaction Identifier number (ID, IDentity) of the affairs is carried, i.e. the Transaction Information of the first affairs carries the thing of the first affairs
Business identification number, the second affairs Transaction Information carry the Transaction Identifier number of the second affairs.
Step 220, server flows into information according to affairs and affairs flow out information, determines the currently processed negative of the node
Lotus.
Wherein, it can also include that affairs flow into log that above-mentioned affairs, which flow into information, which flows into log and carry first
The Transaction Identifier number of affairs, affairs outflow information further include affairs outflow log, and affairs outflow log carries the second thing
The Transaction Identifier number of business.So, in an exemplary embodiment, server flows into information and above-mentioned affairs according to above-mentioned affairs
Information is flowed out, determines that the present processing load of above-mentioned node may include:
Affairs entrained by server Transaction Identifier number according to entrained by affairs inflow log and affairs outflow log
Identification number calculates the quantity for the Transaction Information of the first affairs for flowing into the node and flows out the number of transactions of the second affairs of the node
According to quantity between difference, present processing load as the node.
Step 230, server judges whether present processing load is greater than processing load threshold value, if present processing load is big
In processing load threshold value, triggering executes step 240;If present processing load terminates this process no more than processing load threshold value.
In the exemplary embodiment, optionally, server can receive the maximum processing load that each node reports, and according to
The corresponding processing load threshold value of the node is arranged in the maximum processing load that each node is reported, and can be improved allotment issued transaction
The flexibility of work.
Step 240, server produces the Transaction Information of at least one target transaction on the node to replacement node.
Wherein, which is in all affairs removed except the second affairs in the first affairs on above-mentioned node
At least one affairs.
In the present exemplary embodiment, for a certain node, the first affairs can indicate that historical statistics flows into so far and be somebody's turn to do
The affairs of node, the second affairs can indicate that the affairs of the node are flowed out in historical statistics so far, then, which can be
Flow at least one affairs removed in all affairs except the second affairs in the first affairs of the node;It needs to illustrate
It is that above-mentioned historical statistics is used to indicate target time section so far, which is flowed into information with the affairs of the node and wrapped
It is initial time at the time of including the Transaction Information of the first affairs, obtains the affairs inflow of the node with server in the present embodiment
It is cut-off time at the time of information and affairs outflow information.For example, the first affairs total quantity is 50, the second affairs total quantity
It is 20, then, the quantity of target transaction can be 1,2 or 10, and the quantity of target transaction is no more than 30.
In one exemplary embodiment, optionally, for a certain node, the first affairs also may indicate that current time
Flow into the affairs of the node, the second affairs can be expressed as the affairs for flowing out the node at current time, and history has affairs can be with
The affairs that the moment in this prior has been positioned at the node are expressed as, then target transaction can be the whole of the first affairs and the node
History has at least one affairs removed in all affairs except the second affairs in affairs;It should be noted that this is current
At the time of moment is that the affairs that server obtains the node in the present embodiment flow into information and affairs outflow information.For example,
The total quantity of first affairs is 15, and the total quantity of the second affairs is 20, and the total quantity that history has affairs is 35, then target thing
The quantity of business can be 1,2 or 10, and the quantity of the target transaction is no more than 30.
In the embodiment of another exemplary, server turns the Transaction Information of at least one target transaction on the node
May include: to replacement node out
Server determines the Current transaction of the node, and Current transaction is to remove the second thing in the first affairs on the node
All affairs except business;
Server determines at least one target transaction from Current transaction;
Server inquires preset transaction types and the node table of comparisons, to find out the affairs of transaction types and target transaction
The identical replacement node of type;
Server produces the Transaction Information of at least one target transaction to replacement node, so that replacement node completion pair
The processing of the Transaction Information of at least one target transaction.
