CN109871400A - A kind of big data calculating control system and method based on cloud service platform - Google Patents
A kind of big data calculating control system and method based on cloud service platform Download PDFInfo
- Publication number
- CN109871400A CN109871400A CN201811602311.XA CN201811602311A CN109871400A CN 109871400 A CN109871400 A CN 109871400A CN 201811602311 A CN201811602311 A CN 201811602311A CN 109871400 A CN109871400 A CN 109871400A
- Authority
- CN
- China
- Prior art keywords
- data
- node
- service platform
- cloud service
- sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention belongs to big data technical fields, disclose a kind of big data calculating control system and method based on cloud service platform；Using data collecting module collected all data, by initial data and by parallel processing module and log processing module treated data by data memory module storage in memory, user can check all data by display module.Using Spark memory computing technique, every common parser is completed, improves efficiency and serious forgiveness, while the calculated result high to requirement of real-time is pushed to application layer or client；The prediction of the monitoring and future trend to real-time dynamic data is effectively realized using time series analysis method；Analytical technology is calculated using real-time big data, completes dispatching algorithm, dynamic prediction, safety analysis, analyzes purchase volume, the prediction energy-consuming, energy management user of electric power or other energy.Structure of the invention is reasonable, can effectively improve the utilization of the energy.
Description
Technical field
The invention belongs to big data technical fields more particularly to a kind of big data based on cloud service platform to calculate control system
System and method.
Background technique
Currently, the prior art commonly used in the trade is such that as time goes on, big data has permeated what we lived
Every aspect.The data volume generated in economic life or industrial production is increasing, and traditional calculating control system is
It is not enough to cope with such huge data volume.In order to make full use of these data, to further speed up the progress or more of technology
Good serves people's lives, has expedited the emergence of big data system.Energy management platform exists as independent software systems, crucial
Function is to show energy consumption classification indicators, fails to combine with production control system, it is difficult to realize energy consumption automation, intelligence
The target of monitoring and Reverse Turning Control.Existing big data analysis technology memory usage is high, causes arithmetic speed low；Operation efficiency
Not high, data feedback is slow.
In conclusion problem of the existing technology is: existing big data analysis technology memory usage is high, causes to transport
It is low to calculate rate；Operation efficiency is not high, and data feedback is slow.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of, and the big data based on cloud service platform calculates control
System and method.
The invention is realized in this way a kind of big data calculation control method based on cloud service platform, described to be based on cloud
The big data of service platform calculate control the following steps are included:
The first step acquires all data；
Second step carries out time series processing to daily record data, forms data flow；Based on the communication of more granularities
Dijkstra parallel algorithm, Spark memory computing technique, the real-time big data of time series analysis method calculate analysis processing acquisition
Data generate parallel result；
Third step merges parallel as a result, generating processing result；
4th step, using memory is by the data resource of acquisition and treated that data store；
5th step shows the data information in the data memory module using display.
Further, the Dijkstra parallel algorithm of more granularity communications includes: to seek since data source point in the second step
Look for the associated back end of data source point, be then taken out weight minimum data node；Repeated expansion process, and update pair
Answer the weight of back end；When back end weight cannot be updated again, the mark of the back end is revised as permanently
Mark；The weight of back end is the shortest distance of the back end to data source point；It is opened when from data source point and data terminal
When the mark of some node is all revised as permanent identification for the first time by the process of beginning, then it can determine that by source point through the node to terminal
Path be shortest path.
Further comprise:
(1) two processes 1,2 are opened up, are expanded since source point and terminal respectively；
(2) node of permanent identification is sent to process 2, judges whether the point identification is also modified to forever in process 2
Mark long；Likewise, permanent identification node is sent to process 1 by process 2, make same judgement；If existing by two processes
When being all identified as the node of permanent identification, then stop operation；If it does not exist, then continue the expansion of node；
(3) duplicate node selection and node loose operations, the permanent identification node newly obtained is swapped between process,
(2) are repeated until getting final path.
