CN107122457A - Record the method and its device, computer-readable medium of networks congestion control data - Google Patents
Record the method and its device, computer-readable medium of networks congestion control data Download PDFInfo
- Publication number
- CN107122457A CN107122457A CN201710285002.3A CN201710285002A CN107122457A CN 107122457 A CN107122457 A CN 107122457A CN 201710285002 A CN201710285002 A CN 201710285002A CN 107122457 A CN107122457 A CN 107122457A
- Authority
- CN
- China
- Prior art keywords
- data
- congestion control
- control data
- block type
- queue
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
Abstract
The invention discloses a kind of method and its device, computer-readable medium for recording networks congestion control data, it is intended to when solving a large amount of networks (web) user on-line operation simultaneously, the problem of causing very big load to server.This method comprises the following steps：Network (web) user behavior data is obtained, the network (web) user behavior data is write to the block type queue of local cache；The network (web) user behavior data in the block type queue is handled by least one thread asynchronous operation and non-internal memory type database is persisted to.The device includes：Data acquisition module, for obtaining networks congestion control data, the networks congestion control data is write the block type queue of local cache；Data processing memory module, for being handled by least one thread asynchronous operation by the networks congestion control data in the block type queue and being persisted to non-internal memory type database.
Description
Technical field
The present invention relates to network communication technology field, more particularly to a kind of method for recording networks congestion control data and its
Device, computer-readable medium.
Background technology
With developing rapidly for network technology, various websites or network application also emerge in large numbers like the mushrooms after rain.In network industry
Business field, records the operation behavior of network (web) user, and these data are analyzed, and can allow enterprise in further detail, clearly
Understand to Chu the behavioural habits of user, so as to find out the problem of enterprise marketing environment such as website, channels is present, contribute to enterprise
Industry excavates page conversion ratio, makes the marketing of enterprise more accurate, effective, business conversion ratio is improved, so that the advertisement of enterprise
Income.In order to preferably grasp and analyze network (web) user to the situation of the web page access in website and website, it is necessary to hold
Row web analytics (Web Analytics).Web analytics refer to by analyze network (web) user to the behavior of web page access come
A series of analysis results are provided, utilize provided analysis result to facilitate and be visually known network (web) user to net
The trend of access to web page.Specifically, analysis result potentially includes the point of the web page access order inside website there is provided visitor
Routing information is hit, can preferably be organized using such information web technology personnel and adjust the structure of web page inside website and interior
Hold and arrange.
Network (web) user behavior analysis generally comprises data acquisition and data analysis：
Wherein, data acquisition is exactly the process for obtaining Web user behavioral data, for website, and user is obtained automatically
Behavioral data most common method is namely based on the method (Server log) of server log, is exactly by web server institute
The journal file of generation obtains useful data, under normal circumstances existed in journal file largely with user behavior analysis without
The redundant data of pass, how to be extracted from the data of these magnanimity with user behavior analysis is a very crucial step, therefore
Need to pre-process journal file, then analyzed, currently the most important ones technology is exactly the excavation of web daily records.But,
It is many to be difficult to obtain from journal file on the valuable data of user behavior analysis, therefore, in order to further obtain on
The valuable data of user behavior, gradually generate the behavior number that interaction scenario between user and website is directly obtained from client
According to method, mainly include the method based on browser, the method based on website and the method based on agency, wherein, be based on
The method of browser is to go to browse web sites using the client-side program developed by user, is browsed to existing under normal circumstances
Device is customized, for example IE, Navigator or other browsers, and then the client-side program of exploitation is obtained
Data are simultaneously transferred back to server end by behavioral data that user interacts with website.Method based on website is by tested clothes
It is engaged in device end mounting assembly, the function of this component is exactly to be automatically inserted at event in the Html codes for the webpage to be accessed user
Shell script is managed, and the code of insertion is JavaScript scripts, therefore, is resulted in for the browser of current main-stream
Support.Method based on agency is to set up a proxy server between the client and tested server of test user,
Proxy server is used for completing the collection work of the user behavior data of tested server.Can be to many by the method for agency
Individual different user browses different websites and carries out Data Collection.
So-called data analysis is actually a data mining process that valuable information is obtained from mass data, is passed through
To programs such as filtering, pretreatment, the comprehensive analysis processings of the data on flows that collect, valuable analysis result is therefrom obtained,
And showed in accurate and visual mode.
However, in recording user operation action process, that is to say needs to face two problems in data acquisition：Its
One, because the operation behavior for recording user will necessarily expend response time of the user to page operation, so as to influence Consumer's Experience；
Second, when a large amount of web users are operated online simultaneously, record user behavior directly writes non-internal memory type database, may
Great pressure is caused to the server of non-internal memory type database.
The content of the invention
It is a primary object of the present invention to propose a kind of method and its device, computer for recording networks congestion control data
Computer-readable recording medium, it is intended to when solving a large amount of networks (web) user on-line operation simultaneously, the problem of very big load is caused to server.
To achieve the above object, a kind of method of record network (web) user behavior data that the present invention is provided, this method
Comprise the following steps：
Network (web) user behavior data is obtained, the network (web) user behavior data is write to the resistance of local cache
Plug queue；
By at least one thread asynchronous operation by the network (web) user behavior data in the block type queue
Handled and be persisted to non-internal memory type database.
Wherein, network (web) user behavior data at least includes following information type：
The main body of website is accessed, the information such as the IP address of user are primarily referred to as；
The path of website is accessed, mainly user obtains the link of website by any mode；
User's webpage residence time, user queried which webpage of website, the residence time in each webpage；
Whether user, which reaches, browses purpose, and such as user enters shopping website, if be finally completed transaction；Or user enters
Enter file polling website, if eventually find and obtain the file required for user；
The solicited message of user, such as when into shopping website, user queried the commodity of which kind, or browse
Which information of device interface queries；
The frequency of the date and time of user's request, particularly user's request；
The result that user asks, such as success, failure or Website server transmission mistake,
It is varied due to web server type, the information that different web servers is produced be it is different, each
Website can produce specific user behavior data according to the need for itself.
Server log file is either based on, the method for browser, the method based on website is also based on and is based on
The method of agency obtains network (web) user behavior data, is required to that classification is identified to network (web) user behavior data
Storage.
