CN105955814A - Event priority scheduling-based complex event detection method - Google Patents

Event priority scheduling-based complex event detection method Download PDF

Info

Publication number
CN105955814A
CN105955814A CN201610301700.3A CN201610301700A CN105955814A CN 105955814 A CN105955814 A CN 105955814A CN 201610301700 A CN201610301700 A CN 201610301700A CN 105955814 A CN105955814 A CN 105955814A
Authority
CN
China
Prior art keywords
event
priority
detection
events
complex
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
Application number
CN201610301700.3A
Other languages
Chinese (zh)
Inventor
王建华
王涛
程良伦
彭孝东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China Agricultural University
Guangdong University of Technology
Original Assignee
South China Agricultural University
Guangdong University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China Agricultural University, Guangdong University of Technology filed Critical South China Agricultural University
Priority to CN201610301700.3A priority Critical patent/CN105955814A/en
Publication of CN105955814A publication Critical patent/CN105955814A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses an event priority scheduling-based complex event detection method. Different important degrees and response time demand principles are provided according to different application events; an event priority scheduling-based quick detection and response policy is designed; and an event with high priority index is preferentially selected and detected to ensure execution of massive data detection and system response so as to achieve the purpose of time constraint. Conventional event sequence scanning and sequence process are improved, an existing complex event detection technology is extended, and the capability of detecting complex events in massive data flow is greatly improved. According to the method, for the problems of long detection time, low response speed and low detection efficiency of a current complex event detection method in detecting important events on the massive data flow, a current complex event mode detection method is improved, and the existing complex event detection technology is extended, so that the detection of the complex events on massive data can be finished more efficiently.