In the present example embodiment, transaction types replacement node identical with the transaction types of target transaction refers to that this is replaced
It changes node and is capable of handling affairs identical with the transaction types of target transaction.Server can be by inquiring preset transaction types
With the node table of comparisons, transaction types multiple replacement nodes identical with the transaction types of target transaction are found out, are replaced according to multiple
The affairs for changing each replacement node in node flow into information and affairs outflow information, determine the currently processed of each replacement node
Load, and the smallest target replacement node of present processing load in multiple replacement nodes is selected, by least one of above-mentioned node
The Transaction Information of target transaction goes to target replacement node, so that target replacement node completes the number of transactions to target transaction
According to processing.For example, target transaction can be risk assessment, and risk assessment can be handled by risk assessment company,
The total quantity of risk assessment company is 20;The type of risk assessment can be the risk assessment of traffic Claims Resolution, be also possible to disease
The risk assessment that pathology is paid for, if the transaction types of target transaction are the risk assessment of traffic Claims Resolution, then, it above-mentioned node and replaces
Changing node is the risk assessment company for being capable of handling the risk assessment of traffic Claims Resolution;Then, server is preset by inquiry
Transaction types and the node table of comparisons, determine to be capable of handling the risk assessment of traffic Claims Resolution in 20 risk assessment companies
12 risk assessment companies determine present processing load the smallest 1 target risk assessment company in 12 risk assessment companies,
The data of the risk assessment of traffic Claims Resolution on a certain risk assessment company (node) are transferred to target risk assessment company,
So that target risk assessment company handles it.
As it can be seen that implement the intelligent allocation method of distributing real time system described in Fig. 2, it can be to the processing of each node
Load carry out real-time monitoring, when monitor a large amount of affairs Transaction Information block when some node, will exceed the place of the node
The Transaction Information for managing the target transaction of load is produced to replacement node identical with the node type, so that the place of same type node
Reason load obtains reasonable equilibrium, and then improves tediously long dilatory issued transaction status.
Fig. 3 is a kind of process of the intelligent allocation method of the distributing real time system shown according to another exemplary embodiment
Figure.As shown in figure 3, in addition to the step shown in Fig. 2, step 230 judge present processing load no more than processing load threshold value it
Afterwards, display control method is further comprising the steps of in this embodiment.
Step 250, present processing load is input to processing load forecasting model by server.
In one exemplary embodiment, before step 250, method shown in Fig. 4 can with the following steps are included:
Server obtains the history processing load of the node;
Server by utilizing history handles the initial neural network model of weight training, obtains processing load forecasting model.
Further alternative, server by utilizing history handles the initial neural network model of weight training, obtains processing load
Prediction model may include:
Server obtains N number of continuous processing load as sample from history processing load and inputs initial neural network mould
Type, so that initial neural network model exports output load corresponding with sample;N is positive integer;
Output load is determined as target processing load by server;
Target processing load is compared by server with actual treatment load, updates initial nerve net according to comparison result
The parameter of network model, the actual treatment load be server at the time of getting target processing load from the nodal test to
Handle load;
Server judges whether the loss function of initial neural network model meets preset condition, and loss function is for characterizing
Error between the output load and actual value of initial neural network model;
If the loss function of initial neural network model meets preset condition, server is current by initial network model
Parameter is determined as handling the parameter of load forecasting model, and obtains processing load forecasting model according to parameter.
In the present exemplary embodiment, processing load forecasting model can be used for predicting each node at the place of next period
Manage load (prediction processing load), wherein the period of prediction can determine according to the training content of processing load model.Citing comes
It says, the duration of each period can be set as 1 hour by server, using every 1 hour intermediate time as the regulation moment.For appoint
One node obtains the history processing load of any node, arbitrarily chooses N number of continuous regulation moment from history processing load
(such as 5 points 30 minutes, 6 points 30 minutes and 7 points 30 minutes) corresponding to history handle load, N is positive integer;By the N
History corresponding to a regulation moment handles load as sample and is input to initial neural network model (machine learning model), i.e.,
Input vector is x=(x1, x2, x3, x4 ... xn), and the N+1 regulation moment institute after N number of continuous regulation moment is right
The processing load answered is as desired output;It is further continued for executing above-mentioned operation, i.e., is arbitrarily selected from above-mentioned history processing load
History processing load corresponding to other N number of continuous regulation moment is taken as sample and is input to the initial neural network mould
Type so that the initial neural network model export it is pre- corresponding to the N+1 regulation moment after N number of continuous regulation moment
Processing load (desired output) is surveyed, until the loss function of initial neural network model meets preset condition.Wherein, initial nerve
Network model includes input layer, hidden layer and output layer, and server is a in the N+1 for getting initial neural network model output
After providing moment corresponding prediction processing load (desired output), it can also be obtained from any node and is advised at the N+1
The actual treatment load (reality output) that timing is carved, and initial nerve net is determined according to prediction processing load and actual treatment load
The output error of network model, which is the error of the output layer of initial neural network model, by the error of the output layer
The error that can determine the direct preceding conducting shell of output layer determines the mistake of more preceding layer in the error according to the preceding conducting shell of the output layer
Difference so reversely successively determine from output layer to input layer, can determine all layers in the initial neural network model
Error, and determine according to all layers in the initial neural network model of error the loss letter of the initial neural network model
Number.Since input sample repeatedly can be such that the initial neural network model constantly learns to initial neural network model, and learn
Mode be constantly to change the network connection weight of initial neural network model under the stimulation of extraneous input sample so that initially
The loss function of neural network model meets preset condition.It, can when the loss function of initial network model meets preset condition
To determine that the initial network model training is completed, the parameter current of initial network model is determined as handling load prediction by server
The parameter of model, and processing load forecasting model is obtained according to parameter.
