CN107682416A - Mist computing architecture content collaboration distribution method and application system based on broadcast and storage network - Google Patents
Mist computing architecture content collaboration distribution method and application system based on broadcast and storage network Download PDFInfo
- Publication number
- CN107682416A CN107682416A CN201710851632.2A CN201710851632A CN107682416A CN 107682416 A CN107682416 A CN 107682416A CN 201710851632 A CN201710851632 A CN 201710851632A CN 107682416 A CN107682416 A CN 107682416A
- Authority
- CN
- China
- Prior art keywords
- content
- mist
- node
- server
- broadcast
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- H04L67/55—Push-based network services
-
- 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
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1023—Server selection for load balancing based on a hash applied to IP addresses or costs
-
- 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
- H04L67/56—Provisioning of proxy services
- H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a kind of mist computing architecture content collaboration distribution method and application system based on broadcast and storage network.The present invention is lifted the content active push ability of traditional mist computing architecture using broadcast and storage network and reduces its consumption to Internet bandwidth, using software defined network SDN centralized Control thought, propose to take into account more granularities collaboration memory mechanism of broadcasted content and Web cache contents；On this basis, by controller come all nodes in centralized maintenance and control domain, and efficient content forwarding strategy is formulated for domain intrinsic fog server；Finally, the collaboration distribution of the content of more nodes in domain is realized by controller.The present invention can effectively strengthen the content active push ability of mist computing architecture, reduce user and ask time delay, lift Consumer's Experience.
Description
Technical field
The present invention relates to a kind of mist computing architecture content collaboration distribution method and application system based on broadcast and storage network, can subtract
The response delay of few end-user request, lifts Consumer's Experience, belongs to Internet technical field.
Background technology
Cloud computing can overcome in traditional application system that generally existing resource is exclusive, data center traffic density is low and system
The disadvantages such as resource utilization is low, but it still has and can not effectively support delay-sensitive application and consume a large amount of network bandwidths etc.
Defect.Therefore, Cisco (Cisco) proposes the concept that mist calculates (Fog Computing), the network characterized by cloud computing
Calculate and expand to network edge from network center, and the intermediate state calculated between cloud computing and terminal is described with mist calculating.Mist
Computing architecture is that the solution calculated based on mist defines an open system level framework, can be in the data along cloud to thing
Generating source such as is provided about calculating, store, control and network the function.The it is proposed that mist calculates, fully shows the important of edge calculations
Property.However, in current mist computing architecture, the main progress content distribution using edge cache in the form of, be severely limited by by
The form of dynamic " user pulls ", lack the ability of content active push.
Broadcast and storage network is a kind of new network for supporting the Web content whole people to share, and its basic thought is will be from by broadcast
Network edge is arrived in the hot content and its corresponding content metadata crawled in internet, active push and storage, facilitates user
Terminal directly obtains nearby.Broadcast and storage network can lift the access to content experience of user, share the backbone traffic of internet, and
It is and strong with active push dynamics and content distribution does not consume the characteristics such as Internet bandwidth.Deposited in view of current mist computing architecture
Active push ability is weak, content distribution consumption Internet bandwidth the shortcomings of, it is difficult to overcome these shortcomings using conventional art, and
Advantage possessed by broadcast and storage network can effectively overcome these shortcomings.But one side industry currently is still not based on broadcasting depositing net
The mist computing architecture implementation of network；On the other hand, many difficulties are also had based on broadcast and storage network structure mist computing architecture, such as
Storage collaboration, the efficient management of mist server between Web cache contents and broadcasted content link interior with control, multiple nodes
Hold collaboration distribution method etc..The thought of broadcast and storage network is dissolved into mist computing architecture by the present invention, is designed one kind and is based on broadcasting depositing
The mist computing architecture of network, and the new mist computing architecture is based on, using software defined network SDN centralized Control theory, carry
Go out a kind of more granularity contents collaboration distribution method for taking into account broadcasted content and Web cache contents, realize the content of " push-and-pull combination "
Push.
The content of the invention
Goal of the invention：For problems of the prior art and deficiency, the present invention provides a kind of based on broadcast and storage network
Mist computing architecture content cooperates with distribution method and application system.This method uses the broadcast and storage network of " broadcast distribution+ubiquitous storage "
To aid in mist computing architecture to carry out the active push of content, thought on this basis using software defined network SDN centralized Control
Think, realize the efficient collaboration distribution of content, specifically include the leading information processing of more granularity contents collaboration storage, controller, with
And more node collaboration distributions etc. in domain.This method can reduce consumption of the mist computing architecture to Internet bandwidth, reduce edge and use
The request time delay at family, lift Consumer's Experience.