Further, time series analysis method includes: in the second step
(1) stationary test is carried out to time series data, by the scatter plot of time series or line chart to sequence into
The preliminary stationarity judgement of row；The stationarity of the sequence is accurately judged using ADF unit root test；To the time sequence of non-stationary
Column first carry out taking logarithm or carry out difference processing, then judge the stationarity of sequence after processing to data；Repeat the above mistake
Journey, until becoming stationary sequence；
(2) coefficient using auto-correlation coefficient and the two statistics of PARCOR coefficients identification ARMA (p, q) model is special
The order of point and model；If the deviation―related function of stationary sequence is truncation, and auto-correlation function is hangover, can conclude that sequence
It is suitble to AR model；If the deviation―related function of stationary sequence is hangover, and auto-correlation function is truncation, then can conclude that sequence is suitable
Close MA model；If the deviation―related function and auto-correlation function of stationary sequence are hangovers, sequence is suitble to arma modeling；From phase
Function is closed into the sequence of periodic law, seasonality product model can be selected；The complicated sequence of auto-correlation function rule, it may be necessary to
Make Fitting of Nonlinear Models；
(3) after determining model order, parameter Estimation should be carried out to arma modeling；Parameter is carried out using least square method OLs
Estimation；
(4) after the identification and parameter Estimation of completing model, estimated result should be diagnosed and is examined, in the hope of selected by discovery
Whether model is suitable；If improper, it should know which kind of modification made in next step.
Further, using memory is by the data resource of acquisition and treated that data carry out storage tool in the 4th step
Body includes:
(1) data resource of acquisition and treated data are divided, each data is handled respectively；
(2) according to acquisition time data processing precision, each all time points are marked in order；
(3) Hash, 256 data one obtained by Hash are carried out with 256 algorithm of SHA-1 with 128 random number seeds S, S
It is divided into two, i.e. S11And S12；S11And S12It repeats the above process, continues through SHA-1 256 and be extended and divide；It extends and divides
Splitting operation will continue to that generated binary tree leaf node can cover selected precision corresponding all time points；
(4) label for generating (2) corresponds on the leaf node of third step generation, in other words each in numerical order
Time point has been converted to 256 hash values；
(5) it is merged according to leaf node situation, is changed into upper layer node after merging, until can not merge；
(6) node after merging carries out out-of-order processing, stores as beginning and ending time Node data.
Another object of the present invention is to provide the big datas described in a kind of realize based on cloud service platform to calculate controlling party
The big data calculating control system based on cloud service platform of method, the big data calculating control system based on cloud service platform
Include:
Data acquisition module is connected with parallel processing module, log processing module, acquires all data；
Parallel processing module handles the data of data collecting module collected, generates parallel result；
Merging module merges the parallel of parallel processing module generation as a result, generating processing result；
Log processing module carries out time series to by daily record data, forms data flow and is handled；
Data memory module stores the data resource of acquisition and treated data；
Display module shows the data information in data memory module.
Another object of the present invention is to provide the big datas described in a kind of realize based on cloud service platform to calculate controlling party
The computer program of method.
Another object of the present invention is to provide the big datas described in a kind of realize based on cloud service platform to calculate controlling party
The information data processing terminal of method.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer
When upper operation, so that computer executes the big data calculation control method based on cloud service platform.
In conclusion advantages of the present invention and good effect are as follows: data collecting module collected all data is utilized, it will be original
Data and memory is stored in by data memory module by parallel processing module and log processing module treated data
In, user can check all data by display module.Utilize the purchase volume of big data analysis electric power or other energy, prediction
Energy-consuming, energy management user improve energy efficiency, reduce energy cost, and structure of the invention is reasonable, can effectively improve energy
The utilization in source.The problems in background technique can effectively be solved.