Step 260, server is born based on the output of processing load forecasting model as a result, determining that the prediction of the node is handled
Lotus.
Step 270, server judges whether prediction processing load is greater than processing load threshold value, if prediction processing load is big
In processing load threshold value, triggering executes step 240;If prediction processing load terminates this process no more than processing load threshold value.
Processing may be instantly available since the Transaction Information of affairs flows into after node, it is also possible to will not be instantly available place
Therefore reason predicts that the prediction of the node handles load based on above-mentioned processing load forecasting model, can know the node in advance
The situation of change for handling load is conducive to carry out rational allocation to the Transaction Information of node, optimizes the balanced effect of processing load
Fruit;In addition, processing moment for being handled after flowing into the node of the Transaction Information of affairs can be current according to the node
Sequence of the Transaction Information of processing speed and the affairs in the node, wherein the sequence is flowed by the Transaction Information of affairs
The sequencing of the node.
In a further exemplary embodiment, after step 240, display control method further includes following step in this embodiment
Suddenly.
Step 280, server obtains the flow direction of target transaction.
Step 290, server updates the flow direction record of target transaction according to the flow direction of target transaction, and flow direction record is used
In the circulation path of tracking statistics target transaction.
As it can be seen that the intelligent allocation method of distributing real time system described in implementing Fig. 3, it can be to the processing of each node
Load carry out real-time monitoring, when monitor a large amount of affairs Transaction Information block when some node, will exceed the place of the node
The Transaction Information for managing the target transaction of load is produced to replacement node identical with the node type, so that the place of same type node
Reason load obtains reasonable equilibrium, and then improves tediously long dilatory issued transaction status;In addition, being based on above-mentioned processing load prediction
The prediction of the model prediction node handles load, can know the situation of change of the processing load of the node in advance, is conducive to pair
The Transaction Information of node carries out rational allocation, optimizes the portfolio effect of processing load.
It is the device of the invention embodiment below.
Fig. 4 is a kind of block diagram of server shown according to an exemplary embodiment.As shown in figure 4, the server includes:
Module 310 is obtained, the affairs for obtaining a certain node flow into information and the affairs of the node flow out information, and will
The affairs flow into information and affairs outflow information is supplied to the first determining module 320.
Wherein, it includes flowing into the Transaction Information of all the first affairs of the node, affairs outflow which, which flows into information,
Information includes flowing out the Transaction Information of all the second affairs of the node.
Optionally, the Transaction Information of affairs can carry the Transaction Identifier number (ID, IDentity) of the affairs, i.e., and first
The Transaction Information of affairs carries the Transaction Identifier number of the first affairs, the Transaction Information of the second affairs carries the things of the second affairs
Business identification number.
First determining module 320 determines the current of the node for flowing into information and affairs outflow information according to affairs
Load is handled, and the present processing load is supplied to judgment module 330.
Judgment module 330 is provided for judging whether present processing load is greater than processing load threshold value, and by judging result
To shift module 340.
In the exemplary embodiment, optionally, judgment module 330 is judging whether present processing load is greater than processing load
Before threshold value, the maximum processing load that each node reports can also be received, and negative according to the maximum processing that each node is reported
The corresponding processing load threshold value of the node is arranged in lotus, so that the allotment of issued transaction is more flexible.
Shift module 340, for inciting somebody to action when judgment module 330 judges that present processing load is greater than processing load threshold value
The Transaction Information of at least one target transaction on the node is produced to replacement node;Wherein, which is on the node
The first affairs in remove in all affairs except the second affairs at least one of affairs.