Description

A kind of complex events detecting methods based on Event Priority scheduling
Technical field
The present invention relates to mass data flow processing technology field, relate more specifically to mass data flow process in one Plant complex events detecting methods based on Event Priority scheduling.
Background technology
Technology of Internet of things is by electronic information technology and biographies such as network, embedded, RFID, sensor and executors System production technology blends, it is achieved contains the full-range data perception of production, transmit, process, control and service Deng, improve traditional product technical value added, strengthen the management and control ability produced with service process, expedite the emergence of modern production A kind of Information Service Method of pattern.
In modern production environment, due to production-scale maximization, production process spatial and temporal distributions, produce stream The complexity day by day of journey and the multi-source of production environment do melancholy, substantial amounts of awareness apparatus, such as RFID label tag, pass Sensor nodes etc. are deployed to production scene and go the people of monitoring field, material, equipment, production process, product, Service, the various objects such as environmental change.These equipment are with automatic quick mode Quick Acquisition and form various magnanimity Creation data stream. owing to these magnanimity creation data stream exists: 1) data volume is the hugest, in magnanimity, often Second can reach TB level even PB level scale；2), there is people, material, equipment, work in data content multi-source Skill process, product, the multiple data content such as service；3) data structure is complicated, structural data, half structure Change data and unstructured data coexists；4) data renewal speed is fast, data every point each seconds all producing and Update；5) data response requires height, and data need to process in real time the data characteristicses such as quick response, result in life During product, magnanimity creation data stream processes and is faced with: magnanimity creation data stream is difficult to process in time problem.And show Some data processing methods are difficult to directly support the process completely of above-mentioned magnanimity creation data, cause enterprise directly from The various mass data flow of above-mentioned generation directly obtains oneself information needed and becomes the most difficult, thus impact produces Enterprise's control real-time to production process effect.Owing to complicated event detection technique can utilize between event attribute Association, is constantly filtered the magnanimity creation data stream of arrival continuously, finds out needs by matched rule or algebraic manipulation Meet the sequence of events of certain interconnection constraint, thus in manufacturing enterprise, in recent years obtain increasingly extensive concern.
Currently, about the research of complex events detecting methods, mainly carry out and have based on automat, based on Petri Net, based on coupling tree and detection method based on aspects such as directed graphs.But owing to they exist in the design process , there is range of application little in the shortcoming that object-oriented limits, semantic description is not enough, it is achieved the problems such as cost is high, leads Cause the detection that they cannot directly be efficiently applied on mass data flow.For the problems referred to above, their some change Enter method, as based on Time Petri Nets complex events detecting methods, tree complicated event based on compressed compositions Detection method, based on improve the complex events detecting methods of figure, complicated event based on pushdown automata structure inspection The complex events detecting methods etc. that survey method, matching result based on coupling tree-model are shared is suggested detection magnanimity Data stream.But owing to said method is only applicable to process event according to the data sequencing time of advent, Causing the detection overlong time of urgent critical event, response speed is slow, the problem that overall detection efficiencies is low, thus Have impact on whole data-flow detection efficiency.
Summary of the invention
The present invention is for overcoming above-mentioned existing complex events detecting methods important thing on detection magnanimity creation data stream Occurring during part that the detection time is long, response speed is slow, the problem that detection efficiency is low, it is proposed that a kind of excellent based on event The complex events detecting methods of first level scheduling.
For solving above-mentioned technical problem, technical scheme is as follows:
A kind of complex events detecting methods based on Event Priority scheduling, said method comprising the steps of:
S1: according to different application affairs, there is different significance levels and response time demand, to all kinds of events Give different priority；
S2: read all kinds of event action from flow of event；
S3: according to all kinds of event informations read, calculate the priority index of all kinds of event；
S4: according to the priority index size of all kinds of events, the event that prioritizing selection priority index is high is examined Survey；
S5: output detections result event.
In the preferred scheme of one, in event detection procedure, it is to there is sequencing between event and event , there is not the competitive relation that two events arrive simultaneously, and the internal memory of the event of reception exists all the time and can expire Foot testing requirement.
In the preferred scheme of one, in event detection, when single mode event is to there being multiple direct sub-thing During part, preferentially perform coupling by allowing the subevent that priority is high.
In the preferred scheme of one, in the matching process, excellent in the response allowing the high event schema of priority First obtain under implementation status, it is also desirable to ensure that the event schema that matching capacity is relatively low in given timing constraint also can Mated timely and responded.
Compared with prior art, technical solution of the present invention provides the benefit that: the invention discloses a kind of based on thing The complex events detecting methods of part priority scheduling.According to different application affairs have different significance levels and Response time demand principle, devises quickly detection and response policy of based on Event Priority scheduling, preferentially selects The event selecting priority index high detects, to ensure that the execution of mass data detection and system response is to meet The purpose of confinement time.Improve regular event sequence scanning and sequence process, extend the inspection of existing complicated event Survey technology, substantially increases complicated event power of test in mass data flow.The present invention is directed to current complicated event When detecting critical event on mass data flow, detection method occurs that the detection time is long, response speed is slow, detection effect The problem that rate is low, improves current complicated event mode detection method, enters existing complicated event detection technique Row extension so that it is the detection completing complicated event efficiently in mass data can be compared.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of the complex events detecting methods that the present invention dispatches based on Event Priority.
Fig. 2 is the present invention with complex events detecting methods based on event time of advent to be compared in terms of the detection time and shows It is intended to.
Fig. 3 is the present invention to be compared in terms of event handling capacity with the complex events detecting methods based on event time of advent Schematic diagram.
Detailed description of the invention
Accompanying drawing being merely cited for property explanation, it is impossible to be interpreted as the restriction to this patent；