Parallel processing module of the invention utilizes Spark memory computing technique, completes every common parser, improves
Efficiency and serious forgiveness, while the calculated result high to requirement of real-time is pushed to application layer or client；Parallel processing module
Real-time big data, which calculates, to be analyzed, completion dispatching algorithm, dynamic prediction, safety analysis, the purchase volume of analysis electric power or other energy,
Predict energy-consuming, energy management user.
It compared with prior art, will the invention has the following beneficial effects: data collecting module collected all data is utilized
Initial data and treated that data are stored in by data memory module by parallel processing module and log processing module
In reservoir, user can check all data by display module.Using Spark memory computing technique, every common point is completed
Algorithm is analysed, improves efficiency and serious forgiveness, while the calculated result high to requirement of real-time is pushed to application layer or client；Benefit
The prediction of the monitoring and future trend to real-time dynamic data is effectively realized with time series analysis method；Utilize real-time big data meter
Analytical technology is calculated, dispatching algorithm, dynamic prediction, safety analysis are completed, analyzes purchase volume, the prediction energy of electric power or other energy
Consumption, energy management user.Structure of the invention is reasonable, can effectively improve the utilization of the energy.
Detailed description of the invention
Fig. 1 is the big data calculating control system structural schematic diagram provided in an embodiment of the present invention based on cloud service platform；
In figure: 1, data acquisition module；2, parallel processing module；3, merging module；4, log processing module；5, data are deposited
Store up module；6, display module.
Fig. 2 is the big data calculation control method flow chart provided in an embodiment of the present invention based on cloud service platform.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
It is high for existing big data analysis technology memory usage, cause arithmetic speed low；Operation efficiency is not high, data
Feed back slow problem.The present invention effectively realizes monitoring to real-time dynamic data and future trend using time series analysis method
Prediction；Using real-time big data calculate analytical technology, complete dispatching algorithm, dynamic prediction, safety analysis, analyze electric power or other
Purchase volume, the prediction energy-consuming, energy management user of the energy.Structure of the invention is reasonable, can effectively improve the utilization of the energy.
Application principle of the invention is explained in detail with reference to the accompanying drawing.
As shown in Figure 1, the big data calculating control system provided in an embodiment of the present invention based on cloud service platform includes number
According to acquisition module 1, parallel processing module 2, merging module 3, log processing module 4, data memory module 5 and display module 6.
Data acquisition module 1: being connected with parallel processing module 2, log processing module 4, acquires all data.
Parallel processing module 2: the data that processing data acquisition module 1 acquires generate parallel result.
Merging module 3: merge the parallel of the generation of parallel processing module 2 as a result, generating processing result.
Log processing module 4: carrying out time series to by daily record data, forms data flow and is handled.
Data memory module 5: the data resource of acquisition and treated data are stored.
Display module 6: the data information in display data memory module 5.
As shown in Fig. 2, the big data calculation control method provided in an embodiment of the present invention based on cloud service platform include with
Lower step:
S201: acquisition all data；
S202: time series processing is carried out to daily record data, forms data flow；Dijkstra based on the communication of more granularities
Parallel algorithm, Spark memory computing technique, the real-time big data of time series analysis method calculate analysis processing acquisition data, raw
At parallel result；
S203: merge parallel as a result, generating processing result；
S204: using memory is by the data resource of acquisition and treated that data store；
S205: the data information in the data memory module is shown using display.
In a preferred embodiment of the invention, in step S202, the Dijkstra parallel algorithm of more granularity communications includes:
Dijkstra's algorithm is since data source point, then searching and the associated back end of data source point therefrom take
Weight minimum data node out；Above-mentioned process of expansion is repeated, and updates the weight of corresponding data node；When back end weight not
When can be updated again, the mark of the back end is revised as permanent identification；The weight of back end is the data section at this time
Point arrives the shortest distance of data source point；In view of dijkstra's algorithm every time be choose network in the smallest back end of weight into
Row expansion, when the mark of some node is all revised as permanent identification for the first time by the process since data source point and data terminal
When, then it can determine that the path by source point through the node to terminal is shortest path；
Specific steps include:
1, two processes 1,2 are opened up, are expanded since source point and terminal respectively；
2, the node of permanent identification is sent to process 2, judges whether the point identification is also modified to permanently in process 2
Mark；Likewise, permanent identification node is sent to process 1 by process 2, make same judgement；If it is existing by two processes all
When being identified as the node of permanent identification, then stop operation；If it does not exist, then continue the expansion of node；
3, duplicate node selection and node loose operations, the permanent identification node newly obtained is swapped between process,
Repetitive operation 2 is until getting final path.