Technical scheme：A kind of mist computing architecture content collaboration distribution method based on broadcast and storage network, it is different from existing method
, this method lifted using broadcast and storage network traditional mist computing architecture content active push ability and reduce its to internet
The consumption of bandwidth, and using software defined network SDN centralized Control thought, realize broadcasted content distribution and Web cache contents
The efficient collaboration of distribution.The present invention is broadly divided into 3 steps：
Step 1, more granularity content collaboration storages.First, the thought of broadcast and storage network is dissolved into mist computing architecture, built
Mist computing architecture based on broadcast and storage network, the framework include mist server, controller, multiple user terminals and internet content
Acquisition server and broadcast content source, multiple mist servers carry out centralized Control by controller, and whole framework uses broadcasted content
Source to be gathered service by internet content in full to multiple mist servers distribution content full text and its corresponding UCL, these contents
Device collects from internet, and the UCL as corresponding to the generation of internet content acquisition server, while user terminal also receives
The UCL of broadcast distribution.Here UCL is the metadata of description content, the main title including content, summary, keyword and
The information such as content generic.Two kinds of storage modes are included in the mist computing architecture simultaneously：Web is cached and broadcasted content storage.
It (is usually the constitution element in webpage, such as picture embedded in webpage, music that Web cachings, which are responsible for the fine-grained content of storage,
Deng), and broadcasted content storage is then responsible for the thicker web page contents of storage granularity in full.In the mist computing architecture based on broadcast and storage network
In, while two kinds of varigrained content storage modes are safeguarded, two kinds in full of webpage constitution element and web page contents are realized with this
The collaboration storage of granularity content.First hot value is selected in broadcasted content storage be more than given threshold value PsWeb page contents in full,
And web page contents are parsed to obtain fine-grained web page element in full, then select wherein all with white list (system
Manager customization website list) matching web page element, and store to Web caching in.The hot value of web page contents full text is pressed
Calculated according to formula (1)：
Wherein, PiThe temperature of the web page contents full text within i-th of time-count cycle is represented, and meets P1=0；NiRepresent
The number that web page contents are accessed in full in i-th of time-count cycle；NiWeight coefficient a>1.
Step 2, the leading information processing of controller.Mist computing architecture based on broadcast and storage network is made up of multiple domains, every
A controller is configured in individual domain, multiple mist servers in its management and control domain.When increase, deletion occur for the content in mist server
During with renewal, mist server all can be to the corresponding modification information of controller report；Once there is terminal request to reach mist server, mist
Server also can be to the specifying information of this request of controller report.The information that controller reports according to each mist server, move
State adjusts index information table and service information table in this domain that it is recorded.It is therein index information table mainly include node,
The information such as port, URL and hash value, service information table mainly include source mark, node identification, hash value, type, hit-count
Etc. information.
Controller periodically calculates the class weight ratio of each mist server according to service information table, and is each
Mist server safeguards a class weight ratio information table, and broadcasted content and UCL foundation are received next time as the mist server.
User asks the content corresponding to UCL by UCL.It is assumed that the classification sum of all the elements is Nc, then any classification c weighted value
Calculated according to formula (2)：
Wherein RcRepresent in current period, the number that the UCL requests that user sends are classification c, andFor this cycle
The sum of interior all user UCL requests.Then class weight ratio is
Controller can be also worth by calculating the forwarding of each user UCL requests, to determine to forward in the domain of mist server
Strategy.Forwarding value is calculated according to formula (3)：
F=α W+ β H+ (1- alpha-betas) R formula (3)
WhereinAnd W, H and R weight that to be controller calculate respectively according to required parameter, section
Point hop count and hit rate, x and y are the constant parameter for characterizing tri- factor proportions of W, H, R, and specific value can be according to reality
Demand dynamic adjusts, but must is fulfilled for condition x+y≤0.
Step 3, multi-node collaborative is distributed in domain.Mist computing architecture based on broadcast and storage network uses centerized fusion thought,
To realize the collaboration of domain interior nodes, this causes mist server itself to safeguard the data message and control information of complexity, and
Only need to be responsible for realizing service logic, therefore compared to complete distributed treatment more quickly and efficiently.When mist server local
When being stored with the content resource of user's request, content requests of the mist server directly in response to terminal user；If mist server local
When not storing requested content resource, mist server directly sends collaborative query in a domain to controller and asked.Control
When device receives the content query requests that mist server is sent ,-in the controller institute compass of competency and being stored with requested
Other mist servers of content, it is worth by calculating these mist servers for the forwarding of user UCL requests, selection forwarding value
Highest mist server returns to requesting node as optimal forward node, and by optimal node information.Requesting node receives control
After the information that device processed returns, content requests are directly forwarded to IP and the port corresponding to optimal node, obtain content resource simultaneously
Return to terminal user.
