CN103152396A - Data placement method and device applied to content distribution network system - Google Patents

Data placement method and device applied to content distribution network system Download PDF

Info

Publication number
CN103152396A
CN103152396A CN2013100473458A CN201310047345A CN103152396A CN 103152396 A CN103152396 A CN 103152396A CN 2013100473458 A CN2013100473458 A CN 2013100473458A CN 201310047345 A CN201310047345 A CN 201310047345A CN 103152396 A CN103152396 A CN 103152396A
Authority
CN
China
Prior art keywords
data
server
edge server
data cell
cell
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
CN2013100473458A
Other languages
Chinese (zh)
Inventor
范冰冰
陈振洲
柳杨
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 Normal University
Original Assignee
South China Normal University
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 Normal University filed Critical South China Normal University
Priority to CN2013100473458A priority Critical patent/CN103152396A/en
Publication of CN103152396A publication Critical patent/CN103152396A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The invention belongs to the field of computer network and specifically discloses a data placement method and device applied to a content distribution network system. The data placement method comprises the steps of dividing data into a plurality of data groups and placing the data according to access frequency of a server in a main server. According to a network access accelerating method, common access objects are buffered in an edge server, and a data group in an original server is copied in the edge server and is converted into an object to be buffered in cache of the edge server through the data placement method, so that the access speed of the Internet is effectively increased.