In a preferred embodiment of the invention, in step S202, Spark memory computing technique includes:
Spark is the cluster Computing Platform of a realization Universal-purpose quick, is a universal memory parallel computation frame, is used to
Building large size, low latency data analysis application program；
Data are read in memory from disk for the first time and generate a kind of abstract memory object, i.e. elasticity distribution by Spark system
Formula data set (abbreviation RDD), hereafter user program only operates the RDD in memory, and calculating process pertains only to memory read-write, because
This greatly improves data-handling efficiency；And Spark uses the interface of the conversion based on coarseness, this conversion operation is being counted
A kind of directed acyclic graph (DAG) is formed during calculating, referred to as " blood lineage (lineage) ", " blood lineage " is substantially to establish a kind of number
According to a relationship for conversion, rather than data itself, therefore upon system failure, the letter that this " blood lineage " provides can be passed through
Breath, calculates the data of loss, i.e., exchanges that data are mobile and duplication for calculate, can largely save in this way it is fault-tolerant brought by open
Pin；
In a preferred embodiment of the invention, in step S202, time series analysis method includes:
Time series is chronological set of number sequence, and time series analysis is exactly to utilize this group of ordered series of numbers, application
Mathematical statistics method is pocessed, to predict the development of the following things；
Its basic principle: as soon as it is to recognize that the continuity of things development: using past data, it can speculate that the development of things becomes
Gesture；Two allow for the randomness of things development；Anything development may all be influenced by accidentalia, utilize statistics thus
Weighted mean method handles historical data in analysis；
Time series analysis method basic step:
1, data tranquilization is handled:
It first has to carry out stationary test to time series data, the scatter plot or line chart pair of time series can be passed through
Sequence carries out preliminary stationarity judgement；The stationarity of the sequence is generally accurately judged using ADF unit root test；To non-flat
Steady time series first carries out taking logarithm or carries out difference processing, then judges the stationarity of sequence after processing to data.Weight
Multiple above procedure, until becoming stationary sequence.
2, model identifies
The coefficient feature of ARMA (p, q) model is identified using auto-correlation coefficient and the two statistics of PARCOR coefficients
With the order of model；If the deviation―related function of stationary sequence is truncation, and auto-correlation function is hangover, can conclude that sequence is suitable
Close AR model；If the deviation―related function of stationary sequence is hangover, and auto-correlation function is truncation, then can conclude that sequence is suitble to
MA model；If the deviation―related function and auto-correlation function of stationary sequence are hangovers, sequence is suitble to arma modeling；Auto-correlation
Seasonality product model can be selected at the sequence of periodic law in function；The complicated sequence of auto-correlation function rule, it may be necessary to make
Fitting of Nonlinear Models.
3, parameter Estimation
After determining model order, parameter Estimation should be carried out to arma modeling；Parameter is carried out using least square method OLs to estimate
Meter, it should be noted that the parameter Estimation relative difficulty of MA model, Ying Jinliang avoid the moving average model(MA model) or packet using high-order
The arma modeling of the item of rolling average containing high-order.
4, model testing
After the identification and parameter Estimation of completing model, estimated result should be diagnosed and be examined, in the hope of selected by discovery
Model it is whether suitable；If improper, it should know which kind of modification made in next step.This stage main test fitting model be
It is no reasonable；First is that whether the estimated value of testing model parameter has conspicuousness；Second is that whether the residual sequence of testing model is white
Noise.