With embodiment, technical scheme is described further below in conjunction with the accompanying drawings.
Embodiment 1
A kind of complex events detecting methods based on Event Priority scheduling, comprises the following steps:
S1: according to different application affairs, there is different significance levels and response time demand, to all kinds of events Give different priority；
S2: read all kinds of event action from flow of event；
S3: according to all kinds of event informations read, calculate the priority index of all kinds of event；
S4: according to the priority index size of all kinds of events, the event that prioritizing selection priority index is high is examined Survey；
S5: output detections result event.
In event detection procedure, between event and event, there is sequencing, there are not two events same Time arrive competitive relation, and receive event internal memory exist all the time and disclosure satisfy that testing requirement.
The inventive method implements priority scheduling condition:
(1) building and performing in matching process at Event Pattern Match model, when there is individual event pattern When having multiple direct subevent pattern, will allow the direct subevent pattern match that priority is high, preferential execution is excellent First level high subevent pattern i.e. forms the priority height subevent of event schema；
(2) when multiple event schemas response executed in parallel, it is possible to allow the sound of the event schema that priority is high Should preferentially perform, and the relatively low event schema of the ability that ensures to complete coupling in given timing constraint obtains in time Coupling and response, it is achieved Event Pattern Match scheduling based on priority.
In condition (1), when meeting individual event pattern to when should have multiple direct subevent pattern, will allow The direct subevent pattern that priority is high preferentially performs coupling.
In condition (2), when in multiple event schemas response executed in parallel, allowing the event that priority is high The response of pattern preferentially obtains under implementation status, it is also desirable to ensure to complete the ability of coupling in given timing constraint Relatively low event schema is mated timely and is responded.
Condition (1) and condition (2) realize implementing priority scheduling strategy and realize purpose mainly for improving magnanimity The real-time that data event pattern match processes, it is achieved low time delay fast-response testing goal.
Below concrete matching process is described in detail.In this example, data generator module is utilized to go Simulation produces magnanimity and produces flow of event, generates the specification of event type by controlling data generator module parameter, The probability distribution etc. of flow of event, to realize the needs of experiment demand.The experimental development instrument of the present invention is: Visual C++6.0, test index is: detection time and handling capacity two aspect, and experiment comparative approach is: based on data The time of advent detection method.The detection mode expansions used in experiment is SEQ (A, B, C, D).Table 1 It it is some other major parameter used in experiment.
Table 1
Parameter is set Parameter value
Event type number 10
Each event attribute number 3
Data flow specification 2-8(105)
Event detection modal length 4
Sliding window size 4000
Event detection pattern SEQ (A, B, C, D)
Fig. 1 is that the inventive method is put forward event based on priority scheduling and quickly detects and response policy figure.It is main Content is wanted to include: priority is specifically dispatched, the mapping of business to event schema, the description of Event Pattern Match, The complexity quantization etc. of event schema.
In the design, for ensureing that event quickly responds detection and response demand in mass data, design to be detected The priority index of event should meet equation below:
Object function:
p ( j ) = Σ i = 1 m ( p ( i ) × w i j ) + 1 R + 1 C + n + d - - - ( 1 )
σ (t)=argmaxjp(j) (2)
Wherein: in formula (1), p (j) represents the priority index of event j to be detected, it reflects this event tune The degree influence degree to whole processing system, priority index is the highest, and the process of this event produces shadow to whole system The power of sound is the biggest.P (i) represents the priority index being dispatched to event j to be detected from event i to be detected；wijFor treating Detection event i is dispatched to the probit of event j to be detected；R is system response during customized service logic, C is the impact for event schema (including coupling and the response order of event schema) of the event schema complexity. Before n represents that this event arrives this stage, the upstream phase number it having been carried out process, d is priority Changed factor, is modified dynamic priority according to different conditions, and span is: 0≤d≤1, d=0 For without priority, expression event has just been received by the stage and has joined the team, and d=1 is greatest priority, has maximum scheduling Authority, the stage that therefore event is crossed over is the most, and priority changed factor is the biggest, and its dynamic priority is the highest.Formula (2) limit priority index during σ (t) represents all p (j) of acquisition in.
Fig. 2 is the present invention and complex events detecting methods ratio in terms of the detection time based on the event time of advent Relatively schematic diagram.It can be observed from fig. 2 that under the conditions of same test, compare based on event complexity time of advent thing Part detection method, the inventive method can greatly reduce the event detection time, improves event detection efficiency.Analyze It main reason is that employing inspection policies based on Event Priority in the inventive method goes to detect data stream Middle event, has different significance levels and response time demand principle according to different application affairs, preferentially selects The event selecting priority index high detects, and improves the response speed of the event high to priority, saves It is many unnecessary etc. to be operated, thus realizes saving the lookup time.
Fig. 3 is the present invention with complex events detecting methods based on event time of advent in terms of event handling capacity Relatively schematic diagram.As seen from Figure 3, under the conditions of same test, the inventive method is in event handling capacity side Face is apparently higher than based on the event complex events detecting methods time of advent.Analyze it to main reason is that in the present invention The use of Event Priority scheduling.In the methods of the invention, have according to different application affairs different important Degree and response time demand principle, be provided with different priority index to different application affairs, in detection Time the high event of prioritizing selection priority index detect, it is achieved that critical event is used for quickly detecting behaviour Make, it is to avoid the detection operation of many unnecessary events, thus improve event detection whole efficiency.
The corresponding same or analogous parts of same or analogous label；
Term the being merely cited for property explanation of position relationship described in accompanying drawing, it is impossible to be interpreted as the limit to this patent System；
Obviously, the above embodiment of the present invention is only for clearly demonstrating example of the present invention, and not It it is the restriction to embodiments of the present invention.For those of ordinary skill in the field, in described above On the basis of can also make other changes in different forms.Here without also cannot be to all of enforcement Mode gives exhaustive.All any amendment, equivalent and improvement made within the spirit and principles in the present invention Deng, within should be included in the protection domain of the claims in the present invention.