In the present exemplary embodiment, for a certain node, the first affairs can indicate that historical statistics flows into so far and be somebody's turn to do
The affairs of node, the second affairs can indicate that the affairs of the node are flowed out in historical statistics so far, then, which can be
Flow at least one affairs removed in all affairs except the second affairs in the first affairs of the node;It needs to illustrate
It is that above-mentioned historical statistics is used to indicate target time section so far, which is flowed into information with the affairs of the node and wrapped
It is initial time at the time of including the Transaction Information of the first affairs, obtains the affairs inflow of the node with server in the present embodiment
It is cut-off time at the time of information and affairs outflow information.For example, the first affairs total quantity is 50, the second affairs total quantity
It is 20, then, the quantity of target transaction can be 1,2 or 10, and the quantity of target transaction is no more than 30.
In one exemplary embodiment, optionally, for a certain node, the first affairs also may indicate that current time
Flow into the affairs of the node, the second affairs can be expressed as the affairs for flowing out the node at current time, and history has affairs can be with
The affairs that the moment in this prior has been positioned at the node are expressed as, then target transaction can be the whole of the first affairs and the node
History has at least one affairs removed in all affairs except the second affairs in affairs;It should be noted that this is current
At the time of moment is that the affairs that server obtains the node in the present embodiment flow into information and affairs outflow information.For example,
The total quantity of first affairs is 15, and the total quantity of the second affairs is 20, and the total quantity that history has affairs is 35, then target thing
The quantity of business can be 1,2 or 10, and the quantity of the target transaction is no more than 30.
As it can be seen that implement server described in Fig. 4, can processing load to each node carry out real-time monitoring, work as prison
The Transaction Information for measuring a large amount of affairs blocks when some node, will exceed the affairs of the target transaction of the processing load of the node
Data are produced to replacement node, so that the processing load of node obtains reasonable equilibrium, and then improve tediously long dilatory office
Manage status.
Referring to Fig. 5, the block diagram of Fig. 5 another server shown according to an exemplary embodiment, wherein shown in Fig. 5
Server be that server further progress as shown in Figure 4 optimizes.Compared with server shown in Fig. 4, Fig. 5 institute
The server shown further include:
Input module 350, for judgment module 330 judge present processing load no more than processing load threshold value when,
Present processing load is input to processing load forecasting model, and triggers the starting of the second determining module 360.
Wherein, judgment module 330 can will judge after judging whether present processing load is greater than processing load threshold value
As a result above-mentioned shift module 340 and the input module 350 are supplied to.
Second determining module 360, for the output based on processing load forecasting model as a result, determining at the prediction of the node
Load is managed, and prediction processing load is supplied to judgment module 330.
Above-mentioned judgment module 330, is also used to judge whether prediction processing load is greater than processing load threshold value, and judgement is tied
Fruit is supplied to shift module 340.
Above-mentioned shift module 340 is also used to judge that prediction processing load is greater than processing load threshold value in judgment module 330
When, the Transaction Information of at least one target transaction on the node is produced to replacement node identical with the node type.
In one exemplary embodiment, as shown in figure 5, above-mentioned acquisition module 310, can be also used for obtaining going through for the node
History handles load, and history processing load is supplied to training module 370.
Training module 370 obtains processing load prediction for handling the initial neural network model of weight training using history
Model.
Above-mentioned input module 350 is specifically used for handling the initial neural network of weight training using history in training module 370
Model, obtain processing load forecasting model after and judgment module 330 judge present processing load no more than processing load
When threshold value, present processing load is input to processing load forecasting model.
Further alternative, above-mentioned training module 370 handles the initial neural network model of weight training using history, obtains
The mode of processing load forecasting model is specifically as follows:
N number of continuous processing load, which is obtained, as sample from history processing load inputs initial neural network model, with
Initial neural network model is set to export output load corresponding with sample;N is positive integer;
Output load is determined as target processing load;
Target processing load is compared with actual treatment load, initial neural network model is updated according to comparison result
Parameter, the actual treatment load be server at the time of getting target processing load from the nodal test to processing it is negative
Lotus;
Judge whether the loss function of initial neural network model meets preset condition, the loss function is initial for characterizing
Error between the output load and actual value of neural network model;
If the loss function of initial neural network model meets preset condition, and the parameter current of initial network model is true
It is set to the parameter of processing load forecasting model, and processing load forecasting model is obtained according to parameter.