In a preferred embodiment of the invention, using memory is by the data resource of acquisition and treated in step S204
Data carry out storage and specifically include:
The data resource of acquisition and treated data are divided, are handled respectively each data by the first step；
Each all time points are marked in order according to acquisition time data processing precision for second step；
Third step carries out Hash, 256 digits obtained by Hash with 128 random number seeds S, S with 256 algorithm of SHA-1
According to being divided into two, i.e. S11And S12；S11And S12It repeats the above process, continues through SHA-1256 and be extended and divide；Extension and
Splitting operation will continue to that generated binary tree leaf node can cover selected precision corresponding all time points；
4th step, the label that second step is generated correspond on the leaf node of third step generation in numerical order, or
Person says that each time point has been converted to 256 hash values；
5th step merges according to leaf node situation, is changed into upper layer node after merging, until that can not merge into
Only；
6th step, the node after merging carry out out-of-order processing, store as beginning and ending time Node data.
A kind of big data calculating control system based on cloud service platform provided in an embodiment of the present invention, is acquired using data
Module 1 acquires all data, and by initial data and by parallel processing module 2 and log processing module 4, treated that data are led to
Cross the storage of data memory module 5 in memory, user can check all data by display module 6.Utilize big data point
The purchase volume of electric power or other energy is analysed, prediction energy-consuming, energy management user, energy efficiency is improved, reduces energy cost,
Structure of the invention is reasonable, can effectively improve the utilization of the energy.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or
Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to
Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network
Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one
Computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from one
A web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)
Or wireless (such as infrared, wireless, microwave etc.) mode is carried out to another web-site, computer, server or data center
Transmission).The computer-readable storage medium can be any usable medium or include one that computer can access
The data storage devices such as a or multiple usable mediums integrated server, data center.The usable medium can be magnetic Jie
Matter, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid
State Disk (SSD)) etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (9)
1. a kind of big data calculation control method based on cloud service platform, which is characterized in that described based on cloud service platform
Big data calculate control the following steps are included:
The first step acquires all data；
Second step carries out time series processing to daily record data, forms data flow；Dijkstra based on the communication of more granularities is simultaneously
Row algorithm, Spark memory computing technique, the real-time big data of time series analysis method calculate analysis processing acquisition data, generate
Parallel result；
Third step merges parallel as a result, generating processing result；
4th step, using memory is by the data resource of acquisition and treated that data store；
5th step shows the data information in the data memory module using display.
2. the big data calculation control method based on cloud service platform as described in claim 1, which is characterized in that described second
The Dijkstra parallel algorithm of more granularity communications includes: to find and the associated data of data source point since data source point in step
Then node is taken out weight minimum data node；Repeated expansion process, and update the weight of corresponding data node；Work as number
When cannot be updated again according to node weight, the mark of the back end is revised as permanent identification；The weight of back end is
The shortest distance of the back end to data source point；When the process since data source point and data terminal for the first time saves some
When the mark of point is all revised as permanent identification, then it can determine that the path by source point through the node to terminal is shortest path.
3. the big data calculation control method based on cloud service platform as claimed in claim 2, which is characterized in that further packet
It includes:
(1) two processes 1,2 are opened up, are expanded since source point and terminal respectively；
(2) node of permanent identification is sent to process 2, judges whether the point identification is also modified to permanently mark in process 2
Know；Likewise, permanent identification node is sent to process 1 by process 2, make same judgement；If existing all marked by two processes
When knowing the node for permanent identification, then stop operation；If it does not exist, then continue the expansion of node；
(3) duplicate node selection and node loose operations, the permanent identification node newly obtained is swapped between process, is repeated
(2) until getting final path.