Description

Be applied to data placement method and device in content distribution network system
Technical field
The invention belongs to computer network field, especially a kind of data placement method and apparatus that is applied in content distribution network system.
Background technology
Along with 3G, WiFi, the networks such as WiMax universal, and the continuous appearance (such as mobile phone reading, mobile TV, Mobile business etc.) of dazzling intelligent terminal and abundant dynamic application, this brings huge challenge for network of operator.According to the report of International Telecommunication Association's in January, 2012 issue, during the end of the year in 2011, whole world cellphone subscriber is 5,900,000,000, and the mobile broadband user is near 1,200,000,000.Show according to the report of in January, 2012 China inter network information center issue in Beijing, by in by the end of December, 2011, Chinese netizen's quantity breaks through 500,000,000, and wherein mobile phone netizen scale reaches 3.56 hundred million, increases by 17.5% on a year-on-year basis.
As shown in Figure 1, in traditional network service mode, user and content supplier are positioned at the two ends of service, and the Internet Service Provider connects both.The user utilizes the process of traditional network schemer access websites as follows:
At first, the user provides the domain name of wanting access websites to browser.Then, browser is resolved domain name to the name server request, to obtain IP address corresponding to this domain name.Then, browser uses resulting IP address, sends data access request to Website server.At last, the content of the data display web page returned according to Website server of browser.
After the user sends service request, need to carry out obtaining after domain name mapping through dns server the real IP of institute's access websites, then utilize this IP access websites.Under this access module, Website server is in Central Position, all visitors must directly be connected the ability access relevant content with Website server, this can bring following several problem: the first, because the open ended visit capacity of Website server is limited, if the website has been issued great or during focus message, will bring pouring in of a large number of users, the flow of website can sharply rise at short notice, causes accelerating website access to descend, even paralysis.The second, if Website server hypothesis from the far position of user, user access request just must through wide area network than multiple network node, cause long time-delay.The 3rd, when two users used the service that the heterogeneous networks service provider provides, these two users may have different services to experience (or quality).
In order to address these problems, scholars have proposed many methods.At present, CDN(content distributing network) technology is the most frequently used technology, is known about CND system and concept thereof.
Figure 2 shows that the structure chart of CDN network system.As shown in Figure 2, the CDN network mainly by original server with are connected the Edge Server that is connected with original server and consist of.Original server comprises that CDN network management center and GSLB DNS are redirected resolution system, are responsible for distribution and the management of whole CDN network.Edge Server mainly refers to the strange land distribution node, is comprised of load-balancing device, cache server two parts.CDN is by being published to web site contents near on user's fringe node, and making the user of different geographical is to obtain nearby in the access same page.So not only can alleviate the burden of source server, and can reduce the situation of flow distribution inequality in whole network, and then improve the performance of whole network.
Figure 3 shows that the browsing process of the system of CDN in Fig. 2, as shown in Figure 3, step 1: the user provides the domain name of wanting access websites to browser, and the request of domain name mapping is addressed to local dns and carries out domain name mapping; Step 2:CDN has carried out adjustment to the domain name mapping process, and what parsing obtained is another name CNAME record corresponding to this domain name, and in order to obtain real ip address, browser resolves to obtain actual fringe node IP address by the CNAME domain name; In this process, the GSLB dns resolution of use can according to geographical position and the actual network traffic information of calling party, make the user obtain the rational Edge Server IP of most probable address.Browser sends access request to Edge Server after obtaining actual IP address; Edge Server judges whether directly processing response according to the access request that browser provides, or by the original server Coordination Treatment.Edge Server will ask to preserve in this locality on the one hand, in order to using later on, on the other hand the data of obtaining be returned to client, complete the data, services process; The data that user side obtains being returned by Edge Server show later on and complete the whole data request process of browsing.
Access for the first time on principle, the Edge Server of CDN system will be got back to original server and obtain the Data Concurrent of user's request to the user.When having other users again to access same content, the CDN system will directly return to data the client, complete request/service process.We can find out by above analysis, when the CDN service provides acceleration to website visiting, can realize the generic access user transparent, namely add the later subscription client of buffer memory to need not to carry out any setting, directly the original domain name in the accelerated website of input is namely addressable.
In the CDN system, data placement is a more crucial problem, its target be with data placement among suitable Edge Server, in order to make the overall delay of user access request minimum.If user's request type comprises renewal seldom, the CDN system can copy data to Edge Server so, improves better the user and asks delay problem.But when if there are a large amount of update requests and a small amount of read request in data families (data, services), the scope that increases copy data is larger, will reduce systematic function.Because more new data may not be certain the Edge Server user who is accessing, access other Edge Servers but come from other user.Therefore, according to this situation, the most often data cell need to be arranged on the Edge Server that occurs to upgrade, and the master server of this Edge Server as this type of data family.
In addition, if when system load is relatively lower, data cell can be copied to many places in order to improve user's time delay.But, if when system load is higher, settle more data will bring larger overhead.Therefore, just need to reduce unnecessary copy this moment and settle, in order to reduce server load, improve inner time delay.
Therefore, a kind of suitable data placement method of the expectation that those skilled in the art are eager, the method copy data as required in Edge Server, can either satisfy the CDN system and have less time delay, can not bring larger load to server again.
Summary of the invention