Wherein, the thread according to access website main body difference sets up one respectively, the thread asynchronous operation processing correspondence
Access website main body network (web) user behavior data.Furthermore it is possible to set each thread contexts to handle multiple institutes
State network (web) user behavior data for accessing website main body.
Wherein, thread, the line according to the information type of the network (web) user behavior data sets up one respectively
Journey asynchronous operation handles network (web) user behavior data of corresponding each described information type.
Further, the method for described record network (web) user behavior data, further comprising the steps of：Obtaining net
Before network (web) user behavior data, a block type queue is preset in local cache.
Wherein, before network (web) user behavior data is obtained, multiple block types can also be set in local cache
Queue, each block type queue association receives different webpages, or the network for accessing website user that different zones are divided
(web) user behavior data, so that the treatment effeciency that further lifting network (web) user behavior data is obtained.
Further, the method for described record network (web) user behavior data, further comprising the steps of：
A default monitoring thread, and block type queue in the local cache is monitored by the monitoring thread in real time
Length.
Further, the method for described record network (web) user behavior data, further comprising the steps of：
The maximum length threshold values of the block type queue in the local cache is preset, when the length of the block type queue is big
When the maximum length threshold values, increase the number of threads.
Further, the method for described record network (web) user behavior data, further comprising the steps of：
The minimum length threshold values of the block type queue in the local cache is preset, when the length of the block type queue is small
When the minimum length threshold values, the number of threads is reduced.
Wherein, the block type queue receives network (web) user behavior data, and the block type queue will by thread
Network (web) user behavior data is distributed to server storage database, according to the server storage database and described
Data transmission rate between ground caching determines the maximum length threshold values and minimum length threshold values.
Another aspect of the present invention, to achieve the above object, the present invention also propose a kind of record networks congestion control data
Device, the device includes：Memory, processor and it is stored in the number that can be run on the memory and on the processor
According to capture program, following record network (web) user behavior is realized when the data acquisition program is by the computing device
The method and step of data：
Network (web) user behavior data is obtained, the network (web) user behavior data is write to the resistance of local cache
Plug queue；
By at least one thread asynchronous operation by the network (web) user behavior data in the block type queue
Handled and be persisted to non-internal memory type database.
Further, the data acquisition program is also realized following record network (web) during the computing device
The method and step of user behavior data：
Before network (web) user behavior data is obtained, a block type queue is preset in local cache；
And a default monitoring thread, and the block type team in the local cache is monitored by the monitoring thread in real time
The length of row.
Further, the data acquisition program is also realized following record network (web) during the computing device
The method and step of user behavior data：
It is more than the maximum length of the block type queue in the default local cache in the length of the block type queue
During threshold values, increase the number of threads.
Further, the data acquisition program is also realized following record network (web) during the computing device
The method and step of user behavior data：
It is less than the minimum length of the block type queue in the default local cache in the length of the block type queue
During threshold values, the number of threads is reduced.
In addition, to achieve the above object, present invention also offers a kind of computer-readable recording medium, the computer can
The networks congestion control data acquisition program that is stored with storage medium is read, the networks congestion control data acquisition is held by processor
The step of method of above-mentioned record network (web) user behavior data is realized during row.
The method and its device, computer-readable medium of record networks congestion control data proposed by the present invention, realize note
Family operation behavior data are employed, write the high efficiency of data to improve Consumer's Experience by local cache, and pass through asynchronous thread
The processing to record data and record are carried out, while in order to prevent during high concurrent, may result in user behavior data generation number
Amount is more than processing quantity, and blocking so as to cause the queue of local cache to be booked generation, therefore add a monitoring thread to enter
Row intelligent control, reduces the possibility of queue obstruction.
Brief description of the drawings
Fig. 1 is the schematic diagram for the Web courses of work for realizing each embodiment of the invention；
Fig. 2 is one of which Web user behavior analysis system frame construction drawing of the present invention；
Fig. 3 is the method flow block diagram of the first record network (web) user behavior data of the embodiment of the present invention；
Fig. 4 performs structural representation for the method for the first record network (web) user behavior data of the embodiment of the present invention
Figure；
Fig. 5 is the method flow block diagram of second of record network (web) user behavior data of the embodiment of the present invention；
Fig. 6 is the method flow block diagram of the third record network (web) user behavior data of the embodiment of the present invention；
The realization, functional characteristics and advantage of the object of the invention will be described further referring to the drawings in conjunction with the embodiments.
Embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Web (World Wide Web) the i.e. global wide area network, also referred to as WWW that technical solution of the present invention is related to, it is
A kind of based on hypertext and HTTP, global, dynamic interaction, cross-platform distributed graphic information system.It is to set up
A kind of network service on internet, search and browse on internet for viewer information provide it is patterned,
Information node on Internet is organized into one and closed each other by easily accessed intuitive interface, document therein and hyperlink
The network structure of connection.
The form of expression of so-called web (World Wide Web) is as follows：
(1), hypertext (Hyper text)
Hypertext is a kind of user interface mode, to show text and the content related to text.Current hypertext is general
All over existing in the way of electronic document, word therein, which includes, can be linked to the hypertext link of other fields or document
Connect, it is allowed to the word pointed by hypertext link is directly switch to from current reading location.The form of hypertext has a lot, at present
That most-often used is HTML (Hyper Text Markup Language, HTML) and rich text format (Rich
Text Format, RTF).Link on our the daily webpages browsed belongs to hypertext.Hypertext link one kind is of overall importance
Message structure, it establishes the link the different piece in document by keyword, information is searched for interactive mode.
(2), hypermedia (hypermedia)
Hypermedia is the abbreviation of hyper-media.It is hypertext (hypertext) and multimedia under information browse environment
With reference to.User can not only jump to another text from a text, and can activate one section of sound, show a figure, very
One section of animation can extremely be played.Internet is using hypertext and the information-organization style of hypermedia, by the Extended Hyperlink of information
Onto whole Internet.Web is exactly a kind of Hypertext Information System, and a Web main concept is exactly hypertext link.
It causes text to be no longer fixed linear as a book, but can be jumped to from a position other position and from
It is middle to obtain more information, it may also pass on other theme.As long as wanting to know about the content of some theme in this theme
On tap, it is possible to jump on the document comprising this theme.Exactly this multi connectivity is referred to as Web.
(3), HTTP (HTTP, HyperText Transfer Protocol) is to interconnect web-based applications most
For a kind of extensive procotol.