4. the big data calculation control method based on cloud service platform as described in claim 1, which is characterized in that described second
Time series analysis method includes: in step
(1) stationary test is carried out to time series data, sequence is carried out just by the scatter plot or line chart of time series
The stationarity of step judges；The stationarity of the sequence is accurately judged using ADF unit root test；To the time series of non-stationary,
First data are carried out taking logarithm or carry out difference processing, then judges the stationarity of sequence after processing；Above procedure is repeated, directly
To as stationary sequence；
(2) using auto-correlation coefficient and the two statistics of PARCOR coefficients identification ARMA (p, q) model coefficient feature and
The order of model；If the deviation―related function of stationary sequence is truncation, and auto-correlation function is hangover, can conclude that sequence is suitble to
AR model；If the deviation―related function of stationary sequence is hangover, and auto-correlation function is truncation, then can conclude that sequence is suitble to MA
Model；If the deviation―related function and auto-correlation function of stationary sequence are hangovers, sequence is suitble to arma modeling；Auto-correlation letter
Seasonality product model can be selected at the sequence of periodic law in number；The complicated sequence of auto-correlation function rule, it may be necessary to make non-
Linear model fitting；
(3) after determining model order, parameter Estimation should be carried out to arma modeling；Parameter Estimation is carried out using least square method OLs；
(4) after the identification and parameter Estimation of completing model, estimated result should be diagnosed and is examined, selected in the hope of discovery
Whether model is suitable；If improper, it should know which kind of modification made in next step.
5. the big data calculation control method based on cloud service platform as described in claim 1, which is characterized in that the described 4th
Using memory is by the data resource of acquisition and treated that data carry out that storage specifically includes in step:
(1) data resource of acquisition and treated data are divided, each data is handled respectively；
(2) according to acquisition time data processing precision, each all time points are marked in order；
(3) Hash is carried out with 256 algorithm of SHA-1 with 128 random number seeds S, S, 256 data one obtained by Hash, which are divided, is
Two, i.e. S11And S12；S11And S12It repeats the above process, continues through SHA-1 256 and be extended and divide；Extension and division behaviour
Work will continue to that generated binary tree leaf node can cover selected precision corresponding all time points；
(4) label for generating (2) corresponds on the leaf node of third step generation, in other words each time in numerical order
Point has been converted to 256 hash values；
(5) it is merged according to leaf node situation, is changed into upper layer node after merging, until can not merge；
(6) node after merging carries out out-of-order processing, stores as beginning and ending time Node data.
6. a kind of big data calculation control method realized described in claim 1 based on cloud service platform based on cloud service platform
Big data calculating control system, which is characterized in that the big data calculating control system based on cloud service platform includes:
Data acquisition module is connected with parallel processing module, log processing module, acquires all data；
Parallel processing module handles the data of data collecting module collected, generates parallel result；
Merging module merges the parallel of parallel processing module generation as a result, generating processing result；
Log processing module carries out time series to by daily record data, forms data flow and is handled；
Data memory module stores the data resource of acquisition and treated data；
Display module shows the data information in data memory module.
7. a kind of realize described in Claims 1 to 5 any one based on the big data calculation control method of cloud service platform
Calculation machine program.
8. a kind of letter for realizing the big data calculation control method described in Claims 1 to 5 any one based on cloud service platform
Cease data processing terminal.
9. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed
Benefit requires the big data calculation control method described in 1-5 any one based on cloud service platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811602311.XA CN109871400A (en) | 2018-12-26 | 2018-12-26 | A kind of big data calculating control system and method based on cloud service platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811602311.XA CN109871400A (en) | 2018-12-26 | 2018-12-26 | A kind of big data calculating control system and method based on cloud service platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109871400A true CN109871400A (en) | 2019-06-11 |
Family
ID=66917145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811602311.XA Pending CN109871400A (en) | 2018-12-26 | 2018-12-26 | A kind of big data calculating control system and method based on cloud service platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871400A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105740431A (en) * | 2016-01-29 | 2016-07-06 | 湖南大学 | Time series data prediction method based on multi-layer period exponent hierarchy |
| CN106407837A (en) * | 2016-09-18 | 2017-02-15 | 成都信息工程大学 | A time data encryption processing method having privacy protection capability |
| CN107329982A (en) * | 2017-06-01 | 2017-11-07 | 华南理工大学 | A kind of big data parallel calculating method stored based on distributed column and system |
| US20180060330A1 (en) * | 2016-08-26 | 2018-03-01 | International Business Machines Corporation | Parallel scoring of an ensemble model |
-
2018
- 2018-12-26 CN CN201811602311.XA patent/CN109871400A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105740431A (en) * | 2016-01-29 | 2016-07-06 | 湖南大学 | Time series data prediction method based on multi-layer period exponent hierarchy |
| US20180060330A1 (en) * | 2016-08-26 | 2018-03-01 | International Business Machines Corporation | Parallel scoring of an ensemble model |
| CN106407837A (en) * | 2016-09-18 | 2017-02-15 | 成都信息工程大学 | A time data encryption processing method having privacy protection capability |
| CN107329982A (en) * | 2017-06-01 | 2017-11-07 | 华南理工大学 | A kind of big data parallel calculating method stored based on distributed column and system |
Non-Patent Citations (2)
| Title |
|---|
| 匿名: "基于时间序列模型的GDP预测", 《道客巴巴URL：HTTP://WWW.DOC88.COM/P-4925905043253.HTML》 * |
| 孙文彬: "基于多粒度通讯的Dijkstra并行算法优化", 《中国矿业大学学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070117B (en) | Data processing method and device | |
| AU2016328959B2 (en) | Updating attribute data structures to indicate trends in attribute data provided to automated modeling systems | |
| US20190213115A1 (en) | Utilizing artificial intelligence to test cloud applications | |
| CN109492858B (en) | Employee performance prediction method and device based on machine learning, equipment and medium | |
| US20170109636A1 (en) | Crowd-Based Model for Identifying Executions of a Business Process | |
| EP1631002A2 (en) | Automatic configuration of network performance models | |
| KR20230002702A (en) | Dynamic automation of pipeline artifact selection | |
| WO2017071369A1 (en) | Method and device for predicting user unsubscription | |
| US11675643B2 (en) | Method and device for determining a technical incident risk value in a computing infrastructure from performance indicator values | |
| WO2022142013A1 (en) | Artificial intelligence-based ab testing method and apparatus, computer device and medium | |
| CN110348578A (en) | A kind of security incident scene deduces construction method, system, equipment and medium | |
| CN114968727B (en) | Database through infrastructure fault positioning method based on artificial intelligence operation and maintenance | |
| WO2021191702A1 (en) | Offloading statistics collection | |
| US20170109637A1 (en) | Crowd-Based Model for Identifying Nonconsecutive Executions of a Business Process | |
| CN109711555B (en) | Method and system for predicting single-round iteration time of deep learning model | |
| Ismaeel et al. | Real-time energy-conserving vm-provisioning framework for cloud-data centers | |
| US20230133541A1 (en) | Alert correlating using sequence model with topology reinforcement systems and methods | |
| US11782923B2 (en) | Optimizing breakeven points for enhancing system performance | |
| CN109871400A (en) | A kind of big data calculating control system and method based on cloud service platform | |
| US20230137184A1 (en) | Incremental machine learning for a parametric machine learning model | |
| JP2022168859A (en) | Computer implementation method, computer program, and system (prediction query processing) | |
| JP2022095895A (en) | Traffic data prediction method, traffic data prediction device, electronic device, storage medium, computer program product, and computer program | |
| CN113934894A (en) | Data display method based on index tree and terminal equipment | |
| Morichetta et al. | Demystifying deep learning in predictive monitoring for cloud-native SLOs | |
| CN117131791B (en) | Model evaluation method, system and storage medium based on big data platform |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Cheng Guogen Inventor after: Li Xinjie Inventor before: Cheng Guogen Inventor before: Li Xinran |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190611 |