Claims (4)

1. a complex events detecting methods based on Event Priority scheduling, it is characterised in that described method Comprise the following steps:
S1: according to different application affairs, there is different significance levels and response time demand, to all kinds of events Give different priority；
S2: read all kinds of event action from flow of event；
S3: according to all kinds of event informations read, calculate the priority index of all kinds of event；
S4: according to the priority index size of all kinds of events, the event that prioritizing selection priority index is high is examined Survey；
S5: output detections result event.
2. a complex events detecting methods based on Event Priority scheduling, it is characterised in that examine in event During survey, there is sequencing between event and event, there is not the competition that two events arrive simultaneously Relation, and receive event internal memory exist all the time and disclosure satisfy that testing requirement.
3. a complex events detecting methods based on Event Priority scheduling, it is characterised in that examine in event In survey, when single mode event is to when should have multiple direct subevent, by preferential for the subevent allowing priority high Perform coupling.
4. a complex events detecting methods based on Event Priority scheduling, it is characterised in that mating Cheng Zhong, preferentially obtains under implementation status in the response allowing the high event schema of priority, it is also desirable to ensure to The event schema determining matching capacity in timing constraint relatively low also can be mated timely and be responded.
CN201610301700.3A 2016-05-06 2016-05-06 Event priority scheduling-based complex event detection method Pending CN105955814A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610301700.3A CN105955814A (en) 2016-05-06 2016-05-06 Event priority scheduling-based complex event detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610301700.3A CN105955814A (en) 2016-05-06 2016-05-06 Event priority scheduling-based complex event detection method

Publications (1)

Publication Number Publication Date
CN105955814A true CN105955814A (en) 2016-09-21

Family

ID=56914921

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610301700.3A Pending CN105955814A (en) 2016-05-06 2016-05-06 Event priority scheduling-based complex event detection method

Country Status (1)

Country Link
CN (1) CN105955814A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106600446A (en) * 2016-11-07 2017-04-26 中国科学院自动化研究所 Dynamic-priority-based real-time processing system and method of product field event
CN108347294A (en) * 2017-01-25 2018-07-31 电信科学技术研究院 A kind of data detection method and terminal
CN111026486A (en) * 2019-12-03 2020-04-17 Tcl 移动通信科技(宁波)有限公司 Application sorting method, application sorting device and computer readable storage medium
CN113448679A (en) * 2020-03-25 2021-09-28 北京沃东天骏信息技术有限公司 Method and device for processing call request