For the deficiencies in the prior art, the objective of the invention is to provide the data placement that is applied in content distribution network system method, expectation the method can be made better balance in server load and Time Delay of Systems.
Based on this, first aspect present invention provides a kind of data placement method that is applied in content distribution network system, described content distribution network system comprises original server and is connected the Edge Server that is connected with described original server, and the method comprises the following steps:
Data in the database of master server are divided into several data cells { D 1, D 2, D 3..., D n;
With described data cell { D 1, D 2, D 3..., D nForm some data families;
Calculate respectively from the summation T of the read-write requests quantity of server,
T = Σ j = 1 m ( r i , j + w i , j ) ;
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
If from server to the data cell D described master server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated in this among server, wherein, and X 1It is a set point.
Accordingly, second aspect present invention provides a kind of data placement device that is applied in content distribution network system, comprising:
The data cell module is used for the data in the database of original server are divided into several data cells { D 1, D 2, D 3..., D n;
Data family module is used for basis to described data cell { D 1, D 2, D 3..., D nAccess stencil with described data cell { D 1, D 2, D 3..., D nForm some data families;
Computing module, for the summation T of the read-write requests quantity of calculating each Edge Server,
T = Σ j = 1 m ( r i , j + w i , j ) ;
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
Judge module is if be used for the judgement Edge Server to the data cell D of described original server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated among this Edge Server, wherein, and X 1It is a set point.
Below the invention will be further described.
All documents that the present invention quotes from, their full content is incorporated this paper by reference into, and, if when the implication that these documents are explained and the present invention are inconsistent, be as the criterion with statement of the present invention.In addition, various terms and phrase that the present invention uses have the general sense of well known to a person skilled in the art, nonetheless, the present invention still wishes at this, these terms and phrase to be described in more detail and to explain, the term of mentioning and phrase are as the criterion with the implication that the present invention was explained if any inconsistent with known implication.
Concrete, according to a kind of data placement method that is applied in content distribution network system that first aspect present invention provides, the method comprises the following steps:
Data in the database of master server are divided into several data cells { D 1, D 2, D 3..., D n;
With described data cell { D 1, D 2, D 3..., D nForm some data families;
Calculate respectively from the summation T of the read-write requests quantity of server,
T = Σ j = 1 m ( r i , j + w i , j ) ;
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
If from server to the data cell D described master server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated in this among server, wherein, and X 1It is a set point.The data access request that produces according to data families accounts for the proportion of read-write requests total amount, and the CDN system provides a plurality of different X in OK range 1Value, for example, described X 1Value can be 5,10,20,30 etc., then the different X of auto judge 1Time delay under value is big or small, selects the X of that minimum-time lag 1Value is selected corresponding master server for this data families at last.So not only computation complexity reduces greatly, and the system of being convenient to selects to settle strategy automatically, and can select different arrangement strategies according to user's access stencil.
Apparent, using phrase " master server " herein can be that original server can be also Edge Server, and " from server " is Edge Server.In case obtaining data family from original server, a certain Edge Server becomes " master server " of this data family, in afterwards access process, other Edge Server also might due to frequently from this master server this data of access families obtain this data families.Therefore, the method has not only effectively reduced system response interval, and does not substantially increase the load of original server.
The data placement method that provides according to first aspect present invention is with data cell { D 1, D 2, D 3..., D nBe divided into several data families and comprise the following steps:
Be provided for the characterization data cells D iThe space vector A of access stencil i, wherein:
A i={r i,1,r i,2···,r i,m,w i,1,···,w i,m}；
According to data cell D iSpace vector be A iWith data cell D jSpace vector A jThe calculated data cells D iWith data cell D jSimilarity Sim (i, j),
Sim ( i , j ) = Σ k = 1 2 * m ( a i , k - a ‾ i ) ( a j , k - a ‾ j ) Σ k = 1 2 * m ( a i , k - a ‾ i ) 2 Σ k = 1 2 * m ( a j , k - a ‾ j ) 2 ;
Wherein, (a i,1, a i,2..., a i,m)=(r i,1, r i,2..., r i,m);
(a i,m+1,a i,m+2,...,a i,2*m)=(w i,1,w i,2,...,w i,m)；
All data cell { D in the iterative data storehouse 1, D 2, D 3..., D n, if data cell D wherein iWith data cell D jBetween similarity greater than a set point X 2, with data cell D iWith data cell D jBe classified as same data families; Otherwise establishment comprises data cell D respectively iData families and comprise data cell D jData family.The data cell that the method will have a similarity method form is carried out cluster and is formed data families, be about to fine granularity (data cell) and change coarseness (data family) into, when carrying out data placement and process, the CDN system can go to process with the granularity (comprising one or more data cells) of data family, not namely that whole data set is copied on all Edge Servers, but each data family is only copied in its Edge Server of frequent access.
Accordingly, a kind of data placement device that is applied in content distribution network system according to second aspect present invention provides comprises:
The data cell module is used for the data in the database of original server are divided into several data cells { D 1, D 2, D 3..., D n;
Data family module is used for basis to described data cell { D 1, D 2, D 3..., D nAccess stencil with described data cell { D 1, D 2, D 3..., D nForm some data families;
Computing module, for the summation T of the read-write requests quantity of calculating each Edge Server,
T = Σ j = 1 m ( r i , j + w i , j ) ;
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
Judge module is if be used for the judgement Edge Server to the data cell D of described original server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated among this Edge Server, wherein, and X 1It is a set point.
According to the data placement device that second aspect present invention provides, described data families module comprises:
The first module is used for providing one to be used for the characterization data cells D iThe space vector A of access stencil i, wherein:
A i={r i,1,r i,2···,r i,m,w i,1,···,w i,m}；
Wherein, r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation;
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
The second module is used for according to data cell D iSpace vector be A iWith data cell D jSpace vector A jThe calculated data cells D iWith data cell D jSimilarity Sim (i, j), wherein,
Sim ( i , j ) = Σ k = 1 2 * m ( a i , k - a ‾ i ) ( a j , k - a ‾ j ) Σ k = 1 2 * m ( a i , k - a ‾ i ) 2 Σ k = 1 2 * m ( a j , k - a ‾ j ) 2
Wherein, (a i,1, a i,2..., a i,m)=(r i,1, r i,2..., r i,m);
(a i,m+1,a i,m+2,...,a i,2*m)=(w i,1,w i,2,...,w i,m)；
Data family judge module is used for the iterative data all data cell { D in storehouse 1, D 2, D 3..., D n, if data cell D wherein iWith data cell D jBetween similarity greater than a set point X 2, with data cell D iWith data cell D jBe classified as same data families.
In addition, a kind of internet access accelerator that is applied in content distribution network system that provides according to second aspect present invention, described content delivering system comprises an original server and is connected the Edge Server that is connected with described original server, and described internet access accelerator comprises
The cache lookup module is used for searching the object corresponding with described access request according to the access request of client in the buffer memory of described Edge Server, if find object corresponding to described access request, returns to the access request of client;
First searches module, is used for searching result set corresponding to described access request in the database of described Edge Server, if find described result set, returns to the access request of client;
Second searches device, is used for searching in other Edge Servers and/or described original server the data family that comprises described result set, and returns to the access request of client;
The data placement module is the data placement module that second aspect present invention provides, and is used for access frequency copy package according to described Edge Server and contains the data family of described result set to described Edge Server.
The present invention has also obtained following beneficial effect:
The present invention copies on the Edge Server of this data of frequent access family by accessing data frequently families, and convert thereof into the object that is stored in buffer memory, not only avoid original server and the excessive problem of Edge Server load, also effectively improved user's access speed.
Method and apparatus of the present invention is applied in existing various CDN network system, yet method and apparatus of the present invention more preferably is applicable among mobile network system.
Description of drawings
Accompanying drawing is described in further detail the specific embodiment of the present invention below in conjunction with accompanying drawing, wherein:
Fig. 1 is traditional network service module structural representation, and in figure, user and content supplier are positioned at the two ends of service, relies on the Internet Service Provider with user and content supplier;
Fig. 2 is the structure chart of CDN network system, in figure the CDN network mainly by original server with are connected the Edge Server that is connected with original server and consist of;
Fig. 3 is the browsing process figure of CDN network configuration;
Fig. 4 is the flow chart that the present invention is applied to the embodiment of the data placement method in the CDN network, according to from server, the access frequency of data families being settled data, has improved the load that has reduced server and has improved user's access speed;
Fig. 5 is the flow chart that in Fig. 4, data cell is divided into the method for data families, and it is processed according to the access stencil of data unit is divided into data families with data cell, has effectively improved the processing speed of data;
Fig. 6 is the flow chart that the present invention is applied to the enforcement of the internet access accelerated method in the CDN network system, becomes object to be cached among the buffer memory of Edge Server data transaction according to user's access frequency, thereby has effectively improved user's access speed;
Fig. 7 is the flow chart that the present invention is applied to the embodiment of lookup result collection in the CDN network system;
Fig. 8 is that the present invention is corresponding to the structural representation of the embodiment of the data placement device of Fig. 4;
Fig. 9 is the structural representation of data families module in Fig. 8;
Figure 10 is that the present invention is corresponding to the structural representation of the internet access accelerator of Fig. 6;
Figure 11 is the structural representation corresponding to the embodiment of the finger of Fig. 7.
Embodiment
Further illustrate the present invention below by object lesson, still, should be understood to, these examples are only used for the use that specifically describes more in detail, and should not be construed as for limiting in any form the present invention.
As shown in Figure 4, a kind of data placement method that is applied in content distribution network system, the method comprises the following steps:
Step S401: the data in the database of master server are divided into several data cells { D 1, D 2, D 3..., D n;
Step S402: with described data cell { D 1, D 2, D 3..., D nBe divided into some data families;
Step S403: calculate respectively from the summation T of the read-write requests quantity of server,
T = Σ j = 1 m ( r i , j + w i , j ) ;
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
Step S404: if from server to the data cell D described master server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated in this among server, wherein, and X 1It is a set point.
By said method, data are settled, not only computation complexity reduces greatly, and the system of being convenient to selects to settle strategy automatically, and can select different arrangement strategies according to user's access stencil.
About data cell being divided into the method for several data family, those skilled in the art can adopt the whole bag of tricks to divide, yet, preferably adopt following steps with data cell { D 1, D 2, D 3..., D nBe divided into several data families, referring to Fig. 5, described step is as follows:
Step S501: provide one to be used for the characterization data cells D iThe space vector A of 2*m dimension of access stencil i, wherein:
A i={r i,1,r i,2···,r i,m,w i,1,···,w i,m}；
Step S502: calculated data cells D iSpace vector be A iWith data cell D jSpace vector A jThe calculated data cells D iWith data cell D jSimilarity Sim (i, j),
Sim ( i , j ) = Σ k = 1 2 * m ( a i , k - a ‾ i ) ( a j , k - a ‾ j ) Σ k = 1 2 * m ( a i , k - a ‾ i ) 2 Σ k = 1 2 * m ( a j , k - a ‾ j ) 2
Wherein, (a i,1, a i,2..., a i,m)=(r i,1, r i,2..., r i,m);
(a i,m+1,a i,m+2,...,a i,2*m)=(w i,1,w i,2,...,w i,m)；
Step S503: all data cell { D in the iterative data storehouse 1, D 2, D 3..., D n, if data cell D wherein iWith data cell D jBetween similarity greater than a set point X 2, with data cell D iWith data cell D jBe classified as same data families; Otherwise establishment comprises data cell D respectively iData families and comprise data cell D jData family.