As shown in figure 1, the Web course of work is segmented into：Connection is set up, solicited message is sent, sends response message, closes
Close connection.Client is terminal user, and server end is website, and by using web browser, client is sent out to Web server
Rise and store corresponding html text and image resource on a HTTP request, the Web server of response.Web page is by multiple
Object is constituted, and object (object) is only the file that can be addressed by single uniform resource position mark URL, for example HTMT texts text
Part, still image, JAVA applets, sound bite, animated video etc..Most of Web pages are by single basic HTML
File and some cited objects are constituted.Basic html file quotes other objects of this page using corresponding URL.Often
Individual URL is constituted by depositing the server host name of the object and two parts of pathname of the object.
As shown in Fig. 2 one of which Web user behavior analysis system frame construction drawing, the behavioural information of Web user is from visitor
Family end, user terminal, proxy server are obtained.The object of web Network Users'Behaviors Analysis is：Internet user individual, we
The behavior that user accesses webpage is referred to as event, and when event occurs, user behavior analysis system first has to what is done
It is the corresponding internet behavior of user is obtained by certain method, which part information is distinguish between to user, other portions
Point be used for calculating the interest-degree of user, but collect to browse information often rambling, and there are some interference letters
Breath, it is impossible to reflect the hobby of user completely, it is necessary to be handled accordingly information, data are used as after pretreatment
The basic input excavated, cluster or classified calculating can obtain the degree of correlation of the webpage for some theme, afterwards with reference to user
The webpage of access is for the degree of correlation and some parameters for setting of theme, and defined formula calculates each user for some theme
The degree of correlation, user is referred to as the interest-degree of user by this problem for the degree of correlation of some theme, finally can be according to user couple
In the size of the degree of correlation of this theme, the personalized service such as webpage recommending is provided the user, or reach other mesh
's.As shown in Fig. 2 Web user behavior analysis system is divided into four modules：Data acquisition and pretreatment module, data mining
Module, interest-degree update module.
The effect of data acquisition module is exactly to obtain information above, and they are sent to other modules or is stored in
In corresponding database table, to treat analysis below., it is necessary to analyze whether user is ready to provide corresponding before data acquisition
Information, user typically all want maintain secrecy some important informations of oneself, the information especially with privacy and sensitiveness, for example
Bank's card number, wage, phone number of user etc., 80% user be ready to website provide oneself name, sex, interest love
OK etc., so must be had friendly relations with monitored user's ditch, it is ensured that the system will not obtain the privacy information that user is reluctant to reveal, simply
Observation user's browses information, to obtain their license.
Data preprocessing module is an indispensable link in user behavior analysis system.Large-scale website is produced daily
Raw data volume is surprising, and database is highly prone to the invasion of noise data, AFR control and some inconsistent datas,
Low-quality data will cause the result of low-quality most, and the performance of data mining data can be reduced containing substantial amounts of redundant data, from
And cannot export or obtain insecure output.Make excavation kernel more effectively Extracting Knowledge, be just necessary for it and provide
Accurately, succinct data, pretreatment refers to the analysis, filtering, the conversion for eliminating repetition record, completing data type, inspection of webpage
Data integrity and uniformity are looked into, irrelevant contents in webpage is removed, information is subjected to necessary arrangement, so that data mining
Process is more effectively easy.The method of common data prediction has：Data scrubbing, data conversion, data regularization.
Bulk information shows as text on internet, and text refers to the Web page formatted in half stored with html format
Face and document, potential knowledge how is excavated from these immense text datas being one needs urgent problem to be solved.
, in advance can be by user for data mining technology, it is contemplated that application of the sorting algorithm in user behavior analysis system
Interest topic define, the purpose for the arrangement is that from the angle of user, different users can be according to the reality of oneself
Demand sets theme, and interest-degree of the user for some theme of network analysis is obtained afterwards, and implements corresponding behavior, real
The existing target of oneself.Analysis web basic content of text is a main aspect in interest-degree calculating process, and it is to obtain to use
The network user, different classifications are grouped into the method for classification analysis by the premise of family interest-degree, and these classifications are exactly the emerging of correlation
Interesting theme.
Embodiment 1
Based on above-mentioned web structures and web user behavior analysis systems, each embodiment of the inventive method is proposed.
As shown in figure 3, first embodiment of the invention proposes a kind of method of record network (web) user behavior data, should
Method comprises the following steps：
S101, acquisition network (web) user behavior data, the write-in of the network (web) user behavior data is local slow
The block type queue deposited；
S102, by least one thread asynchronous operation by the network (web) user's row in the block type queue
Handled for data and be persisted to non-internal memory type database.
Wherein, network (web) user behavior data at least includes following information type：
The main body of website is accessed, the information such as the IP address of user are primarily referred to as；
The path of website is accessed, mainly user obtains the link of website by any mode；
User's webpage residence time, user queried which webpage of website, the residence time in each webpage；
Whether user, which reaches, browses purpose, and such as user enters shopping website, if be finally completed transaction；Or user enters
Enter file polling website, if eventually find and obtain the file required for user；
The solicited message of user, such as when into shopping website, user queried the commodity of which kind, or browse
Which information of device interface queries；
The frequency of the date and time of user's request, particularly user's request；
The result that user asks, such as success, failure or Website server transmission mistake,
It is varied due to web server type, the information that different web servers is produced be it is different, each
Website can produce specific user behavior data according to the need for itself.
Server log file is either based on, the method for browser, the method based on website is also based on and is based on
The method of agency obtains network (web) user behavior data, is required to that classification is identified to network (web) user behavior data
Storage.
As shown in figure 4, user behavior data is put into this by the data acquisition pin for obtaining network (web) user behavior data
In the queue of ground caching, such as data 1, data 2, data 3, data 4.., chain arrangement；Set up at least one thread, such as thread 1,
Thread 2, thread 3, thread 4.., consume data therein respectively from queue, and web services are arrived into user behavior data storage
The non-internal memory type database of device.
Wherein, the thread according to access website main body difference sets up one respectively, the thread asynchronous operation processing correspondence
Access website main body network (web) user behavior data.Furthermore it is possible to set each thread contexts to handle multiple institutes
State network (web) user behavior data for accessing website main body.
Wherein, thread, the line according to the information type of the network (web) user behavior data sets up one respectively
Journey asynchronous operation handles network (web) user behavior data of corresponding each described information type.