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102629352A (en) * 2012-03-08 2012-08-08 中国科学院自动化研究所 Composite event pattern matching method facing real-time perceptual environment and system thereof
US20150205636A1 (en) * 2014-01-17 2015-07-23 Nvidia Corporation Using high priority thread to boost cpu clock rate

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102629352A (en) * 2012-03-08 2012-08-08 中国科学院自动化研究所 Composite event pattern matching method facing real-time perceptual environment and system thereof
US20150205636A1 (en) * 2014-01-17 2015-07-23 Nvidia Corporation Using high priority thread to boost cpu clock rate

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
程良伦,王涛,肖红,王卓薇,王建华,刘军: ""智能制造物联网及应用实例"", 《高科技与产业化》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106600446A (en) * 2016-11-07 2017-04-26 中国科学院自动化研究所 Dynamic-priority-based real-time processing system and method of product field event
CN106600446B (en) * 2016-11-07 2021-04-02 中国科学院自动化研究所 Production field event real-time processing system and method based on dynamic priority
CN108347294A (en) * 2017-01-25 2018-07-31 电信科学技术研究院 A kind of data detection method and terminal
WO2018137719A1 (en) * 2017-01-25 2018-08-02 电信科学技术研究院 Data detection method and terminal
US10827512B2 (en) 2017-01-25 2020-11-03 China Academy Of Telecommunications Technology Data detection method and terminal
CN108347294B (en) * 2017-01-25 2021-03-19 电信科学技术研究院 Data detection method and terminal
CN111026486A (en) * 2019-12-03 2020-04-17 Tcl 移动通信科技(宁波)有限公司 Application sorting method, application sorting device and computer readable storage medium
CN111026486B (en) * 2019-12-03 2022-10-25 Tcl 移动通信科技(宁波)有限公司 Application sorting method, application sorting device and computer readable storage medium
CN113448679A (en) * 2020-03-25 2021-09-28 北京沃东天骏信息技术有限公司 Method and device for processing call request

Similar Documents

Publication Publication Date Title
Gong et al. Community detection in dynamic social networks based on multiobjective immune algorithm
CN104572998B (en) Question and answer order models update method and device for automatically request-answering system
CN105955814A (en) Event priority scheduling-based complex event detection method
CN105577440B (en) A kind of network downtime localization method and analytical equipment
CN107633050A (en) A kind of method that stealing probability is judged based on big data analysis electricity consumption behavior
CN109063939A (en) A kind of wind speed forecasting method and system based on neighborhood door shot and long term memory network
CN107220168B (en) Information system maturity evaluation method based on integration
Brian Hwarng et al. Shift detection and source identification in multivariate autocorrelated processes
He et al. Parametric transformation of timed weighted marked graphs: Applications in optimal resource allocation
CN112990111A (en) Method and device for identifying ozone generation high-value area, storage medium and equipment
CN106447207A (en) Green building pre-evaluation system based on large data analysis
CN110245706A (en) A kind of lightweight target detection network for Embedded Application
CN109816120A (en) Computer room operation management method, electronic device and storage medium
CN109857817A (en) The whole network domain electronic mutual inductor frequent continuous data is screened and data processing method
Yan et al. The application of BP neural network algorithm in optical fiber fault diagnosis
CN103336763B (en) A kind of compound method for measuring similarity of semantic Web service combination result
CN106559290B (en) The method and system of link prediction based on community structure
CN108876210A (en) A kind of recognition methods, system and the device of land use and change causal structure
Szwed Video event recognition with fuzzy semantic petri nets
Duan et al. Extreme learning machine for bank clients classification
CN105468871B (en) A kind of momenttum wheel success Envelope Analysis and application process
CN106097135A (en) A kind of key factor extracting method towards spinning quality fluctuation
CN117474203A (en) Enterprise energy consumption data management system
CN113378514B (en) Multi-label data feature selection processing method and device
Sheng et al. Simulation of the Users' Email Behavior Based on BP-BDI Model

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20160921