A kind of content collaboration distribution application system of the mist computing architecture based on broadcast and storage network for performing the above method, mainly
Include a lower node：Hot content is gathered from internet and carries out the crawler server of content metadata UCL indexes, carries out heat
The broadcasting server of door content broadcast, mist server node (node identification is respectively A, B, C), controller node, and client
Hold (node identification D).Crawler server gathers popular webpage from internet, and carries out semantic parsing to these popular webpages
Corresponding UCL is obtained, popular webpage and UCL are all then transmitted to broadcasting server；Broadcasting server leads to popular webpage and UCL
Broadcast is crossed, radiation is distributed to each mist server.
Mist server is first according to formula (1) and calculates each popular webpage for the popular webpage arrived by broadcast reception
Hot value Pi.Next, it is determined that PiWhether condition is met：Pi≥Ps, if satisfied, then carrying out web page element solution to the popular webpage
Analysis, obtains fine-grained web page element, then selects by wherein all web page elements with white list matching.Then, it is right
In selecting next web page element, hash value is asked to the original URL of web page element using SHA1 algorithms.Finally, will according to hash value
Corresponding catalogue of the web page element storage to Web cachings.
Each mist server, can be actively to controller when receiving the popular webpage and UCL that broadcasting server is sent
Report corresponding information.And UCL content metadatas information will be illustrated in user terminal browser interface, user is accessed by clicking on UCL
Corresponding webpage full text or web page element, it is assumed that these contents are stored in each mist server, once there is user to ask to reach
Mist server, mist server all can be to the hit situations of the controller report request.The content stored in mist server such as increase,
Delete or update, mist server also can be to the corresponding modification information of controller report.And controller believes the index safeguarded in domain
Cease table and service information table.
Controller can calculate each of content in each mist server according to formula (2) within each time-count cycle
Classification c weighted value Wc, and then try to achieve class weight ratio
(3) multi-node collaborative is distributed in domain.If the corresponding classification 1 of UCL requests sent by user, corresponding to URL
Hash value is 4fghadf234hg23bc.UCL requests reach one of mist server A, and are not ordered in mist server A
In, then mist server A inquiry request into controller sending domain.
When controller receives content query requests in the domain that mist server A is sent, first according to hash value from node identification table
In find the set Node={ Node of all nodes for being stored with requested content1,Node2,...,Nodei}.Then, for
Each node Node in set Nodei, Node is obtained from class weight table according to the class label of contentiClass weight Wi,
The hop count H between node is obtained from topology information table according to node identificationi, and Node is calculated according to service information tableiAsked
Seek the hit rate R in classificationi, Node is finally calculated according to formula (3)iForwarding value Fi.When all nodes in set Node
After forwarding value calculation out, controller sorts these nodes according to forwarding value from high to low, picks out before coming most
The node in face returns to requesting node as optimal forward node, and by optimal forward node information.User receives controller and returned
After the information returned, UCL requests are directly sent to optimal forward node, last optimal forward node is right by user UCL requests institute
The web page contents answered are in full or web page element returns to terminal user.
Beneficial effect：Compared with prior art, the mist computing architecture content collaboration provided by the invention based on broadcast and storage network
Distribution method and application system, have the following advantages that：
1. the broadcast and storage network that active push dynamics is strong, Consumer's Experience is good is dissolved into mist computing architecture, using " broadcast
Distribution+ubiquitous storage " aids in the mist computing architecture to carry out the active push of content, can be effectively improved including traditional mist computing architecture
Hold the shortcoming in terms of active push ability, and consumption of the mist computing architecture to Internet bandwidth can be reduced.
2. software defined network SDN centralized Control thought is combined with mist computing architecture, realize broadcasted content distribution and
The efficient collaboration of Web cache contents distribution, the complexity of content collaboration distribution is simplified, improves the distribution of mist computing architecture
Can, and Consumer's Experience can be lifted.
Brief description of the drawings
Fig. 1 is the mist calculating support composition based on broadcast and storage network.
Embodiment
With reference to specific embodiment, the present invention is furture elucidated, it should be understood that these embodiments are merely to illustrate the present invention
Rather than limitation the scope of the present invention, after the present invention has been read, various equivalences of the those skilled in the art to the present invention
The modification of form falls within the application appended claims limited range.