Both multithreading and asynchronous operation, which can reach, avoids the purpose for calling thread block, so that improves software can
Response.Even sometimes we are considered as multithreading and asynchronous operation is equivalent concept.But, multithreading and asynchronous behaviour
Still have what some were distinguished.And these differences cause the difference on the opportunity using multithreading and asynchronous operation.
All programs finally can all be performed by computer hardware, so the sheet in order to be better understood from asynchronous operation
Matter, we are necessary to learn about its hardware foundation.The friend for being familiar with computer hardware affirms that this word is not strange to DMA, firmly
There is clear and definite DMA mode index in disk, the technical specification of CD-ROM drive, network interface card, sound card, video card are also there are DMA functions in fact.
DMA is exactly the meaning of direct memory access, that is to say, that possess the hardware of DMA functions with internal memory carry out data exchange when
Time can not consume cpu resource.As long as CPU sends an instruction when initiating data transfer, hardware begins to oneself and internal memory
Data are exchanged, hardware can trigger one and interrupt to notify operation to complete after completion of the transmission.These need not consume the CPU time
The hardware foundation of I/O operation exactly asynchronous operation.So in one process as DOS (and without thread concept) system
Asynchronous dma operation can similarly be initiated.
Thread is not the function of a computer hardware, but a kind of logic function that operating system is provided, thread essence
On be one section is concurrently run in process code, so thread needs operating system to put into cpu resource to run and dispatch.
It is good in design because asynchronous operation is born without extra thread, and is handled using the mode of readjustment
In the case of, processing function may not necessarily use shared variable (even if can not completely dispense with, shared variable can be reduced at least
Quantity), reduce the possibility of deadlock.
The advantage of multithreading is it is obvious that the processing routine in thread is still order execution, and the thinking for meeting ordinary people is practised
It is used, so programming is simple.But the shortcoming of multithreading is similarly obvious, the use (abuse) of thread can bring context to system
The added burden of switching.And the shared variable of cross-thread is likely to result in the appearance of deadlock.
Asynchronous and multithreading, from dialectical relationship, asynchronous and multithreading and frequently an identity relation, asynchronous is mesh
, multithreading is that we realize an asynchronous means what is asynchronous：Asynchronous adjusted when a call request is sent to
User, and caller is realized without waiting for the return of its result and asynchronous using multithreading or can give other to enter
Journey is handled.
Wherein, local cache refers to that the local physical memory of client is marked off into a part of space is used for buffering client
The data of server are written back to, because of its outstanding contributions on write-back, therefore local cache is commonly referred to as local write-back.The technology
By the no longer first write service device hard disk of the data of client write-back, but write-back is first write to local write-back buffer, when slow
When depositing space and reaching certain threshold values, then by write back data to server.Have after local write-back buffer function, can drop significantly
Low server read-write pressure and network load.
Wherein, persistence (Persistence), i.e., data (object in such as internal memory) be saved in can persistence deposit
Store up in equipment (such as disk).The main application of persistence is that the object in internal memory is stored in the database of relationship type, certainly
It can also be stored in disk file, in XML data file etc..Persistence is in permanent state and instantaneous shape by routine data
The mechanism changed between state.JDBC is exactly a kind of persistence mechanism.File I/O is also a kind of persistence mechanism.Object Persistence can be with
Meet object serialization.
Wherein, memory database has abandoned the traditional approach of data in magnetic disk management, is weighed based on total data all in internal memory
Newly architecture is devised, and be also correspondingly improved in terms of data buffer storage, fast algorithm, parallel work-flow, so
Data processing speed is more many soon than the data processing speed of traditional database, and typically all common example has more than 10 times
MySQL MEMORY storage engines, eXtremeDB, TT, FastDB, SQLite, Microsoft SQL Server Compact
Deng.Traditional Database Systems are relevant databases, develop the purpose of this database, are permanent, the stable data of processing.
Relational database is emphasized to safeguard the integrality of data, uniformity, but it is difficult to takes the timing restriction about data and its processing into account, no
The need for industrial production management application in real time can be met, because Real-time Transaction requires that system can accurately forecast the operation of affairs
Time.
For disk database, due to disk access, the data transfer of interior external memory, buffer management, wait in line and
Delay of lock etc. causes the WCET that affairs actual average performs the time with estimating to differ greatly, if will be whole
Database or its main " work " are partially placed into internal memory, make each affairs in the process of implementation without I/O, then calibrated for system
Really the run time of estimation and arrangement affairs, strong support is provided with preferable dynamic predictability, while
To realize that the timing restriction of affairs lays the foundation.
Data handled by memory database are typically " of short duration ", that is, have certain effective time, out-of-date, there is new
Data are produced, and current decision-making derives and becomes invalid.So, handled in practical application using memory database real-time
Business logic processing data.And traditional database is intended to permanent, the stable data of processing, its performance objective is that high system is gulped down
The amount of telling and low cost, the real-time of processing data will consider relatively less.Traditional database is utilized in practical application
The not high data of the relative requirement of real-time of this characteristic storage.
The non-internal memory type database that technical solution of the present invention is related to includes relevant database and hard disc data storehouse.
As shown in figure 5, the present invention second record network (web) user behavior data method, this method include with
Lower step：
S201, obtain network (web) user behavior data before, local cache preset a block type queue；
S202, acquisition network (web) user behavior data, the write-in of the network (web) user behavior data is local slow
The block type queue deposited；
S203, by least one thread asynchronous operation by the network (web) user's row in the block type queue
Handled for data and be persisted to non-internal memory type database.
When recording web user behavior datas, do not use it is traditional directly write non-internal memory type perdurable data storehouse, but will
Data write the local cache of the bounded queue of a thread-safe obstruction, because the writing speed of local cache is exceedingly fast,
Consumer's Experience can be significantly improved.
Wherein, before network (web) user behavior data is obtained, multiple block types can also be set in local cache
Queue, each block type queue association receives different webpages, or the network for accessing website user that different zones are divided
(web) user behavior data, so that the treatment effeciency that further lifting network (web) user behavior data is obtained.