In addition, the present invention also provides a kind of internet access accelerated method that is applied in content distribution network system.Total institute is known, and described content delivering system comprises original server and are connected the Edge Server that is connected with described original server, and therefore, referring to Fig. 6, this internet access accelerated method comprises the following steps,
Step S601: search the object corresponding with described access request in the buffer memory of access request in described Edge Server according to client, if find object corresponding to described access request, return to the access request of client;
Step S602: otherwise, search result set corresponding to described access request in database, if find described result set, return to the access request of client;
Step S603: otherwise, to the request that other Edge Servers or described original server send visit data, search the data family that comprises described result set in described original server, and return to the access request of client;
Described original server also the above data placement method copy package logical according to the access frequency of described Edge Server contains the data family of described result set to described Edge Server.
In order further to improve the visit data of the Internet, described Edge Server also is converted to described result set the object in the buffer memory that is stored in described Edge Server, and upgrades the object in described buffer memory, to improve the probability of hit access request in buffer memory.
In addition, those skilled in the art also can adopt the whole bag of tricks to search data in original server, and are preferred, can be shown in Figure 7, and described Edge Server is searched described result set from described original server by following steps:
Step S701: set up in Edge Server:
Store the data families member relation table of the identification information of the data cell in each data family;
Figure BDA00002819706800122
The data families attribute list that comprises each data family information;
Step S702: search the data family of this data cell according to data families member relation table, and search the position of original server corresponding to this data families according to the data family that finds from described data families attribute list;
Step S703: read this data cell and return to client-requested from described original server.
Correspondingly, the present invention also provides a kind of data placement device that is applied in content distribution network system, and is shown in Figure 8, and this data placement device comprises:
Data cell module 81 is used for the data in the database of original server are divided into several data cells { D 1, D 2, D 3..., D n;
Data family module 82 is used for basis to described data cell { D 1, D 2, D 3..., D nAccess stencil with described data cell { D 1, D 2, D 3..., D nBe divided into some data families;
Computing module 83, for the summation T of the read-write requests quantity of calculating each Edge Server,
T = Σ j = 1 m ( r i , j + w i , j ) ;
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
Judge module 84 is if be used for the judgement Edge Server to the data cell D of described original server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated among this Edge Server, wherein, and X 1It is a set point.
Wherein, preferred, as shown in Figure 9, the data families module that data cell is divided into some data families comprises:
The first module 91 is used for providing one to be used for the characterization data cells D iThe space vector A of 2*m dimension of access stencil i, wherein:
A i={r i,1,r i,2···,r i,m,w i,1,···,w i,m}；
Wherein, r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation;
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
The second module 92 is used for the calculated data cells D iSpace vector be A iWith data cell D jSpace vector A jThe calculated data cells D iWith data cell D jSimilarity Sim (i, j), wherein,
Wherein, (a i,1, a i,2..., a i,m)=(r i,1, r i,2..., r i,m);
(a i,m+1,a i,m+2,...,a i,2*m)=(w i,1,w i,2,...,w i,m)；
Data family judge module 93 is used for the iterative data all data cell { D in storehouse 1, D 2, D 3..., D n, if data cell D wherein iWith data cell D jBetween similarity greater than a set point X, with data cell D iWith data cell D jBe classified as same data families, otherwise, create respectively and comprise data cell D iData families and comprise data cell D jData family.
Accordingly, the present invention also provides a kind of internet access accelerator that is applied in content distribution network system, and described content delivering system comprises an original server and is connected the Edge Server that is connected with described original server.As shown in figure 10, described internet access accelerator comprises,
Cache lookup module 101 is used for searching the object corresponding with described access request according to the access request of client in the buffer memory of described Edge Server, if find object corresponding to described access request, returns to the access request of client;
First searches module 102, is used for searching result set corresponding to described access request in the database of described Edge Server, if find described result set, returns to the access request of client;
Second searches device 103, is used for the request of sending visit data to other Edge Servers or described original server, searches the data family that comprises described result set in described original server, and returns to the access request of client;
Data placement module 104 is above-mentioned data placement module, is used for access frequency copy package according to described Edge Server and contains the data family of described result set to described Edge Server.
Preferably, described Edge Server also is converted to described result set the object in the buffer memory that is stored in described Edge Server, and upgrades the object in described buffer memory.
As shown in figure 11, comprise that also described finger comprises for search the finger of described result set from described original server:
Form module 111 is used for setting up in Edge Server:
Store the data families member relation table of the identification information of the data cell in each data family;
Figure BDA00002819706800142
The data families attribute list that comprises each data family information;
Search module 112, be used for searching according to data families member relation table the data family of this data cell, and search the position of original server corresponding to this data families according to the data family that finds from described data families attribute list;
Return to module 113, be used for reading this data cell and returning to client-requested from described original server.
The above has done detailed description to embodiments of the present invention by reference to the accompanying drawings, but the present invention is not limited to above-mentioned execution mode, in the ken that one skilled in the relevant art possesses, can also make various variations under the prerequisite that does not break away from aim of the present invention.