Mist computing architecture content collaboration distribution method based on broadcast and storage network, unlike existing method, this method profit
The content active push ability of traditional mist computing architecture is lifted with broadcast and storage network and reduces its consumption to Internet bandwidth, and
Using software defined network SDN centralized Control thought, broadcasted content distribution and the efficient association that Web cache contents are distributed are realized
Together.The present invention is broadly divided into 3 steps：
Step 1, more granularity content collaboration storages.First, the thought of broadcast and storage network is dissolved into mist computing architecture, built
Mist computing architecture based on broadcast and storage network, the framework include mist server, controller, multiple user terminals and internet content
Acquisition server and broadcast content source, multiple mist servers carry out centralized Control by controller, and whole framework uses broadcasted content
Source to be gathered service by internet content in full to multiple mist servers distribution content full text and its corresponding UCL, these contents
Device collects from internet, and the UCL as corresponding to the generation of internet content acquisition server, while user terminal also receives
The UCL of broadcast distribution.Here UCL is the metadata of description content, the main title including content, summary, keyword and
The information such as content generic.Two kinds of storage modes are included in the mist computing architecture simultaneously：Web is cached and broadcasted content storage.
It (is usually the constitution element in webpage, such as picture embedded in webpage, music that Web cachings, which are responsible for the fine-grained content of storage,
Deng), and broadcasted content storage is then responsible for the thicker web page contents of storage granularity in full.In the mist computing architecture based on broadcast and storage network
In, while two kinds of varigrained content storage modes are safeguarded, two kinds in full of webpage constitution element and web page contents are realized with this
The collaboration storage of granularity content.First hot value is selected in broadcasted content storage be more than given threshold value PsWeb page contents in full,
And web page contents are parsed to obtain fine-grained web page element in full, then select wherein all with white list (system
Manager customization website list) matching web page element, and store to Web caching in.The hot value of web page contents full text is pressed
Calculated according to formula (1)：
Wherein, PiThe temperature of the web page contents full text within i-th of time-count cycle is represented, and meets P1=0；NiRepresent
The number that web page contents are accessed in full in i-th of time-count cycle；NiWeight coefficient a>1.
Step 2, the leading information processing of controller.Mist computing architecture based on broadcast and storage network is made up of multiple domains, every
A controller is configured in individual domain, multiple mist servers in its management and control domain.When increase, deletion occur for the content in mist server
During with renewal, mist server all can be to the corresponding modification information of controller report；Once there is terminal request to reach mist server, mist
Server also can be to the specifying information of this request of controller report.The information that controller reports according to each mist server, move
State adjusts index information table and service information table in this domain that it is recorded.It is therein index information table mainly include node,
The information such as port, URL and hash value, service information table mainly include source mark, node identification, hash value, type, hit-count
Etc. information.
Controller periodically calculates the class weight ratio of each mist server according to service information table, and is each
Mist server safeguards a class weight ratio information table, and broadcasted content and UCL foundation are received next time as the mist server.
User asks the content corresponding to UCL by UCL.It is assumed that the classification sum of all the elements is Nc, then any classification c weighted value
Calculated according to formula (2)：
Wherein RcRepresent in current period, the number that the UCL requests that user sends are classification c, andFor this cycle
The sum of interior all user UCL requests.Then class weight ratio is
Controller can be also worth by calculating the forwarding of each user UCL requests, to determine to forward in the domain of mist server
Strategy.Forwarding value is calculated according to formula (3)：
F=α W+ β H+ (1- alpha-betas) R formula (3)
WhereinAnd W, H and R weight that to be controller calculate respectively according to required parameter, section
Point hop count and hit rate, x and y are the constant parameter for characterizing tri- factor proportions of W, H, R, and specific value can be according to reality
Demand dynamic adjusts, but must is fulfilled for condition x+y≤0.
Step 3, multi-node collaborative is distributed in domain.Mist computing architecture based on broadcast and storage network uses centerized fusion thought,
To realize the collaboration of domain interior nodes, this causes mist server itself to safeguard the data message and control information of complexity, and
Only need to be responsible for realizing service logic, therefore compared to complete distributed treatment more quickly and efficiently.When mist server local
When being stored with the content resource of user's request, content requests of the mist server directly in response to terminal user；If mist server local
When not storing requested content resource, mist server directly sends collaborative query in a domain to controller and asked.Control
When device receives the content query requests that mist server is sent, in the controller institute compass of competency and being stored with requested
Other mist servers held, it is worth by calculating these mist servers for the forwarding of user UCL requests, selection forwarding value is most
High mist server returns to requesting node as optimal forward node, and by optimal node information.Requesting node receives control
After the information that device returns, content requests are directly forwarded to IP and the port corresponding to optimal node, content resource is obtained and returns
Back to terminal user.