As shown in fig. 6, the present invention the third record network (web) user behavior data method, this method include with
Lower step：
S301, obtain network (web) user behavior data before, local cache preset a block type queue；
S302, acquisition network (web) user behavior data, the write-in of the network (web) user behavior data is local slow
The block type queue deposited；
S303, a default monitoring thread, and the block type in the local cache is monitored by the monitoring thread in real time
The length of queue；The maximum length threshold values of the block type queue in the local cache is preset, when the length of the block type queue
When degree is more than the maximum length threshold values, increase the number of threads；Block type queue in the default local cache is most
Small length threshold values, when the length of the block type queue is less than the minimum length threshold values, reduces the number of threads；
S304, by least one thread asynchronous operation by the network (web) user's row in the block type queue
Handled for data and be persisted to non-internal memory type database.
Wherein, the block type queue receives network (web) user behavior data, and the block type queue will by thread
Network (web) user behavior data is distributed to server storage database, according to the server storage database and described
Data transmission rate between ground caching determines the maximum length threshold values and minimum length threshold values.
Embodiment 2
Another aspect of the present invention, to achieve the above object, the present invention also propose a kind of record networks congestion control data
Device, the device includes：Memory, processor and it is stored in the number that can be run on the memory and on the processor
According to capture program, following record network (web) user behavior is realized when the data acquisition program is by the computing device
The method and step of data：
Network (web) user behavior data is obtained, the network (web) user behavior data is write to the resistance of local cache
Plug queue；
By at least one thread asynchronous operation by the network (web) user behavior data in the block type queue
Handled and be persisted to non-internal memory type database.
Wherein, the data acquisition program is also realized following record network (web) user during the computing device
The method and step of behavioral data：
Before network (web) user behavior data is obtained, a block type queue is preset in local cache；
And a default monitoring thread, and the block type team in the local cache is monitored by the monitoring thread in real time
The length of row.
Wherein, the data acquisition program is also realized following record network (web) user during the computing device
The method and step of behavioral data：
It is more than the maximum length of the block type queue in the default local cache in the length of the block type queue
During threshold values, increase the number of threads.
Wherein, the data acquisition program is also realized following record network (web) user during the computing device
The method and step of behavioral data：
It is less than the minimum length of the block type queue in the default local cache in the length of the block type queue
During threshold values, the number of threads is reduced.
Specifically, a block type queue LinkedBlockingQueue is created in local cache, the big of queue is set
Small is capacity.When needing to record user behavior, by the related data production write-in local cache queue of user behavior
LinkedBlockingQueue.The data that user behavior is produced are consumed, pass through the asynchronous operation of one or more threads
Data in queue local cache queue are consumed, while being persisted to non-internal memory type database.Entered by a thread
The length of the current local cache queue of row monitoring, if greater than the threshold value of setting, then opens new thread and is consumed and be put in storage.
If queue is less than the threshold value set for a long time, appropriate to reduce consuming thread number.
BlockingQueue is solved in multithreading well, how highly effective and safe " transmission " data the problem of.By this
The queue class of a little efficient and thread-safes, brings great convenience for the high-quality multithread programs of our fast constructions.
BlockingQueue blocks queue, and as its name suggests, it is a queue first, and a queue is played in data structure
Effect approximately as：By a shared queue, data can be caused to be inputted by one end of queue, from other end output.
Conventional queue mainly has following two：(certainly by different implementations, many differences can also be extended
The queue of type, DelayQueue is exactly one kind therein)；First in first out (FIFO)：The element for the queue first inserted is also at first
Dequeue, similar to the function of queuing.This queue also embodies a kind of fairness to some degree.Last in, first out
(LIFO)：The element dequeue, this queue priority treatment recent occurrence part at first of insertion queue afterwards.
In multi-thread environment, data sharing, such as classical " producer " and " consumption can be easily achieved by queue
In person " model, data sharing between the two can advantageously be realized by queue very much.Assuming that we have some producer's lines
Journey, has several consumer threads again in addition.If producer thread is needed ready data sharing to consumer thread,
Data are transmitted using the mode of queue, it is possible to easily solve the problems, such as the data sharing between them.But if production
Person and consumer are within some period, just in case occur the unmatched situation of data processing speedIdeally, if raw
The speed of production person's output data be more than consumer spending speed, and when produce data accumulation to a certain extent when
Wait, then the producer, which must suspend, waits (obstruction producer thread), to wait consumer thread the data of accumulation
It is disposed, vice versa.(in multithreading field：So-called obstruction, can hang up thread (blocking) in some cases, once
Condition is met, and the thread being suspended can be waken up automatically again).
BlockingQueue two common obstruction scenes：In the case of there is no data in queue, the institute at consumer end
There is thread all can be by self-blocking (hang-up), until there are data to be put into queue.In the case of data are filled up in queue, the producer
All threads at end all can be by self-blocking (hang-up), and until the position being had time in queue, thread is automatically waken up.
BlockingQueue core methed：
1. it is put into data
(1)offer(anObject)：Represent if it would be possible, anObject is added in BlockingQueue, i.e.,
If BlockingQueue can be accommodated, true is returned, false is otherwise returned.(this method does not block current execution method
Thread)；
(2) offer (E o, long timeout, TimeUnit unit)：The time of wait can be set, if referred to
In the fixed time, failure can't be then returned to toward BlockingQueue is added in queue.
(3)put(anObject)：AnObject is added in BlockingQueue, if BlockQueue be not it is free
Between, then call the thread of the method to be blocked has space to be further continued for inside BlockingQueue.
2. obtain data
(1)poll(time)：Take the object ranked the first in BlockingQueue away, can be with if can not take out immediately
Deng the time as defined in time parameters, take and do not return to null then；
(2) poll (long timeout, TimeUnit unit)：Pair of a head of the queue is taken out from BlockingQueue
As if within a specified time, queue once has data can use, then returning to the data in queue immediately.Otherwise the time is known
Time-out also can use without data, return to failure.
(3)take()：Take the object ranked the first in BlockingQueue away, if BlockingQueue is sky, block
Into wait state, until BlockingQueue has, new data are added into；
(4)drainTo()：It is disposable (to specify and obtain from all available data objects of BlockingQueue acquisitions
The number for evidence of fetching), with it, acquisition data efficiency can be lifted；It repeatedly need not in batches lock or discharge lock.
Common BlockingQueue is as follows：
1.ArrayBlockingQueue
Obstruction queue based on array is realized, inside ArrayBlockingQueue, maintains a fixed length array, with
Just the data object in buffer queue, this is a conventional obstruction queue, in addition to a fixed length array,
Also in store two shaping variables inside ArrayBlockingQueue, identify respectively queue head and afterbody in array
Position.