Claims (4)

1. a data placement method that is applied in content distribution network system, is characterized in that, the method comprises the following steps:
Data in the database of master server are divided into several data cells { D 1, D 2, D 3..., D n;
With described data cell { D 1, D 2, D 3..., D nForm some data families;
Calculate respectively from the summation T of the read-write requests quantity of server,
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
If from server to the data cell D described master server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated in this among server, wherein, and X 1It is a set point.
2. the data placement method that is applied in content distribution network system as claimed in claim 1, is characterized in that, with data cell { D 1, D 2, D 3..., D nForm several data families and comprise the following steps:
Be provided for the characterization data cells D iThe space vector A of access stencil i, wherein:
A i={r i，1,r i,2…,r i,m,w i，1,…,w i，m}；
According to data cell D iSpace vector be A iWith data cell D jSpace vector A jThe calculated data cells D iWith data cell D jSimilarity Sim (i, j),
Figure FDA00002819706700012
Wherein, (a i,1, a i,2..., a i,m)=(r i,1, r i,2..., r I, m);
(a i,m+1,a i,m+2,...,a i,2*m)=(w i,1,w i,2,...,w i，m)；
All data cell { D in the iterative data storehouse 1, D 2, D 3..., D n, if data cell D wherein iWith data cell D jBetween similarity greater than a set point X 2, with data cell Di and data cell D jBe classified as same data families.
3. a data placement device that is applied in content distribution network system, is characterized in that, comprising:
The data cell module is used for the data in the database of original server are divided into several data cells { D 1, D 2, D 3..., D n;
Data family module is used for basis to described data cell { D 1, D 2, D 3..., D nAccess stencil with described data cell { D 1, D 2, D 3..., D nForm some data families;
Computing module, for the summation T of the read-write requests quantity of calculating each Edge Server,
Figure FDA00002819706700021
Wherein, m is the quantity of described Edge Server,
r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation,
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
Judge module is if be used for the judgement Edge Server to the data cell D of described original server iAccess number be greater than or equal to T* (X 1/ 100), and r iBe greater than or equal to w i, with data cell D iThe data family at place is replicated among this Edge Server, wherein, and X 1It is a set point.
4. the data placement device that is applied in content distribution network system as claimed in claim 3, is characterized in that, described data families module comprises:
The first module is used for providing one to be used for the characterization data cells D iThe space vector A of access stencil i, wherein:
A i={r i,1,r i,2···,r i,m,w i,1,···,w i,m}；
Wherein, r iFor the user by Edge Server to the data cells D iCarry out the quantity of read operation;
w iFor the user by Edge Server to the data cells D iCarry out the quantity of write operation;
The second module is used for according to data cell D iSpace vector be A iWith data cell D jSpace vector.
CN2013100473458A 2013-02-05 2013-02-05 Data placement method and device applied to content distribution network system Pending CN103152396A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2013100473458A CN103152396A (en) 2013-02-05 2013-02-05 Data placement method and device applied to content distribution network system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2013100473458A CN103152396A (en) 2013-02-05 2013-02-05 Data placement method and device applied to content distribution network system