In the present exemplary embodiment, processing load forecasting model can be used for predicting each node at the place of next period
Manage load (prediction processing load), wherein the period of prediction can determine according to the training content of processing load model.Citing comes
It says, the duration of each period can be set as 1 hour by training module 370, using every 1 hour intermediate time as the regulation moment.
For any node, the history processing load of any node is obtained, arbitrarily chooses N number of continuous regulation from history processing load
Moment (such as 5 points 30 minutes, 6 points 30 minutes and 7 points 30 minutes) corresponding to history handle load, N is positive integer;
Initial neural network model (machine learning mould is input to using the processing load of history corresponding to N number of regulation moment as sample
Type), i.e., input vector be x=(x1, x2, x3, x4 ... xn), by it is N number of it is continuous regulation the moment after the N+1 regulation when
Corresponding processing load is carved as desired output;It is further continued for executing above-mentioned operation, i.e., from above-mentioned history processing load
History processing load corresponding to any selection other N number of continuous regulation moment is input to the initial nerve net as sample
Network model, so that the initial neural network model exported corresponding to the N+1 regulation moment after N number of continuous regulation moment
Prediction handle load (desired output), until the loss function of initial neural network model meets preset condition.Wherein, initially
Neural network model includes input layer, hidden layer and output layer, and training module 370 is getting initial neural network model output
The N+1 regulation moment corresponding prediction processing load (desired output) after, can also be obtained from any node its
The actual treatment load (reality output) at the N+1 regulation moment, and determined according to prediction processing load and actual treatment load
The output error of initial neural network model, which is the error of the output layer of initial neural network model, defeated by this
The error of layer can determine the error of the direct preceding conducting shell of output layer out, determine more in the error of the preceding conducting shell according to the output layer
The error of preceding layer so carries out reversed successively calculating from output layer to input layer, can determine the initial neural network mould
All layers of error in type, and the initial neural network model is determined according to all layers in the initial neural network model of error
Loss function.Since input sample repeatedly can be such that the initial neural network model constantly learns to initial neural network model
The mode practised, and learnt is constantly to change the network connection power of initial neural network model under the stimulation of extraneous input sample
Weight, so that the loss function of initial neural network model meets preset condition.When the loss function of initial network model meets in advance
If when condition, training module 370 can determine that the initial network model training is completed, and by the parameter current of initial network model
It is determined as handling the parameter of load forecasting model, and processing load forecasting model is obtained according to parameter.
As it can be seen that implement server described in Fig. 5, can processing load to each node carry out real-time monitoring, work as prison
The Transaction Information for measuring a large amount of affairs blocks when some node, will exceed the affairs of the target transaction of the processing load of the node
Data are produced to replacement node identical with the node type, so that the processing load of same type node obtains reasonable equilibrium, into
And improve tediously long dilatory issued transaction status;In addition, predicting the prediction of the node based on above-mentioned processing load forecasting model
Load is handled, the situation of change of the processing load of the node can be known in advance, be conducive to close the Transaction Information of node
Reason allotment, optimizes the portfolio effect of processing load.
Referring to Fig. 6, the block diagram of Fig. 6 another server shown according to an exemplary embodiment, wherein shown in Fig. 6
Server be that server further progress as shown in Figure 5 optimizes.Compared with server shown in fig. 5, in Fig. 6
Shown in server:
Above-mentioned first determining module 320 flows into information according to affairs and affairs flow out information, determines the current of the node
The mode of processing load is specifically as follows:
Affairs entrained by server Transaction Identifier number according to entrained by affairs inflow log and affairs outflow log
Identification number determines the Current transaction of the node;Wherein, which is to remove the second thing in the first affairs on the node
All affairs except business;Using the Transaction Information of Current transaction as the present processing load of the node.
Wherein, it further includes that affairs flow into log that affairs, which flow into information, and affairs flow into the affairs that log carries the first affairs
Identification number, affairs outflow information further include affairs outflow log, and affairs outflow log carries the Transaction Identifier number of the second affairs.