In the specific implementation, structure follows the content of the mist computing architecture based on broadcast and storage network shown in Fig. 1 to the present invention first
Collaboration distribution application system.Without loss of generality, it is assumed that content collaboration distribution application system mainly includes 7 nodes：1 from mutual
Hot content is gathered in networking and carries out the crawler server of content metadata UCL indexes, 1 carries out the wide of hot content broadcast
Broadcast server, 3 mist server nodes (node identification is respectively A, B, C), 1 controller node, and 1 client (section
Point identification is D).It is assumed that crawler server acquires 100 popular webpages from internet, and these popular webpages are entered
The semantic parsing of row obtains 100 corresponding UCL, and 100 popular webpages and 100 UCL have all then been transmitted into broadcasting server.
100 popular webpages and 100 UCL are distributed to 3 mist servers by broadcasting server by broadcast, radiation, then of the invention
Specific implementation process is as follows：
(1) more granularity content storage collaborations.3 mist servers are for the 100 popular webpages arrived by broadcast reception, head
The hot value P of each popular webpage is first calculated according to formula (1)i.Next, it is determined that PiWhether condition is met：Pi≥Ps, if satisfied,
Web page element parsing then is carried out to the popular webpage, obtains fine-grained web page element (picture that is embedded in such as webpage, music
Deng), then wherein all web page elements with white list matching are selected.Then, for selecting the web page element come, make
Hash value is asked to the original URL of web page element with SHA1 algorithms.Finally, web page element storage is cached to Web according to hash value
Corresponding catalogue.
(2) the leading information processing of controller.Each mist server is in 100 warm for receiving broadcasting server and sending
Door webpage and during 100 UCL, can be actively to controller report corresponding information.And 100 UCL content metadatas information will be shown
In user terminal browser interface, user is by clicking on webpage full text or web page element corresponding to UCL access, it is assumed that these contents are deposited
Storage is in mist server A, B and C, once there is user to ask to reach mist server, mist server be able to should all ask to controller report
The hit situation asked.In addition, in this dynamic process, the content stored in mist server may increase, deletes or update, mist
Server also can be to the corresponding modification information of controller report.And index information table and the information on services that controller will be safeguarded in domain
Table, its structure is respectively as shown in table 1, table 2：
Table 1 indexes information table
| Node | Port | URL | Hash value |
| A | 8080 | www.sina.com.cn | 52df5gd2fg5dvbn6 |
| B | 9009 | www.baidu.com | 4fghadf234hg23bc |
| ┆ | ┆ | ┆ | ┆ |
| C | 9009 | www.jd.com | ac45b5b52c2gfg0 |
The service information table of table 2
| Source identifies | Node identification | Hash value | Type | Hit |
| D | A | 1g34h121h32lc145 | 1 | 0 |
| D | B | efgl19034hg2abcd | 5 | 1 |
| ┆ | ┆ | ┆ | ┆ | ┆ |
| D | C | 5acb632fg4987de | 4 | 2 |
Controller within each time-count cycle, can calculate every one kind of content in 3 mist servers according to formula (2)
Other c weighted value Wc, and then try to achieve class weight ratio
(3) multi-node collaborative is distributed in domain.It is assumed that the UCL sent by user asks corresponding classification 1, corresponding to URL
Hash value is 4fghadf234hg23bc.UCL requests reach mist server A, and miss in mist server A, then mist takes
Business device A inquiry requests into controller sending domain.
When controller receives content query requests in the domain that mist server A is sent, first according to hash value from node identification table
In find the set Node={ Node of all nodes for being stored with requested content1,Node2,...,Nodei}.Then, for
Each node Node in set Nodei, Node is obtained from class weight table according to the class label of contentiClass weight Wi,
The hop count H between node is obtained from topology information table according to node identificationi, and Node is calculated according to service information tableiAsked
Seek the hit rate R in classificationi, Node is finally calculated according to formula (3)iForwarding value Fi.When all nodes in set Node
After forwarding value calculation out, controller sorts these nodes according to forwarding value from high to low, picks out before coming most
The node in face returns to requesting node as optimal forward node, and by optimal forward node information.User receives controller and returned
After the information returned, UCL requests are directly sent to optimal forward node, last optimal forward node is right by user UCL requests institute
The web page contents answered are in full or web page element returns to terminal user.
Claims (9)
1. a kind of mist computing architecture content collaboration distribution method based on broadcast and storage network, it is characterised in that be broadly divided into 3 steps
Suddenly：
Step 1, more granularity content collaboration storages；The mist computing architecture based on broadcast and storage network is built first, in the mist computing architecture
Include two kinds of storage modes simultaneously：Web is cached and broadcasted content storage, realizes two kinds in full of webpage constitution element and web page contents
The collaboration storage of granularity content；
Step 2, the leading information processing of controller；Mist computing architecture based on broadcast and storage network is made up of multiple domains, in each domain
One controller of middle configuration, multiple mist servers in its management and control domain；
Step 3, multi-node collaborative is distributed in domain；When mist server local is stored with the content resource of user's request, mist service
Content requests of the device directly in response to terminal user；If mist server local does not store requested content resource, mist service
Device directly sends collaborative query in a domain to controller and asked.Controller receives the content query requests that mist server is sent
When, calculating institute's scope memory storage by corresponding forwarding strategy has the optimal node information of requested content, and optimal node is believed
Breath returns to requesting node.After requesting node receives the information of controller return, content requests are directly forwarded to optimal node
Corresponding IP and port, obtain content resource and return to terminal user.