Publications (1)

Publication Number Publication Date
CN103152396A true CN103152396A (en) 2013-06-12

Family

ID=48550255

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2013100473458A Pending CN103152396A (en) 2013-02-05 2013-02-05 Data placement method and device applied to content distribution network system

Country Status (1)

Country Link
CN (1) CN103152396A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103973788A (en) * 2014-05-08 2014-08-06 浪潮电子信息产业股份有限公司 Load balancing method based on transmission widespread network architecture
CN104065543A (en) * 2014-05-19 2014-09-24 浪潮电子信息产业股份有限公司 Network card drive design method based on data transmission management
CN105407005A (en) * 2015-12-09 2016-03-16 中国联合网络通信集团有限公司 Content distribution method and device
CN106790552A (en) * 2016-12-23 2017-05-31 航天星图科技（北京）有限公司 A kind of content providing system based on content distributing network
CN108600777A (en) * 2018-04-09 2018-09-28 北京大米科技有限公司 A kind of stream media playback server, client
CN108737372A (en) * 2018-04-09 2018-11-02 北京大米科技有限公司 A kind of stream media playback method, computer equipment and readable medium
CN110213774A (en) * 2019-05-23 2019-09-06 超讯通信股份有限公司 A kind of 5G network automatic evaluation system neural network based, method and device
CN112291381A (en) * 2017-11-13 2021-01-29 华为技术有限公司 Application server switching method, device and system
CN113301184A (en) * 2021-07-08 2021-08-24 凌锐蓝信科技(北京)有限公司 Remote access method, device, computer equipment and storage medium
US11671373B2 (en) 2019-07-31 2023-06-06 Huawei Technologies Co., Ltd. Systems and methods for supporting traffic steering through a service function chain

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1481635A (en) * 2000-10-26 2004-03-10 普里斯梅迪亚网络有限公司 Method and appts. for large payload distribution in network
CN101557499A (en) * 2008-04-09 2009-10-14 中兴通讯股份有限公司 System and method for delivering media and method for renewing resource of media delivering system
CN102752381A (en) * 2012-06-28 2012-10-24 北京邮电大学 Multi-movable-duplicate mechanism applied to distributed storage and access method thereof