In one exemplary embodiment, as shown in fig. 6, affairs, which flow into log, can also carry the transactions classes of the first affairs
Type, affairs outflow log also carry the transaction types of the second affairs;Above-mentioned shift module 340 is by least one on the node
The Transaction Information of item target transaction, which is produced to the mode of replacement node, to be specifically as follows:
It determines the Current transaction of the node, and determines at least one target transaction from Current transaction, wherein really
The Current transaction of the node is made, the Current transaction is busy except the second affairs to remove in the first affairs on the node
Business;
Preset transaction types and the node table of comparisons are inquired, to find out transaction types phase of the transaction types with target transaction
Same replacement node;
The Transaction Information of at least one target transaction is produced to replacement node, so that replacement node is completed at least one
The processing of the Transaction Information of target transaction.
In a further exemplary embodiment, as shown in fig. 6, above-mentioned acquisition module 310, being also used to will in shift module 340
The Transaction Information of at least one target transaction on the node is produced to replacement node, obtains the flow direction of target transaction, and
The flow direction of the target transaction is supplied to update module 380.
Update module 380 updates the flow direction record of target transaction, flow direction record for the flow direction according to target transaction
For tracking the circulation path of statistics target transaction.
Wherein, server can be recorded according to the flow direction that the flow direction of affairs updates the target transaction, can be realized to target
The circulation path that the Transaction Information of affairs flows to node carries out global monitoring;In addition, server can also receive user institute
Equipment issue inquiry request after, according to the target transaction updated at the time of receiving the inquiry request
Flow direction records the inquiry request for the user that makes an immediate response, i.e., the flow direction record of the target transaction, energy are sent to the equipment where user
Response speed is enough improved, and then improves the interactive experience of user.
As it can be seen that implement server described in Fig. 6, can processing load to each node carry out real-time monitoring, work as prison
The Transaction Information for measuring a large amount of affairs blocks when some node, will exceed the affairs of the target transaction of the processing load of the node
Data are produced to replacement node identical with the node type, so that the processing load of same type node obtains reasonable equilibrium, into
And improve tediously long dilatory issued transaction status;And the situation of change of the processing load of the node can be known in advance, have
Rational allocation is carried out conducive to the Transaction Information to node, optimizes the portfolio effect of processing load;Furthermore it is possible to improve response speed
Degree, and then improve the interactive experience of user.
The present invention also provides a kind of electronic equipment, which includes:
Processor;
Memory is stored with computer-readable instruction on the memory, when which is executed by processor,
Realize the intelligent allocation method of distributing real time system as previously shown.
The electronic equipment can be Fig. 1 shown device 100.
In one exemplary embodiment, the present invention also provides a kind of computer readable storage mediums, are stored thereon with calculating
Machine program when the computer program is executed by processor, realizes the intelligent allocation method of distributing real time system as previously shown.
It should be understood that the present invention is not limited to the precise structure already described above and shown in the accompanying drawings, and
And various modifications and change can executed without departing from the scope.The scope of the present invention is limited only by the attached claims.
Claims (10)
1. a kind of intelligent allocation method of distributing real time system, which is characterized in that the described method includes:
The affairs that server obtains a certain node flow into the affairs outflow information of information and the node;Wherein, the transaction flow
Entering information includes flowing into the Transaction Information of all the first affairs of the node, and the affairs outflow information includes flowing out the section
The Transaction Information of all the second affairs of point;
The server flows into information according to the affairs and the affairs flow out information, determines the currently processed of the node
Load;
The server judges whether the present processing load is greater than processing load threshold value;
If the present processing load be greater than the processing load threshold value, the server by the node at least one of
The Transaction Information of target transaction is produced to replacement node, and the target transaction is described in removing in the first affairs on the node
At least one affairs in all affairs except second affairs.
2. the method according to claim 1, wherein the affairs flow into information further include affairs flow into log,
The affairs flow into the Transaction Identifier number that log carries first affairs, and the affairs outflow information further includes affairs outflow
Log, the affairs outflow log carry the Transaction Identifier number of second affairs, and the server is according to the transaction flow
Enter information and affairs outflow information, determine the present processing load of the node, comprising:
Server Transaction Identifier number according to entrained by affairs inflow log and affairs outflow log are taken
The Transaction Identifier number of band determines that the Current transaction of the node, the Current transaction are in the first affairs on the node
Remove all affairs except second affairs;
The server is using the Transaction Information of the Current transaction as the present processing load of the node.