ArrayBlockingQueue is put into data in the producer and consumer obtains data, is all to share same lock pair
As thus also implying that both really can not run parallel, this point is especially different from LinkedBlockingQueue；According to reality
Existing principle is analyzed, and ArrayBlockingQueue completely can be using separation lock, so as to realize that producers and consumers operate
Complete parallel operation.Why Doug Lea do not do so, may is that because ArrayBlockingQueue data are write
Enter and obtain to have operated enough lightly, so that independent lock mechanism is introduced, in addition to carrying out extra complexity to code band,
It accounts for cheap less than any completely in performance.Also have one between ArrayBlockingQueue and LinkedBlockingQueue
It is individual obvious difference is that, the former insert or delete element when will not produce or destroy any extra object instance,
And the latter can then generate an extra Node object.This needs efficiently that concomitantly processing high-volume data are in a long time
In system, still there is certain difference in it for GC influence.And when creating ArrayBlockingQueue, we can be with
Whether the internal lock of control object is using fair lock, and acquiescence is using non-fair lock.
2.LinkedBlockingQueue
Obstruction queue based on chained list, it is similar with ArrayListBlockingQueue, a number is also maintain inside it
According to buffering queue (queue is made up of a chained list), when the producer is toward when being put into a data in queue, queue can be from production
Data are obtained in person's hand, and are buffered in inner queue, and the producer returns immediately；Only when queue buffer reaches that maximum is delayed
When depositing capacity (LinkedBlockingQueue can specify the value by constructed fuction), it can just block producer's queue, until
Consumer consumes a data from queue, and producer thread can be waken up, vice versa for the processing also base at this end of consumer
In same principle.And why LinkedBlockingQueue can efficiently handle concurrent data, also as it is for life
Production person end and consumer end have been respectively adopted independent lock and have carried out control data synchronization, this also means that raw in the case of high concurrent
Production person and consumer can concurrently in operation queue data, the concurrency performance of whole queue is improved with this.
It is used as developer, it would be desirable to if it is noted that one LinkedBlockingQueue object of construction, and does not have
There is its specified amount of capacity, LinkedBlockingQueue can give tacit consent to the capacity (Integer.MAX_ of a similar unlimited size
VALUE), like this, if perhaps the speed of the producer do not wait until that queue full blocks also once being more than the speed of consumer
Produce, Installed System Memory is possible to be consumed totally.
ArrayBlockingQueue and LinkedBlockingQueue be two it is most common be also the most frequently used obstruction team
Row, generally, the producer consumer problem between processing multithreading are enough using the two classes.
3.DelayQueue
Element in DelayQueue is only arrived when its time delay specified, and this yuan can be got from queue
Element.DelayQueue is a queue limited without size, therefore eternal toward the operation (producer) of insertion data in queue
It will not be blocked, and the operation (consumer) for only obtaining data can be just blocked.
Usage scenario：DelayQueue usage scenarios are less, but all quite ingenious, and common example such as uses one
DelayQueue manages the connection queue that a time-out is not responded.
4.PriorityBlockingQueue
(judgement of priority is determined by the incoming Compator objects of constructed fuction for obstruction queue based on priority
It is fixed), but should be noted PriorityBlockingQueue can't the blocking data producer, and can only not disappear
During the data taken, the consumer of blocking data.Therefore to pay special attention to when use, the speed of producer's creation data is absolute
The speed of consumption data of consumers can not be faster than, otherwise the time one grow, can finally exhaust all available heap memory spaces.In reality
During existing PriorityBlockingQueue, the synchronous lock of internal control thread uses fair lock.
5.SynchronousQueue
A kind of non-cushioned waiting list, similar to the direct dealing without intermediary, is somewhat like the producer in primitive society
And consumer, the producer holds the ultimate consumer that product goes to fairground to be sold to product, and consumer must go to fairground to look in person
To the immediate producer of wanted commodity, if a side does not find suitable target, then sorry, everybody is in fairground etc.
Treat.For the BlockingQueue for having buffering, the link (buffering area) of an interim distributors is lacked, if warp
Business is sold, the producer is directly wholesale to dealer product, disappears without taking notice of that these products are eventually sold to those by dealer
Fei Zhe, because dealer can be with a part of commodity of stock, therefore relative to direct dealing pattern, is sold using centre on the whole
The pattern meeting handling capacity of business is higher (can buy and sell in batches)；But then, and because dealer introducing so that product from
The producer is to adding extra bargain link in the middle of consumer, and single product timely responds to performance and may reduced.
One SynchronousQueue of statement has two kinds of different modes, there is not quite alike behavior between them.
The difference of fairness scheme and non-fairness scheme：If using fairness scheme：SynchronousQueue can be using fair lock, and matches somebody with somebody
Unify a fifo queue to block unnecessary producers and consumers, so that the overall fairness policy of system；But if being non-public affairs
Flat-die type powdered (SynchronousQueue acquiescences)：SynchronousQueue is using non-fair lock, while with unification LIFO teams
Row manage unnecessary producers and consumers, and latter mode, if the processing speed of producers and consumers has gap,
Then it is easy to the situation for occurring hungering and thirst, it is possible to which the data for having some producers either consumer cannot all be handled forever.
BlockingQueue not only realizes the basic function that a complete queue has, while in multi-thread environment
Under, he has also managed multi-thread automatic and has waited arousal function automatically, so that ignoring these details, pays close attention to higher
The function of level.
Embodiment 3
In addition, to achieve the above object, present invention also offers a kind of computer-readable recording medium, the computer can
The networks congestion control data acquisition program that is stored with storage medium is read, the networks congestion control data acquisition is held by processor
The step of method of above-mentioned record network (web) user behavior data is realized during row：
S301, obtain network (web) user behavior data before, local cache preset a block type queue；
S302, acquisition network (web) user behavior data, the write-in of the network (web) user behavior data is local slow
The block type queue deposited；
S303, a default monitoring thread, and the block type in the local cache is monitored by the monitoring thread in real time
The length of queue；The maximum length threshold values of the block type queue in the local cache is preset, when the length of the block type queue
When degree is more than the maximum length threshold values, increase the number of threads；Block type queue in the default local cache is most
Small length threshold values, when the length of the block type queue is less than the minimum length threshold values, reduces the number of threads；
S304, by least one thread asynchronous operation by the network (web) user's row in the block type queue
Handled for data and be persisted to non-internal memory type database.