2. the mist computing architecture content based on broadcast and storage network as described in power requires 1 cooperates with distribution method, it is characterised in that Web
Caching is responsible for the fine-grained webpage constitution element content of storage, and broadcasted content storage is then responsible in the thicker webpage of storage granularity
Hold in full.
3. the mist computing architecture content collaboration distribution method based on broadcast and storage network as claimed in claim 1, it is characterised in that
In mist computing architecture based on broadcast and storage network, while two kinds of varigrained content storage modes are safeguarded, webpage structure is realized with this
Into element and web page contents, two kinds of granularity contents cooperate with storage in full；First broadcasted content storage in select hot value be more than to
Determine threshold value PsWeb page contents in full, and to web page contents in full parsed to obtain fine-grained web page element, then at it
In select all web page elements with the white list website list of customization (system operator) matching, and store in Web cachings；
The hot value of web page contents full text is calculated according to formula (1)：
Wherein, PiThe temperature of the web page contents full text within i-th of time-count cycle is represented, and meets P1=0；NiRepresent at i-th
The number that web page contents are accessed in full in time-count cycle；NiWeight coefficient a>1.
4. the mist computing architecture content collaboration distribution method based on broadcast and storage network as claimed in claim 1, it is characterised in that when
When content in mist server occurs increase, deletes and update, mist server all can be to the corresponding modification information of controller report；
Once there is terminal request to reach mist server, mist server also can be to the specifying information of this request of controller report；Controller
The information reported according to each mist server, dynamically adjust index information table and service information table in this domain that it is recorded.
5. the mist computing architecture content collaboration distribution method based on broadcast and storage network as claimed in claim 1, it is characterised in that control
Device processed periodically calculates the class weight ratio of each mist server according to service information table, and is tieed up for each mist server
A class weight ratio information table is protected, broadcasted content and UCL foundation are received next time as the mist server；User passes through
Content corresponding to UCL requests UCL；It is assumed that the classification sum of all the elements is Nc, then any classification c weighted value is according to formula
(2) calculated：
Wherein RcRepresent in current period, the number that the UCL requests that user sends are classification c, andFor institute in this cycle
The sum of some user UCL requests；Then class weight ratio is
6. the mist computing architecture content collaboration distribution method based on broadcast and storage network as claimed in claim 1, it is characterised in that control
Forwarding of the device processed by calculating each user UCL requests is worth, to determine forwarding strategy in the domain of mist server；Forwarding value is shone
Formula (3) is calculated：
F=α W+ β H+ (1- alpha-betas) R formula (3)
WhereinAnd W, H and R weight that to be controller calculate respectively according to required parameter, node are jumped
Number and hit rate, x and y are the constant parameter for characterizing tri- factor proportions of W, H, R.
7. the content collaboration distribution application system of a kind of mist computing architecture based on broadcast and storage network, it is characterised in that mainly include
One lower node：Hot content is gathered from internet and carries out the crawler server of content metadata UCL indexes, is carried out in hot topic
Hold the broadcasting server of broadcast, mist server node, controller node, and client；Crawler server gathers from internet
Popular webpage, and semantic parsing is carried out to these popular webpages and obtains corresponding UCL, then popular webpage and UCL are passed
To broadcasting server；Popular webpage and UCL are distributed to each mist server by broadcasting server by broadcast, radiation；
Mist server is for the hot value P of the popular webpage, the first each popular webpage of calculating that are arrived by broadcast receptioni.Then,
Judge PiWhether condition is met：Pi≥Ps, if satisfied, then carrying out web page element parsing to the popular webpage, obtain fine-grained net
Page element, then wherein all web page elements with white list matching are selected；Then, for selecting the webpage come member
Element, hash value is asked to the original URL of web page element using SHA1 algorithms；Finally, web page element storage is arrived by Web according to hash value
The corresponding catalogue of caching.