3. according to the method described in claim 2, it is characterized in that, the affairs, which flow into log, also carries first affairs
Transaction types, affairs outflow log also carries the transaction types of second affairs, if described currently processed negative
Lotus is greater than the processing load threshold value, and the server produces the Transaction Information of at least one target transaction on the node
To replacement node, comprising:
The server determines at least one target transaction from the Current transaction;
The server inquires preset transaction types and the node table of comparisons, to find out transaction types and the target transaction
The identical replacement node of transaction types;
The server produces the Transaction Information of at least one of described target transaction to the replacement node, so that the replacement
Node completes the processing of the Transaction Information at least one of described target transaction.
4. described in any item methods according to claim 1~3, which is characterized in that the server by the node extremely
The Transaction Information of one item missing target transaction is produced to replacement node, the method also includes:
The server obtains the flow direction of the target transaction;
The server updates the flow direction record of the target transaction, the flow direction record according to the flow direction of the target transaction
For tracking the circulation path for counting the target transaction.
5. the method according to claim 1, wherein the method also includes:
If the present processing load is not more than the processing load threshold value, the server is defeated by the present processing load
Enter to processing load forecasting model;
The server is based on the output of the processing load forecasting model as a result, determining that the prediction of the node handles load;
The server judges whether the prediction processing load is greater than the processing load threshold value;
If the prediction processing load is greater than the processing load threshold value, will be on the node described in the server execution
At least one of target transaction Transaction Information produce to replacement node.
6. according to the method described in claim 5, it is characterized in that, if the present processing load is negative no more than the processing
Lotus threshold value, before the present processing load is input to processing load forecasting model by the server, the method also includes:
The server obtains the history processing load of the node;
History described in the server by utilizing handles the initial neural network model of weight training, obtains the processing load prediction mould
Type.
7. according to the method described in claim 6, it is characterized in that, at the beginning of the processing weight training of history described in the server by utilizing
Beginning neural network model obtains the processing load forecasting model, comprising:
The server obtains N number of continuous processing load as sample from history processing load and inputs initial nerve net
Network model, so that the initial neural network model exports output information corresponding with the sample;The N is positive integer;
The output load is determined as target processing load by the server;
Target processing load is compared by the server with actual treatment load, is updated according to comparison result described first
The parameter of beginning neural network model, the actual treatment load are the servers when getting the target processing load
Carve from the nodal test to processing load;
The server judges whether the loss function of the initial neural network model meets preset condition, the loss function
The error between output load and actual value for characterizing the initial neural network model;
If the loss function of the initial neural network model meets the preset condition, the server is by the original net
The parameter current of network model is determined as handling the parameter of load forecasting model, and it is pre- according to the parameter to obtain the processing load
Survey model.
8. a kind of server, which is characterized in that the server includes:
Module is obtained, the affairs for obtaining a certain node flow into information and the affairs of the node flow out information;Wherein, described
It includes flowing into the Transaction Information of all the first affairs of the node that affairs, which flow into information, and the affairs outflow information includes outflow
The Transaction Information of all the second affairs of the node;
First determining module determines the node for flowing into information and affairs outflow information according to the affairs
Present processing load;
Judgment module, for judging whether the present processing load is greater than processing load threshold value;
Shift module, for the judgment module judge the present processing load be greater than the processing load threshold value when,
The Transaction Information of at least one target transaction on the node is produced to replacement node, the target transaction is the node
On the first affairs in remove in all affairs except second affairs at least one of affairs.
9. a kind of computer readable storage medium, which is characterized in that it stores computer program, and the computer program makes to succeed in one's scheme
Calculation machine perform claim requires the intelligent allocation method of 1~7 described in any item distributing real time systems.