The method and its device, computer-readable medium of record networks congestion control data proposed by the present invention, realize note
Family operation behavior data are employed, write the high efficiency of data to improve Consumer's Experience by local cache, and pass through asynchronous thread
The processing to record data and record are carried out, while in order to prevent during high concurrent, may result in user behavior data generation number
Amount is more than processing quantity, and blocking so as to cause the queue of local cache to be booked generation, therefore add a monitoring thread to enter
Row intelligent control, reduces the possibility of queue obstruction.
It should be noted that herein, term " comprising ", "comprising" or its any other variant are intended to non-row
His property is included, so that process, method, article or device including a series of key elements not only include those key elements, and
And also including other key elements being not expressly set out, or also include for this process, method, article or device institute inherently
Key element.In the absence of more restrictions, the key element limited by sentence " including one ... ", it is not excluded that including
Also there is other identical element in process, method, article or the device of the key element.
The embodiments of the present invention are for illustration only, and the quality of embodiment is not represented.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can add the mode of required general hardware platform to realize by software, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Understood based on such, technical scheme is substantially done to prior art in other words
Going out the part of contribution can be embodied in the form of software product, and the computer software product is stored in a storage medium
In (such as ROM/RAM, magnetic disc, CD), including some instructions are to cause a station terminal equipment (can be mobile phone, computer, clothes
It is engaged in device, air conditioner, or network equipment etc.) perform method described in each embodiment of the invention.
The preferred embodiments of the present invention are these are only, are not intended to limit the scope of the invention, it is every to utilize this hair
Equivalent structure or equivalent flow conversion that bright specification and accompanying drawing content are made, or directly or indirectly it is used in other related skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of method for recording networks congestion control data, it is characterised in that the described method comprises the following steps：
Networks congestion control data are obtained, the networks congestion control data are write to the block type queue of local cache；
The networks congestion control data in the block type queue are handled by least one thread asynchronous operation
And it is persisted to non-internal memory type database.
2. a kind of method for recording networks congestion control data according to claim 1, it is characterised in that also including following
Step：Before networks congestion control data are obtained, a block type queue is preset in local cache.
3. a kind of method for recording networks congestion control data according to claim 1, it is characterised in that also including following
Step：
A default monitoring thread, and the length of the block type queue in the local cache is monitored by the monitoring thread in real time
Degree.
4. a kind of method for recording networks congestion control data according to claim 3, it is characterised in that also including following
Step：
The maximum length threshold values of the block type queue in the local cache is preset, when the length of the block type queue is more than institute
When stating maximum length threshold values, increase the number of threads.
5. a kind of method for recording networks congestion control data according to claim 3, it is characterised in that the record net
The method of network user behavior data, it is further comprising the steps of：
The minimum length threshold values of the block type queue in the local cache is preset, when the length of the block type queue is less than institute
When stating minimum length threshold values, the number of threads is reduced.
6. a kind of device for recording networks congestion control data, it is characterised in that described device includes：Memory, processor and
The data acquisition program that can be run on the memory and on the processor is stored in, the data acquisition program is described
The method and step of following record networks congestion control data is realized during computing device：
Networks congestion control data are obtained, the networks congestion control data are write to the block type queue of local cache；
The networks congestion control data in the block type queue are handled by least one thread asynchronous operation
And it is persisted to non-internal memory type database.
7. a kind of device for recording networks congestion control data according to claim 6, it is characterised in that the data are adopted
Collection program is also realized the method and step of following record networks congestion control data during the computing device：
Before networks congestion control data are obtained, a block type queue is preset in local cache；
And a default monitoring thread, and block type queue in the local cache is monitored by the monitoring thread in real time
Length.
8. a kind of device for recording networks congestion control data according to claim 6, it is characterised in that the data are adopted
Collection program is also realized the method and step of following record networks congestion control data during the computing device：
It is more than the maximum length threshold values of the block type queue in the default local cache in the length of the block type queue
When, increase the number of threads.
9. a kind of device for recording networks congestion control data according to claim 8, it is characterised in that the data are adopted
Collection program is also realized the method and step of following record networks congestion control data during the computing device：
It is less than the minimum length threshold values of the block type queue in the default local cache in the length of the block type queue
When, reduce the number of threads.