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1481635A (en) * 2000-10-26 2004-03-10 普里斯梅迪亚网络有限公司 Method and appts. for large payload distribution in network
CN101557499A (en) * 2008-04-09 2009-10-14 中兴通讯股份有限公司 System and method for delivering media and method for renewing resource of media delivering system
CN102752381A (en) * 2012-06-28 2012-10-24 北京邮电大学 Multi-movable-duplicate mechanism applied to distributed storage and access method thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
柳杨，范冰冰: "《一种通用的动态Web应用加速方案》", 《计算机工程》 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103973788A (en) * 2014-05-08 2014-08-06 浪潮电子信息产业股份有限公司 Load balancing method based on transmission widespread network architecture
CN104065543A (en) * 2014-05-19 2014-09-24 浪潮电子信息产业股份有限公司 Network card drive design method based on data transmission management
CN105407005A (en) * 2015-12-09 2016-03-16 中国联合网络通信集团有限公司 Content distribution method and device
CN105407005B (en) * 2015-12-09 2018-09-14 中国联合网络通信集团有限公司 Content distribution method and device
CN106790552B (en) * 2016-12-23 2019-04-09 中科星图股份有限公司 A kind of content providing system based on content distributing network
CN106790552A (en) * 2016-12-23 2017-05-31 航天星图科技（北京）有限公司 A kind of content providing system based on content distributing network
CN112291381A (en) * 2017-11-13 2021-01-29 华为技术有限公司 Application server switching method, device and system
CN112291381B (en) * 2017-11-13 2022-01-21 华为技术有限公司 Application server switching method, device and system
CN108737372A (en) * 2018-04-09 2018-11-02 北京大米科技有限公司 A kind of stream media playback method, computer equipment and readable medium
CN108600777A (en) * 2018-04-09 2018-09-28 北京大米科技有限公司 A kind of stream media playback server, client
CN110213774A (en) * 2019-05-23 2019-09-06 超讯通信股份有限公司 A kind of 5G network automatic evaluation system neural network based, method and device
US11671373B2 (en) 2019-07-31 2023-06-06 Huawei Technologies Co., Ltd. Systems and methods for supporting traffic steering through a service function chain
US12088501B2 (en) 2019-07-31 2024-09-10 Huawei Technologies Co., Ltd. Systems and methods for supporting traffic steering through a service function chain
CN113301184A (en) * 2021-07-08 2021-08-24 凌锐蓝信科技(北京)有限公司 Remote access method, device, computer equipment and storage medium
CN113301184B (en) * 2021-07-08 2021-10-26 凌锐蓝信科技(北京)有限公司 Remote access method, device, computer equipment and storage medium

Similar Documents

Publication Publication Date Title
CN103139301B (en) Be applied to the internet access accelerated method in content distribution network system and device
CN103152396A (en) Data placement method and device applied to content distribution network system
CN103428267B (en) A kind of wisdom caching system and the method distinguishing user preferences dependency thereof
US9614889B2 (en) Autonomic content load balancing
CN106202581B (en) A kind of data search method and device
CN109542613A (en) Distribution method, device and the storage medium of service dispatch in a kind of CDN node
US20160335243A1 (en) Webpage template generating method and server
WO2009062424A1 (en) Method and system for search
CN103297472A (en) Redirection method and content distribution node applied to content distribution network
CN105930528A (en) Webpage cache method and server
EP3724805B1 (en) Enhanced online privacy
CN102624914B (en) Method for detecting local DNS (Domain Name Server) used by client side in Web form
CN107241372A (en) Configuration information generation, sending method and resource loading method and apparatus and system
CN102638483A (en) Method, device and system for determining content distribution nodes
Ma et al. An improved web cache replacement algorithm based on weighting and cost
CN103957282B (en) Terminal user's domain name mapping acceleration system and its method in a kind of domain
US11416291B1 (en) Database server management for proxy scraping jobs
CN104935653A (en) Bypass cache method for visiting hot spot resource and device
CN104021137B (en) A kind of client based on catalogue mandate is locally opened and closed the method and system of file
CN106210022A (en) A kind of system and method for processing forum's height concurrent data requests
CN109451089A (en) A kind of access method, device, computer equipment and computer storage medium
CN113127420A (en) Metadata request processing method, device, equipment and medium
CN103107944A (en) Content locating method and route equipment
CN103825916A (en) Resource downloading method and resource downloading system
CN101751408A (en) System and method of internet in text advertisements

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20130612