10. a kind of electronic equipment, including memory and processor, the memory are stored with computer program, which is characterized in that
The step of processor realizes method according to any one of claims 1 to 7 when executing the computer program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811191526.7A CN109523123B (en) | 2018-10-12 | 2018-10-12 | Intelligent allocation method for distributed transaction processing and server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811191526.7A CN109523123B (en) | 2018-10-12 | 2018-10-12 | Intelligent allocation method for distributed transaction processing and server |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109523123A true CN109523123A (en) | 2019-03-26 |
CN109523123B CN109523123B (en) | 2024-04-05 |
Family
ID=65771813
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811191526.7A Active CN109523123B (en) | 2018-10-12 | 2018-10-12 | Intelligent allocation method for distributed transaction processing and server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109523123B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20000073142A (en) * | 1999-05-06 | 2000-12-05 | 서평원 | Apparatus And Method For Overload State Sensing Of Message |
CN102055675A (en) * | 2011-01-21 | 2011-05-11 | 清华大学 | Multipath routing distribution method based on load equilibrium |
CN104580396A (en) * | 2014-12-19 | 2015-04-29 | 华为技术有限公司 | Task scheduling method, node and system |
CN105991458A (en) * | 2015-02-02 | 2016-10-05 | 中兴通讯股份有限公司 | Load balancing method and load balancing device |
CN106101232A (en) * | 2016-06-16 | 2016-11-09 | 北京思源置地科技有限公司 | Load-balancing method and device |
CN106302161A (en) * | 2016-08-01 | 2017-01-04 | 广东工业大学 | Perception data transmission method based on load estimation, device, path control deivce |
CN106484530A (en) * | 2016-09-05 | 2017-03-08 | 努比亚技术有限公司 | A kind of distributed task dispatching O&M monitoring system and method |
-
2018
- 2018-10-12 CN CN201811191526.7A patent/CN109523123B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20000073142A (en) * | 1999-05-06 | 2000-12-05 | 서평원 | Apparatus And Method For Overload State Sensing Of Message |
CN102055675A (en) * | 2011-01-21 | 2011-05-11 | 清华大学 | Multipath routing distribution method based on load equilibrium |
CN104580396A (en) * | 2014-12-19 | 2015-04-29 | 华为技术有限公司 | Task scheduling method, node and system |
CN105991458A (en) * | 2015-02-02 | 2016-10-05 | 中兴通讯股份有限公司 | Load balancing method and load balancing device |
CN106101232A (en) * | 2016-06-16 | 2016-11-09 | 北京思源置地科技有限公司 | Load-balancing method and device |
CN106302161A (en) * | 2016-08-01 | 2017-01-04 | 广东工业大学 | Perception data transmission method based on load estimation, device, path control deivce |
CN106484530A (en) * | 2016-09-05 | 2017-03-08 | 努比亚技术有限公司 | A kind of distributed task dispatching O&M monitoring system and method |
Non-Patent Citations (2)
Title |
---|
陈世浩等: "基于预测的计算网格负载平衡研究", vol. 26, no. 26, pages 82 - 85 * |
齐璇;: "支持动态修改的MINI工作流过程元模型", 计算机工程与科学, no. 07, 15 July 2007 (2007-07-15), pages 138 - 140 * |
Also Published As
Publication number | Publication date |
---|---|
CN109523123B (en) | 2024-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106982359B (en) | A kind of binocular video monitoring method, system and computer readable storage medium | |
CN109582793A (en) | Model training method, customer service system and data labeling system, readable storage medium storing program for executing | |
CN104144204B (en) | A kind of method and system for the simulation for carrying out industrial automation system | |
CN106549772A (en) | Resource prediction method, system and capacity management device | |
CN105989441A (en) | Model parameter adjustment method and device | |
CN109614238A (en) | A kind of recongnition of objects method, apparatus, system and readable storage medium storing program for executing | |
CN109726664A (en) | A kind of intelligence dial plate recommended method, system, equipment and storage medium | |
CN110442737A (en) | The twin method and system of number based on chart database | |
CN108809694A (en) | Arranging service method, system, device and computer readable storage medium | |
CN110139067A (en) | A kind of wild animal monitoring data management information system | |
CN108320045A (en) | Student performance prediction technique and device | |
CN109840111A (en) | A kind of patterned transaction processing system and method | |
US20200372428A1 (en) | Automated and efficient personal transportation vehicle sharing | |
CN109389518A (en) | Association analysis method and device | |
CN109740965A (en) | A kind of engineering verification analysis method and device | |
CN104866922B (en) | A kind of off-grid prediction technique of user and device | |
CN110535850A (en) | Treating method and apparatus, storage medium and the electronic device that account number logs in | |
CN113486584B (en) | Method and device for predicting equipment failure, computer equipment and computer readable storage medium | |
CN107273979A (en) | The method and system of machine learning prediction are performed based on service class | |
CN107924492A (en) | Classified using normalization the value of the confidence to mobile equipment behavior | |
CN109995592A (en) | Quality of service monitoring method and equipment | |
CN106909560A (en) | Point of interest sort method | |
CN110458572A (en) | The determination method of consumer's risk and the method for building up of target risk identification model | |
CN110263136B (en) | Method and device for pushing object to user based on reinforcement learning model | |
CN109409780B (en) | Change processing method, device, computer equipment and storage medium |
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 |