10. a kind of computer-readable recording medium, it is characterised in that the network that is stored with the computer-readable recording medium is used
Family behavioral data capture program, is realized when the networks congestion control data acquisition is executed by processor as in claim 1 to 5
The step of method of record networks congestion control data described in any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710285002.3A CN107122457A (en) | 2017-04-26 | 2017-04-26 | Record the method and its device, computer-readable medium of networks congestion control data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710285002.3A CN107122457A (en) | 2017-04-26 | 2017-04-26 | Record the method and its device, computer-readable medium of networks congestion control data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107122457A true CN107122457A (en) | 2017-09-01 |
Family
ID=59725039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710285002.3A Pending CN107122457A (en) | 2017-04-26 | 2017-04-26 | Record the method and its device, computer-readable medium of networks congestion control data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107122457A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107633087A (en) * | 2017-09-29 | 2018-01-26 | 南威软件股份有限公司 | A kind of date storage method based on database |
| CN107665243A (en) * | 2017-09-11 | 2018-02-06 | 安徽慧视金瞳科技有限公司 | A kind of across language environment data interactive method based on database access |
| CN107704605A (en) * | 2017-10-13 | 2018-02-16 | 北京中教在线科技有限公司 | It is a kind of based on the user behavior recording method being locally stored |
| CN109753479A (en) * | 2018-12-28 | 2019-05-14 | 杭州迪普科技股份有限公司 | A kind of data distributing method and device, equipment, medium |
| CN110221987A (en) * | 2019-05-05 | 2019-09-10 | 福建天泉教育科技有限公司 | Date storage method, storage medium in the case of system emergency |
| CN110413428A (en) * | 2019-08-06 | 2019-11-05 | 中国联合网络通信集团有限公司 | Processing method, electronic equipment and the storage medium of reconciliation data |
| CN111190899A (en) * | 2019-11-26 | 2020-05-22 | 泰康保险集团股份有限公司 | Buried point data processing method and device, server and storage medium |
| CN112445596A (en) * | 2020-11-27 | 2021-03-05 | 平安普惠企业管理有限公司 | Multithreading-based data import method and system and storage medium |
| WO2021121203A1 (en) * | 2019-12-17 | 2021-06-24 | 中兴通讯股份有限公司 | Method and apparatus for configuring service table, network device, and storage medium |
| CN113535721A (en) * | 2018-08-31 | 2021-10-22 | 创新先进技术有限公司 | Data writing method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110289097A1 (en) * | 2010-05-20 | 2011-11-24 | Salesforce.Com, Inc. | Metrics-based accessing of social threads |
| CN102346667A (en) * | 2011-09-19 | 2012-02-08 | 北京金和软件股份有限公司 | Realization method for automatically recording behavioral data of user |
| CN102622426A (en) * | 2012-02-27 | 2012-08-01 | 杭州闪亮科技有限公司 | Database writing system and database writing method |
| CN105681132A (en) * | 2016-02-26 | 2016-06-15 | 深圳市共进电子股份有限公司 | Method and device for collecting user internet-surfing behavior |
-
2017
- 2017-04-26 CN CN201710285002.3A patent/CN107122457A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110289097A1 (en) * | 2010-05-20 | 2011-11-24 | Salesforce.Com, Inc. | Metrics-based accessing of social threads |
| CN102346667A (en) * | 2011-09-19 | 2012-02-08 | 北京金和软件股份有限公司 | Realization method for automatically recording behavioral data of user |
| CN102622426A (en) * | 2012-02-27 | 2012-08-01 | 杭州闪亮科技有限公司 | Database writing system and database writing method |
| CN105681132A (en) * | 2016-02-26 | 2016-06-15 | 深圳市共进电子股份有限公司 | Method and device for collecting user internet-surfing behavior |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665243B (en) * | 2017-09-11 | 2021-03-09 | 安徽慧视金瞳科技有限公司 | Cross-language environment data interaction method based on database access |
| CN107665243A (en) * | 2017-09-11 | 2018-02-06 | 安徽慧视金瞳科技有限公司 | A kind of across language environment data interactive method based on database access |
| CN107633087A (en) * | 2017-09-29 | 2018-01-26 | 南威软件股份有限公司 | A kind of date storage method based on database |
| CN107704605A (en) * | 2017-10-13 | 2018-02-16 | 北京中教在线科技有限公司 | It is a kind of based on the user behavior recording method being locally stored |
| CN107704605B (en) * | 2017-10-13 | 2021-04-09 | 北京中教在线科技有限公司 | User behavior recording method based on local storage |
| CN113535721A (en) * | 2018-08-31 | 2021-10-22 | 创新先进技术有限公司 | Data writing method and device |
| CN109753479A (en) * | 2018-12-28 | 2019-05-14 | 杭州迪普科技股份有限公司 | A kind of data distributing method and device, equipment, medium |
| CN109753479B (en) * | 2018-12-28 | 2021-05-25 | 杭州迪普科技股份有限公司 | Data issuing method, device, equipment and medium |
| CN110221987A (en) * | 2019-05-05 | 2019-09-10 | 福建天泉教育科技有限公司 | Date storage method, storage medium in the case of system emergency |
| CN110413428A (en) * | 2019-08-06 | 2019-11-05 | 中国联合网络通信集团有限公司 | Processing method, electronic equipment and the storage medium of reconciliation data |
| CN111190899A (en) * | 2019-11-26 | 2020-05-22 | 泰康保险集团股份有限公司 | Buried point data processing method and device, server and storage medium |
| CN111190899B (en) * | 2019-11-26 | 2023-08-15 | 泰康保险集团股份有限公司 | Buried data processing method, buried data processing device, server and storage medium |
| WO2021121203A1 (en) * | 2019-12-17 | 2021-06-24 | 中兴通讯股份有限公司 | Method and apparatus for configuring service table, network device, and storage medium |
| CN112445596A (en) * | 2020-11-27 | 2021-03-05 | 平安普惠企业管理有限公司 | Multithreading-based data import method and system and storage medium |
| CN112445596B (en) * | 2020-11-27 | 2024-02-02 | 上海睿量私募基金管理有限公司 | Data importing method, system and storage medium based on multithreading |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107122457A (en) | Record the method and its device, computer-readable medium of networks congestion control data | |
| US11907237B2 (en) | Gathering and contributing content across diverse sources | |
| US10007933B2 (en) | Systems and methods for integrating dynamic content into electronic media | |
| US9081789B2 (en) | System for prefetching digital tags | |
| US8683379B2 (en) | Dynamic layout for a search engine results page based on implicit user feedback | |
| CN103412890B (en) | A kind of webpage loading method and device | |
| US7062756B2 (en) | Dynamic object usage pattern learning and efficient caching | |
| CN102576366B (en) | Pre-fetching content items based on social distance | |
| CN109902220B (en) | Webpage information acquisition method, device and computer readable storage medium | |
| CN104111847B (en) | A kind of method and device for improving game running speed | |
| JP2019514091A (en) | Optimizing user interface data caching for future actions | |
| CN106250464B (en) | Training method and device of ranking model | |
| CN104052809B (en) | A kind of flow-dividing control method and apparatus of website test | |
| CN108366096A (en) | A kind of information subscribing method, terminal and computer readable storage medium | |
| US9830304B1 (en) | Systems and methods for integrating dynamic content into electronic media | |
| CN107526718A (en) | Method and apparatus for generating text | |
| CN111259220A (en) | Data acquisition method and system based on big data | |
| CN110266555A (en) | Method for analyzing web site service request | |
| US20140317044A1 (en) | Method And Device for Real-Time Knowledge Processing Based on an Ontology With Temporal Extensions | |
| CN108205555A (en) | Information recommendation method, device, browser and terminal device | |
| Tsai et al. | Analysis of Application Data Mining to Capture Consumer Review Data on Booking Websites | |
| US10719202B2 (en) | System for dynamically rendering a graphical user interface | |
| da Cunha | Data analysis and recommender system architecture for e-commerce platforms | |
| US11438426B1 (en) | Systems and methods for intelligent session recording | |
| You et al. | DeepScraper: A complete and efficient tweet scraping method using authenticated multiprocessing |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170901 |