8. the content collaboration distribution application system of the mist computing architecture based on broadcast and storage network as claimed in claim 7, its feature
It is, each mist server, can be actively to controller report when receiving the popular webpage and UCL that broadcasting server is sent
Corresponding information；And UCL content metadatas information will be illustrated in user terminal browser interface, user is corresponding by clicking on UCL access
Webpage in full or web page element, it is assumed that these contents are stored in each mist server, once have user ask reach mist clothes
Business device, mist server all can be to the hit situation of the controller report request；The content stored in mist server is as increased, deleting
Or renewal, mist server also can be to the corresponding modification information of controller report；And the index information table that controller will be safeguarded in domain
And service information table.
9. the content collaboration distribution application system of the mist computing architecture based on broadcast and storage network as claimed in claim 7, its feature
It is, controller within each time-count cycle, can calculate the weighted value W of each classification c of content in each mist serverc,
And then try to achieve class weight ratio
If the corresponding classification of UCL requests sent by user, the one of mist server A of UCL request arrival, and
Mist server A is miss, then mist server A inquiry request into controller sending domain；
When controller receives content query requests in the domain that mist server A is sent, first looked for according to hash value from node identification table
To the set Node={ Node for all nodes for being stored with requested content1,Node2,...,Nodei}；Then, for set
Each node Node in Nodei, Node is obtained from class weight table according to the class label of contentiClass weight Wi, according to
Node identification obtains the hop count H between node from topology information tablei, and Node is calculated according to service information tableiIn requested class
Hit rate R on noti, finally calculate NodeiForwarding value Fi；When the forwarding value calculation of all nodes in set Node goes out
After coming, controller sorts these nodes according to forwarding value from high to low, picks out the node for coming foremost as most
Good forward node, and optimal forward node information is returned into requesting node；After user receives the information of controller return, directly
UCL requests are sent to optimal forward node, user UCL is asked corresponding web page contents in full by last optimal forward node
Or web page element returns to terminal user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710851632.2A CN107682416B (en) | 2017-09-19 | 2017-09-19 | Broadcast-storage network-based fog computing architecture content collaborative distribution method and application system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710851632.2A CN107682416B (en) | 2017-09-19 | 2017-09-19 | Broadcast-storage network-based fog computing architecture content collaborative distribution method and application system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107682416A true CN107682416A (en) | 2018-02-09 |
| CN107682416B CN107682416B (en) | 2020-05-19 |
Family
ID=61137383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710851632.2A Active CN107682416B (en) | 2017-09-19 | 2017-09-19 | Broadcast-storage network-based fog computing architecture content collaborative distribution method and application system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107682416B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108156267A (en) * | 2018-03-22 | 2018-06-12 | 山东大学 | Improve the method and system of website visiting time delay in a kind of mist computing architecture using caching |
| CN109167842A (en) * | 2018-10-24 | 2019-01-08 | 清华大学 | Content distribution and transmission service service system and method based on mixing broadcast mode |
| CN109802998A (en) * | 2018-12-28 | 2019-05-24 | 上海无线通信研究中心 | A kind of mist network cooperating scheduling motivational techniques and system based on game |
| CN110457575A (en) * | 2019-07-08 | 2019-11-15 | 视联动力信息技术股份有限公司 | File push method, apparatus and storage medium |
| CN110601812A (en) * | 2019-09-17 | 2019-12-20 | 电子科技大学 | Privacy protection encrypted data query method based on fog assistance |
| CN112287102A (en) * | 2019-08-29 | 2021-01-29 | 北京京东尚科信息技术有限公司 | Data mining method and device |
| US11481259B2 (en) | 2020-01-07 | 2022-10-25 | International Business Machines Corporation | Distributing computation workloads based on calculated compute gravity within differing computing paradigms |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150078333A1 (en) * | 2013-09-19 | 2015-03-19 | Cisco Technology, Inc. | High-speed mobile broadband network access via programmed tracking of a sequence of wireless broadband data links |
| CN106357743A (en) * | 2016-08-29 | 2017-01-25 | 北京邮电大学 | Fog computing network service transmission method based on grading caching |
| US20170048308A1 (en) * | 2015-08-13 | 2017-02-16 | Saad Bin Qaisar | System and Apparatus for Network Conscious Edge to Cloud Sensing, Analytics, Actuation and Virtualization |
-
2017
- 2017-09-19 CN CN201710851632.2A patent/CN107682416B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150078333A1 (en) * | 2013-09-19 | 2015-03-19 | Cisco Technology, Inc. | High-speed mobile broadband network access via programmed tracking of a sequence of wireless broadband data links |
| CN105794268A (en) * | 2013-09-19 | 2016-07-20 | 思科技术公司 | High-speed mobile broadband network access via programmed tracking of a sequence of wireless broadband data links |
| US20170048308A1 (en) * | 2015-08-13 | 2017-02-16 | Saad Bin Qaisar | System and Apparatus for Network Conscious Edge to Cloud Sensing, Analytics, Actuation and Virtualization |
| CN106357743A (en) * | 2016-08-29 | 2017-01-25 | 北京邮电大学 | Fog computing network service transmission method based on grading caching |
Non-Patent Citations (3)
| Title |
|---|
| FLAVIO BONOMI 等: "Fog computing and its role in the internet of things", 《MCC WORKSHOP ON MOBILE CLOUD COMPUTING》 * |
| 杨鹏 等: "播存网络体系结构普适模型及实现模式", 《电子学报》 * |
| 顾梁 等: "一种播存网络环境下的UCL协同过滤推荐算法", 《计算机研究与发展》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108156267A (en) * | 2018-03-22 | 2018-06-12 | 山东大学 | Improve the method and system of website visiting time delay in a kind of mist computing architecture using caching |
| CN108156267B (en) * | 2018-03-22 | 2020-12-29 | 山东大学 | Method for improving website access time delay by using cache in fog computing architecture |
| CN109167842A (en) * | 2018-10-24 | 2019-01-08 | 清华大学 | Content distribution and transmission service service system and method based on mixing broadcast mode |
| CN109802998A (en) * | 2018-12-28 | 2019-05-24 | 上海无线通信研究中心 | A kind of mist network cooperating scheduling motivational techniques and system based on game |
| CN109802998B (en) * | 2018-12-28 | 2021-09-17 | 上海无线通信研究中心 | Game-based fog network cooperative scheduling excitation method and system |
| CN110457575A (en) * | 2019-07-08 | 2019-11-15 | 视联动力信息技术股份有限公司 | File push method, apparatus and storage medium |
| CN110457575B (en) * | 2019-07-08 | 2023-08-18 | 视联动力信息技术股份有限公司 | File pushing method, device and storage medium |
| CN112287102A (en) * | 2019-08-29 | 2021-01-29 | 北京京东尚科信息技术有限公司 | Data mining method and device |
| CN112287102B (en) * | 2019-08-29 | 2024-04-16 | 北京京东尚科信息技术有限公司 | Data mining method and device |
| CN110601812A (en) * | 2019-09-17 | 2019-12-20 | 电子科技大学 | Privacy protection encrypted data query method based on fog assistance |
| US11481259B2 (en) | 2020-01-07 | 2022-10-25 | International Business Machines Corporation | Distributing computation workloads based on calculated compute gravity within differing computing paradigms |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107682416B (en) | 2020-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107682416A (en) | Mist computing architecture content collaboration distribution method and application system based on broadcast and storage network | |
| Chen et al. | Leveraging social networks for P2P content-based file sharing in disconnected MANETs | |
| US20140280606A1 (en) | Method and Apparatus for Content Management | |
| CN104394236B (en) | Node combines the distributed collaborative caching method perceived with message status | |
| CN101588468B (en) | Medium playing method, medium playing device and medium playing system based on P2P | |
| CN106161204B (en) | A kind of data transmission method in mobile social networking based on swarm intelligence | |
| CN103260050B (en) | Based on the video on-demand system of Google App Engine Cloud platform | |
| CN106454430B (en) | For the preparatory dissemination method of video traffic in Autonomous Domain in NDN/CCN | |
| CN104507124A (en) | Management method for base station cache and user access processing method | |
| CN108881445A (en) | A kind of mist calculate in the cooperation caching method based on ancient promise game | |
| CN102045392A (en) | Interest-based adaptive topology optimization method for unstructured P2P (peer-to-peer) network | |
| CN101217565A (en) | A network organization method of classification retrieval in peer-to-peer network video sharing system | |
| Fei | A novel approach to managing consistency in content distribution networks | |
| Wang et al. | Social-aware video delivery: Challenges, approaches, and directions | |
| Al-Ward et al. | Caching transient data in Information-Centric Internet-of-Things (IC-IoT) networks: A survey | |
| WO2021017968A1 (en) | Method, apparatus and system for processing access request in content delivery system | |
| Zhang | A literature survey of cooperative caching in content distribution networks | |
| Alduayji et al. | PF-EdgeCache: Popularity and freshness aware edge caching scheme for NDN/IoT networks | |
| CN107302571B (en) | The routing of information centre's network and buffer memory management method based on drosophila algorithm | |
| Li et al. | QoS-aware streaming in overlay multicast considering the selfishness in construction action | |
| Reshadinezhad et al. | An efficient adaptive cache management scheme for named data networks | |
| Almobaideen et al. | Application based caching in fog computing to improve quality of service | |
| Jia et al. | Cluster-based content caching driven by popularity prediction | |
| Balasubramaniam et al. | Biologically inspired future service environment | |
| CN108769251A (en) | A kind of heuristic CCN network cooperations caching method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |