[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN102684817A - Digital fountain code based cloud storage system structure and service providing method - Google Patents

Digital fountain code based cloud storage system structure and service providing method Download PDF

Info

Publication number
CN102684817A
CN102684817A CN2011100581237A CN201110058123A CN102684817A CN 102684817 A CN102684817 A CN 102684817A CN 2011100581237 A CN2011100581237 A CN 2011100581237A CN 201110058123 A CN201110058123 A CN 201110058123A CN 102684817 A CN102684817 A CN 102684817A
Authority
CN
China
Prior art keywords
cloud
data
edge
user
core
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
CN2011100581237A
Other languages
Chinese (zh)
Inventor
万鑫明
王奇凡
王争
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.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to CN2011100581237A priority Critical patent/CN102684817A/en
Publication of CN102684817A publication Critical patent/CN102684817A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Information Transfer Between Computers (AREA)

Abstract

The invention discloses a digital fountain code based cloud storage system structure and a service providing method. By means of the system structure and the service providing method, the quality of service (QoS) of a cloud storage is effectively guaranteed. Edge clouds are deployed on different networks and in different regions, and the digital fountain code based cloud storage system structure is formed by a core cloud, the edge clouds, a terminal user and a digital fountain code coding and decoding layer. According to the system structure, services of common data, frequently-used data by a user and the like are distributed on the edge clouds of the networks or the regions where the terminal user is located, and the core cloud and the edge clouds cooperate to transparently provide a cloud storage service for the user. On the above basis, the service providing method specific to the digital fountain code based cloud storage system structure, the cooperating method of the core cloud and the edge clouds, and a user data migration method are provided. According to the digital fountain code based cloud storage system structure and the service providing method, most of network flow in the cloud storage is limited to local terminal users, the flow of a backbone network is reduced, the problem of the dependence of a centralized type cloud storage mode on the network bandwidth, the time delay and the like is effectively avoided, and the QoS of the cloud storage is guaranteed.

Description

A kind of cloud storage architecture and service providing method based on digital fountain sign indicating number
Technical field
The present invention relates to a kind of cloud storage architecture,, alleviated the dependence of cloud storage, improve the QoS of cloud storage the network bandwidth through introducing the notion of edge cloud and digital fountain code coding/decoding layer based on digital fountain sign indicating number.The present invention relates to areas of information technology such as cloud storage, digital fountain sign indicating number, distributed storage, computer network.
Background technology
Cloud storage (Cloud Storing) be in a kind of network storage more rationally, effective service provides and operation mode, under this pattern, be in local various stores service traditionally and will be replaced by the cloud stores service by large-scale, centralized memory module.In the cloud storage, the user can the convenient various storage resources that use cloud storage service provider to provide pellucidly as required.These storage resources can provide and obtain apace, have reduced enterprise or personal user and have purchased, administer and maintain cost, have and use future widely.The cloud storage is the stores service operation mode that the computer business provides industry to propose, and correlation technique relates to grid storage (Grid Storing), distributed storage (Distributed Storing), parallel computation (Parallel Computing), the network storage (Network Storage Technologies), digital fountain (Digital Fountain), virtual (Virtualization), load balancing (Load Balance) etc.
A core concept of cloud storage is exactly through improving constantly the overall storage capacity of " cloud " end (memory node); And then the storage that reduces user terminal is born; Finally make user terminal be simplified to a simple store instruction input-output equipment, and can enjoy the powerful storage capacity of " cloud " as required.
Under existing centralized cloud storage system framework, set up centralized large-scale data center and storage center, form the high in the clouds of cloud storage, by high in the clouds the user is provided externally required stores service; The user directly obtains system data from high in the clouds through storage networking, and is as shown in Figure 1.Because user's data is stored and moved and all carry out beyond the clouds, cloud is stored the dependence of the bandwidth of network and data transmission delay stronger.The application of cloud storage must cause lot of data mutual, and the outlet network of storage center is caused heavy pressure.The traffic of high in the clouds outlet network is the traffic sum in each terminal and high in the clouds.Under current network environment, the service quality QoS that centralized cloud storage architecture can provide (availability, throughput, time delay, time delay variation etc.) is difficult to satisfy the demand of user to the cloud storage; For example, the internet virtual disk application receives the puzzlement of problems such as network congestion, high time-delay.In order to guarantee the QoS of cloud storage, various relevant technology are studied, and wherein most typical is that data are encoded, and to reduce the data volume of transmission over networks, reduce the active demand of cloud storage to the network bandwidth.But these technology only can partly alleviate the network burden under the cloud storage environment, not from basic head it off.
Based on above problem, the present invention has designed the cloud storage architecture based on digital fountain sign indicating number, from computer network and memory location equal angles, solves the problem that the existing network bandwidth, time delay etc. can not satisfy cloud stores service demand.Because under the cloud storage environment; Has bigger redundancy for user data (virtual disk etc.); And these data volumes are bigger, simultaneously for unique user, and the data relative fixed of use and customer location relative fixed; Therefore adopt cloud storage architecture, user instruction data and user's frequently-used data are stored on the nearer server of distance users based on digital fountain sign indicating number.This mode can reduce the data traffic of backbone network with the limiting transmission of data of cloud stores service in subscriber's local, fundamentally solves the dependency problem of cloud storage to the network bandwidth, time delay, guarantees the availability of cloud storage.The present invention is provided with the server node with storage capacity in each network and area and forms the edge cloud, has formed the cloud storage architecture based on digital fountain sign indicating number of " core cloud ", " edge cloud ", user terminal and four level of digital fountain code coding/decoding layer.Node in the edge cloud is through collaborative with the core cloud; (correlation between communication data comprises correlation and the correlation between parallel data between its data to utilize correlation between communication data; Through technology such as digital fountain sign indicating number, compress technique, automatic buffer memory and local shape libraries; Can effectively reduce data volume) reduce the data volume that the core cloud exports, solve the dependence of cloud storage to the network bandwidth, time delay.
Summary of the invention
Problem to be solved by this invention is the active computer network is difficult to satisfy cloud stores service demand at aspects such as bandwidth, time delays a problem.For head it off; Improvement to existing cloud memory model; Designed cloud storage architecture based on digital fountain sign indicating number; Be provided with corresponding server node and form the edge cloud, form the cloud storage architecture of forming by " core cloud ", " edge cloud ", user terminal and digital fountain code coding/decoding layer based on digital fountain sign indicating number.Edge cloud and core cloud are collaborative, utilize the correlation between communication data that it is handled, and reduce the data volume of core cloud outlet, solve the dependency problem of cloud storage to the network bandwidth, time delay.
Cloud storage architecture based on digital fountain sign indicating number of the present invention is made up of core cloud, edge cloud, terminal use and digital fountain code coding/decoding layer:
The core cloud: have powerful storage capacity, under the support of cloud, cooperation flexibly provides transparent cloud stores service to the user rapidly each other on the edge of.The core cloud is the cloud stores service center that has the server cluster composition of powerful storage capacity, the various stores service that the user uses the core cloud to provide according to self-demand.
Edge cloud: form with the server node in area by being distributed in heterogeneous networks,, also flexibly the cloud stores service is provided rapidly to the user with the cooperation of core cloud for the core cloud provides corresponding miscellaneous function.The edge cloud is connected with the core cloud through backbone network, user terminal under the coordination of core cloud, the service of using the edge cloud to provide nearby.The edge cloud is responsible for the data flow between core cloud and the user terminal is carried out processed (like compression, coding and decoding, encryption and decryption etc.) on the one hand, utilizes the correlation between communication data, reduces network overhead, reduces time delay, guarantees cloud storage QoS; On the other hand, edge cloud storage terminal visit needed general character data of cloud stores service and frequently-used data (for example, the file that user's frequentation is asked generates the needed dynamic link library of information report etc.).
Terminal: the request entity that uses cloud storage related service.Terminal use's prime responsibility is to accomplish storage to wait command function with obtaining: on the one hand, land the cloud storage center, make up the user job environment, ask needed data to the cloud storage; On the other hand, for the user interface of browsing and showing the cloud stores service is provided.
Numeral fountain code coding/decoding layer: before the user submits data to, the data of desiring to upload are carried out digital fountain sign indicating number coding, upload the coded identification data again; When the user wanted to obtain data, advanced line of numbers fountain sign indicating number decoding sent the successful data of decoding to user again.
In the cloud storage architecture based on digital fountain sign indicating number, core cloud and edge cloud are cooperated each other, are the give security cloud stores service of QoS of user.
The cooperation of edge cloud and core cloud: the core cloud provides cloud stores service such as user storage, data migration; The edge cloud is shared the correlation function of core cloud, and the storage terminal visit needed general character data of cloud stores service and user's frequently-used data under the coordination of core cloud, provide the cloud stores service for the user pellucidly.Utilize high bandwidth and the network characteristic of low time delay between edge cloud and the terminal, guarantee the QoS of cloud stores service.Core cloud and the cooperation of edge cloud are that the terminal use provides the cloud stores service with the mode to user transparent.
The cooperation at edge cloud and terminal: the edge cloud utilizes high bandwidth and the local network of low time delay between itself and the terminal to send the needed general character data of terminal access cloud stores service and user's frequently-used data to terminal fast, and transparent provides real-time cloud stores service to the terminal.
The cooperation at core cloud and terminal: under cloud auxiliary, the core cloud is accomplished the concrete stores service that ask at the terminal on the edge of.
Based on the cloud storage architecture of digital fountain sign indicating number, core cloud and edge cloud are cooperated jointly, transparent the cloud stores service are provided for the user.The user at first proposes services request to the core cloud, the edge cloud negotiate service strategy at core cloud and user place then, and at last according to the result who consults, cooperating jointly is the give security service of QoS of user.
Under cloud storage architecture according to the invention,, utilize data dependence that the core cloud is had following three kinds of methods for the terminal provides the processing procedure of data M according to different scene:
Method one: the situation that will send to data M existence self correlation at a certain terminal for the core cloud; The core cloud is consulted according to the characteristics of data M and the edge Cloud Server at target terminal place; According to the result who consults; M handles to data, generates the littler data m of data volume, and data m is sent to the edge Cloud Server; After the edge Cloud Server is received m, use inverse algorithm, obtain initial data M, and data M is sent to target terminal.Data are sent to the core cloud in the terminal, take opposite process.
Method two: for the data cases with correlation being provided will for a plurality of terminals, might as well establish data is M 1, M 2..., M n, core cloud and edge cloud are consulted, and use the result who consults, to data M 1, M 2..., M nHandle, generate the littler data set of data volume, and send it to the edge Cloud Server; After the edge Cloud Server is received data, it is reduced, and send the data to relevant terminal.Data are sent to the core cloud in the terminal, take opposite process.
Method three: the third situation is the target data M that the core cloud will send 1, M 2..., M n, the process that generates target data is similar, and has common input (for example, using identical function library); Therefore, target data no longer is directly to be generated by the core cloud, but under the coordination of core cloud, by the edge cloud data is sent to relevant terminal.For example, in the internet virtual disk application, move to the edge cloud by the core cloud for the generation of user side cloud interface; A large amount of request msgs only need on the edge of cloud with and its terminal that links to each other between transmit, thereby reduced the data volume that exports Network Transmission at the core cloud.
When terminal use's position changes, need move with user-dependent data, to satisfy user experience better.Whether the user is using the cloud stores service when changing based on occurrence positions, and migration is divided into real-time migration and plan migration.
Real-time migration refers to that user's occurrence positions in the process of using the cloud stores service moves, by institute on the edge of cloud move into new edge cloud, user's service data should be under the coordination of core cloud at this moment, cloud is moved to new edge cloud from former edge.This stylish edge cloud replaces former edge cloud provides service to the user.
The plan migration refers to that the user informs high in the clouds in advance with the routing of oneself; The cloud storage center is in a planned way arranged the transfer of user-dependent data between different edge clouds according to user's stroke, the service that the use cloud storage that the assurance user all can be real-time in predetermined stroke provides.
Description of drawings
Fig. 1 is centralized cloud storage network architecture structure chart
Fig. 2 is the cloud storage architecture figure based on digital fountain sign indicating number that the present invention proposes
Fig. 3 is the service providing method sketch map of layered distribution type cloud storage
Fig. 4 is a traffic sketch map under the centralized cloud storage architecture
Fig. 5 utilizes correlation minimizing traffic sketch map between its data
Fig. 6 utilizes correlation minimizing traffic sketch map between parallel data
Fig. 7 moves to the edge cloud with the critical storage that causes data expansion to reduce traffic sketch map
Fig. 8 is core cloud and edge cloud negotiations process sketch map
Fig. 9 is a real-time migration each side reciprocal process sketch map
Figure 10 is a plan migration each side reciprocal process sketch map
Embodiment
The present invention is provided with the edge cloud that corresponding server node is formed beyond the clouds, forms the cloud storage architecture of being made up of " core cloud ", " edge cloud " and user terminal based on digital fountain sign indicating number, and is as shown in Figure 2.Through the traffic that memory node reduces core cloud exit is set, thereby reduce time delay between core cloud storage center and terminal, guaranteed the QoS of cloud storage.
The core cloud: have powerful storage capacity, under the support of cloud, cooperation flexibly provides transparent cloud stores service to the user rapidly each other on the edge of.The core cloud is the cloud stores service center that has the server cluster composition of powerful storage capacity, the various stores service that the user uses the core cloud to provide according to self-demand.
Edge cloud: form with the server node in area by being distributed in heterogeneous networks,, also flexibly the cloud stores service is provided rapidly to the user with the cooperation of core cloud for the core cloud provides corresponding miscellaneous function.The edge cloud is connected with the core cloud through backbone network, user terminal under the coordination of core cloud, the service of using the edge cloud to provide nearby.The edge cloud is responsible for the data flow between core cloud and the user terminal is carried out processed (like compression, coding and decoding, encryption and decryption etc.) on the one hand, utilizes the correlation between communication data, reduces network overhead, reduces time delay, guarantees cloud storage QoS; On the other hand, edge cloud storage terminal visit needed general character data of cloud stores service and frequently-used data (for example, the file that user's frequentation is asked generates the needed dynamic link library of information report etc.).
Terminal: the request entity that uses cloud storage related service.Terminal use's prime responsibility is to accomplish storage to wait command function with obtaining: on the one hand, land the cloud storage center, make up the user job environment, ask needed data to the cloud storage; On the other hand, for the user interface of browsing and showing the cloud stores service is provided.
Numeral fountain code coding/decoding layer: before the user submits data to, the data of desiring to upload are carried out digital fountain sign indicating number coding, upload the coded identification data again; When the user wanted to obtain data, advanced line of numbers fountain sign indicating number decoding sent the successful data of decoding to user again.
Under the architecture based on digital fountain sign indicating number, core cloud and edge cloud are cooperated jointly, and be transparent in the user provides the cloud stores service, as shown in Figure 3.Cloud stores service based on digital fountain sign indicating number provides step following:
Step a. user sends services request to the core cloud;
Step b. core cloud is searched the edge cloud at user place, and with the edge cloud negotiate service strategy at user place;
Step c is according to the service strategy of consulting, and the core cloud uses response data, the edge cloud use its data of storing (the needed general character data of terminal access cloud stores service and user's frequently-used data) common as the user transparent service is provided.
In the cloud stores service provided, the data transfer mode under the centralized cloud storage architecture of tradition was as shown in Figure 4.The core cloud directly directly is transferred to each terminal with the data that each terminal needs, and the wherein transmission of all data of demand burden on the backbone network brings bigger burden to network.Under cloud storage architecture,, utilize data dependence that the core cloud is handled for the terminal provides data M according to different scene based on digital fountain sign indicating number.According to the characteristics of data self, processing procedure has following three kinds of methods:
Method one: the situation that will send to data M existence self correlation at a certain terminal for the core cloud; The edge Cloud Server negotiate service strategy that the core cloud links to each other with target terminal according to the characteristics of data M; Use the result who consults; Core cloud and edge cloud are jointly for the user provides service, and be as shown in Figure 5.The practical implementation step is following:
Step a: there is self correlation in the target data that the core cloud sends to each terminal;
Step b: the core cloud is initiated the cooperation request to the edge cloud at place, terminal;
Step c: after the edge cloud is received the cooperation request, detect own resource and whether meet the demands, return acceptance response if satisfy; Otherwise, return the refusal response;
Steps d: the core cloud is then searched for new available edge cloud if receive that refusal is corresponding, and find and return step b, otherwise failure;
Step e: the core cloud is received acceptance response, initiates the negotiations process with the edge cloud; Negotiate the data processing algorithm of use, accomplish negotiations process, successfully set up cooperation relation;
Step f: the core cloud is processed f (M to the data M that will send i)=m i, generate overall data m i, and send to the edge cloud; The edge cloud to its reverse process, restores data, and sends to the corresponding terminal terminal after receiving the data of core cloud transmission.
Method two: for the data cases with correlation being provided will for a plurality of terminals, might as well establish data is M 1, M 2..., M n, core cloud and edge cloud are consulted, and use the result who consults, to data M 1, M 2..., M nHandle, generate the littler data set of data volume, and send it to the edge Cloud Server; After the edge Cloud Server is received data, it is reduced, and send the data to relevant terminal.Data are sent to the core cloud in the terminal, take opposite process.Data transmission procedure is as shown in Figure 6, and the practical implementation step is following:
Step a: the core cloud sends to and has very high correlation between the data of different terminals;
Step b: the core cloud is initiated the cooperation request to this edge cloud;
Step c: after the edge cloud is received the cooperation request, detect own resource and whether meet the demands, return acceptance response if satisfy; Otherwise, return the refusal response;
Steps d: the core cloud is then searched for new available edge cloud if receive that refusal is corresponding, and find and return step b, otherwise failure;
Step e: the core cloud is received acceptance response, initiates the negotiations process with the edge cloud; Negotiate the data processing algorithm of use, accomplish negotiations process, successfully set up cooperation relation;
Step f: the core cloud is handled v (M together to the data that will send each terminal 1, M 2..., M n)=Msg generates overall data Msg, and sends to the edge cloud; The edge cloud to its reverse process, decomposites the data component that will send to each terminal, and sends to the corresponding terminal terminal after receiving the data of core cloud transmission.
Method three: the third situation is the target data M that the core cloud will send 1, M 2..., M n, the process that generates target data is similar, and has common input (for example, using identical shape library); Therefore, target data no longer is directly to be generated by the core cloud, but under the coordination of core cloud, by the edge cloud data is sent to relevant terminal.For example, in remote desktop is used, move to the edge cloud by the core cloud for the generation of graphical user desktop; A large amount of graph datas only need on the edge of cloud with and its terminal that links to each other between transmit, thereby reduced the data volume that exports Network Transmission at the core cloud.Data transmission procedure is as shown in Figure 7, and the practical implementation step is following:
Step a: the generative process of target data that the core cloud sends to a plurality of terminals is similar, and has common input;
Step b: the core cloud is initiated the cooperation request to the edge cloud;
Step c: after the edge cloud is received the cooperation request, detect own resource and whether meet the demands, return acceptance response if satisfy; Otherwise, return the refusal response;
Steps d: the core cloud is then searched for new available edge cloud if receive that refusal is corresponding, and find and return step b, otherwise failure;
Step e: the core cloud receives that acceptance is corresponding, initiates the negotiations process with the edge cloud; The core cloud will be ordered, and Processing Algorithm semantic, that consult sends to the edge cloud, accomplishes negotiations process, has successfully set up cooperation relation;
Step f: when the core cloud wants to send data M iWhen giving certain terminal, the order that generation is changed data sends to the edge cloud that the terminal belongs to relevant low volume data;
Step g: the edge cloud receive orders with data after, call correlation function, calculate the result, and this result sent to associated terminal.
Provide in the process in the cloud stores service, the negotiations process of core cloud and edge cloud is as shown in Figure 8.Concrete negotiation step is following:
Step a: after the user sent services request, the core cloud was confirmed the edge cloud according to customer location;
Step b: the core cloud is initiated the cooperation request to the edge cloud;
Step c: the edge cloud is the cooperation Resources allocation, and sends agreement to the core cloud and cooperate corresponding after the request of receiving;
Steps d: the core cloud is consulted allocating task etc. with the edge cloud after receiving and agreeing response alternately;
Step e: core cloud and edge cloud are carried out cooperation by negotiation result, and service are provided for the user.
When terminal use's position changes, need move with user-dependent data, to satisfy user experience better.Whether the user is using the cloud stores service when changing based on occurrence positions, and migration is divided into real-time migration and plan migration.
In the real-time migration, the user by on the edge of cloud move into new edge cloud, this moment, user's service data should be under the coordination of core cloud, cloud is moved to new edge cloud from former edge.This stylish edge cloud replaces former edge cloud provides service to the user.The implementation step of real-time migration is as shown in Figure 9:
Step a: terminal user position sends and changes, and by the perception of high in the clouds institute;
Step b: the edge cloud judges according to the change information of customer location whether oneself still can provide required service;
Step c: still required service can be provided if the edge cloud is judged oneself, then repeating step b; Otherwise send the real-time migration request to the core cloud;
Steps d: the information that the core cloud provides according to the edge cloud, select new available edge cloud;
Step e: core cloud and new edge cloud are consulted and are disposed required environment for the user;
Step f: the user job environmental data has former edge cloud to move to new edge cloud;
Step g: core cloud and new edge cloud provide the cloud stores service for the user jointly.
In the plan migration, the cloud storage center is in a planned way arranged the transfer of user-dependent data between different edge clouds based on user's stroke, the service that the use cloud storage that the assurance user all can be real-time in predetermined stroke provides.The implementation step of plan migration is shown in figure 10:
Step a: the user submits the routing of oneself to high in the clouds;
Step b: the core cloud is worked out the data transfer plan according to user's routing;
Step c: the core cloud is according to data migration plan, and for the user service is provided with the respective edges cloud.

Claims (4)

1. cloud storage architecture based on digital fountain sign indicating number, it is characterized in that: this architecture is made up of core cloud, edge cloud, terminal and digital fountain code coding/decoding layer:
Core cloud: calculating, storage capacity are provided, to the user cloud stores service are provided under the support of cloud on the edge of;
The edge cloud: form with the server node in area by being distributed in heterogeneous networks, be connected with the core cloud through backbone network, for the core cloud provides corresponding miscellaneous function, cooperation provides the cloud stores service nearby to the user;
Terminal: use core cloud and edge cloud that the entity of cloud stores service is provided.
Numeral fountain code coding/decoding layer: user data encoded upload and decipher download.
2. the cloud storage architecture based on digital fountain sign indicating number according to claim 1, its service providing method are characterised in that the cooperation mode of core cloud, edge cloud and terminal room and the process that service provides:
The cooperation of edge cloud and core cloud: core cloud and the cooperation of edge cloud, the QoS of assurance cloud stores service is that the terminal use provides the cloud stores service with the mode to user transparent;
The cooperation at edge cloud and terminal: the edge cloud utilizes high bandwidth and the local network of low time delay between itself and the terminal to send the needed general character data of terminal access cloud stores service and user's frequently-used data to terminal fast, and transparent provides real-time cloud stores service to the terminal;
The cooperation at core cloud and terminal: under cloud auxiliary, the core cloud is accomplished the concrete cloud stores service that ask at the terminal on the edge of;
Service provides process: at first the user proposes services request to the core cloud, the edge cloud negotiate service strategy at core cloud and user place then, and at last according to the result who consults, cooperating jointly is the give security service of QoS of user.
3. cloud storage architecture and claim 2 described service providing method based on digital fountain sign indicating number according to claim 1, utilize data dependence that the core cloud is had following three kinds of methods for the terminal provides the processing procedure of data:
Method one: the situation that will send to data M existence self correlation at a certain terminal for the core cloud; The core cloud is consulted according to the characteristics of data M and the edge Cloud Server at target terminal place; According to the result who consults; M handles to data, generates the littler data m of data volume, and data m is sent to the edge Cloud Server; After the edge Cloud Server is received m, use inverse algorithm, obtain initial data M, and data M is sent to target terminal.Data are sent to the core cloud in the terminal, take opposite process.
Method two: for the data cases with correlation being provided will for a plurality of terminals, might as well establish data is M 1, M 2..., M n, core cloud and edge cloud are consulted, and use the result who consults, to data M 1, M 2..., M nHandle, generate the littler data set of data volume, and send it to the edge Cloud Server; After the edge Cloud Server is received data, it is reduced, and send the data to relevant terminal.Data are sent to the core cloud in the terminal, take opposite process.
Method three: the third situation is the target data M that the core cloud will send 1, M 2..., M n, the process that generates target data is similar, and has common input (for example, using identical function library); Therefore, target data no longer is directly to be generated by the core cloud, but under the coordination of core cloud, by the edge cloud data is sent to relevant terminal.For example, in the internet virtual disk application, move to the edge cloud by the core cloud for the generation data of user terminal; A large amount of storage data only need on the edge of cloud with and its terminal that links to each other between transmit, thereby reduced the data volume that exports Network Transmission at the core cloud.
4. cloud storage architecture and the described service providing method of claim 2 based on digital fountain sign indicating number according to claim 1; Migration of subscriber data need move according to customer location; It is characterized in that: whether the user is using the cloud stores service when changing according to occurrence positions, and migration is divided into real-time migration and plan migration:
Real-time migration refers to that user's occurrence positions in the process of using the cloud stores service moves, by institute on the edge of cloud move into new edge cloud, user's service data should be under the coordination of core cloud at this moment, cloud is moved to new edge cloud from former edge.This stylish edge cloud replaces former edge cloud provides service to the user.
The plan migration refers to that the user informs high in the clouds in advance with the routing of oneself; The cloud storage center is in a planned way arranged the transfer of user-dependent data between different edge clouds according to user's stroke, the service that the use cloud storage that the assurance user all can be real-time in predetermined stroke provides.
CN2011100581237A 2011-03-10 2011-03-10 Digital fountain code based cloud storage system structure and service providing method Pending CN102684817A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2011100581237A CN102684817A (en) 2011-03-10 2011-03-10 Digital fountain code based cloud storage system structure and service providing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2011100581237A CN102684817A (en) 2011-03-10 2011-03-10 Digital fountain code based cloud storage system structure and service providing method

Publications (1)

Publication Number Publication Date
CN102684817A true CN102684817A (en) 2012-09-19

Family

ID=46816235

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2011100581237A Pending CN102684817A (en) 2011-03-10 2011-03-10 Digital fountain code based cloud storage system structure and service providing method

Country Status (1)

Country Link
CN (1) CN102684817A (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018000647A1 (en) * 2016-06-30 2018-01-04 乐视控股(北京)有限公司 Method, device, terminal, and server for file downloading
CN109067859A (en) * 2018-07-20 2018-12-21 北京航空航天大学 A kind of double-deck cloud framework system and implementation method towards cross-domain cooperation with service
CN109117856A (en) * 2018-08-23 2019-01-01 中国联合网络通信集团有限公司 People and object method for tracing, apparatus and system based on intelligent edge cloud
CN109783190A (en) * 2018-12-12 2019-05-21 北京世纪互联宽带数据中心有限公司 Data migration method and its device, electronic equipment, computer-readable medium
CN110430266A (en) * 2019-08-06 2019-11-08 腾讯科技(深圳)有限公司 A kind of side cloud synergistic data transmission method, device, equipment and storage medium
CN110493304A (en) * 2019-07-04 2019-11-22 上海数据交易中心有限公司 Edge calculations system and transaction system
CN112328174A (en) * 2020-10-29 2021-02-05 北京京东尚科信息技术有限公司 Data pushing method, device, system and medium for cloud computing platform
CN112584439A (en) * 2020-11-27 2021-03-30 重庆邮电大学 Caching method in edge calculation
CN113132427A (en) * 2019-12-30 2021-07-16 中移智行网络科技有限公司 Edge cloud cooperation method and edge cloud cooperation system
CN113273162A (en) * 2019-05-23 2021-08-17 Oppo广东移动通信有限公司 Communication method and device and network architecture
CN113518111A (en) * 2021-05-07 2021-10-19 西安交通大学 Fountain code driven multi-path side unit cooperative caching method
CN113572813A (en) * 2021-06-22 2021-10-29 复旦大学 Data backup method based on network coding

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018000647A1 (en) * 2016-06-30 2018-01-04 乐视控股(北京)有限公司 Method, device, terminal, and server for file downloading
CN109067859A (en) * 2018-07-20 2018-12-21 北京航空航天大学 A kind of double-deck cloud framework system and implementation method towards cross-domain cooperation with service
CN109067859B (en) * 2018-07-20 2021-03-16 北京航空航天大学 Cross-domain collaborative service-oriented double-layer cloud architecture system and implementation method
CN109117856B (en) * 2018-08-23 2021-01-29 中国联合网络通信集团有限公司 Intelligent edge cloud-based person and object tracking method, device and system
CN109117856A (en) * 2018-08-23 2019-01-01 中国联合网络通信集团有限公司 People and object method for tracing, apparatus and system based on intelligent edge cloud
CN109783190A (en) * 2018-12-12 2019-05-21 北京世纪互联宽带数据中心有限公司 Data migration method and its device, electronic equipment, computer-readable medium
CN113273162A (en) * 2019-05-23 2021-08-17 Oppo广东移动通信有限公司 Communication method and device and network architecture
EP3972219A4 (en) * 2019-05-23 2022-05-25 Guangdong Oppo Mobile Telecommunications Corp., Ltd. Communication method and apparatus, and network architecture
CN113273162B (en) * 2019-05-23 2024-01-26 Oppo广东移动通信有限公司 Communication method and device and network architecture
US12001890B2 (en) 2019-05-23 2024-06-04 Guangdong Oppo Mobile Telecommunications Corp., Ltd. Communication method and apparatus, and network architecture
CN110493304A (en) * 2019-07-04 2019-11-22 上海数据交易中心有限公司 Edge calculations system and transaction system
CN110430266B (en) * 2019-08-06 2021-07-13 腾讯科技(深圳)有限公司 Edge cloud cooperative data transmission method, device, equipment and storage medium
CN110430266A (en) * 2019-08-06 2019-11-08 腾讯科技(深圳)有限公司 A kind of side cloud synergistic data transmission method, device, equipment and storage medium
CN113132427A (en) * 2019-12-30 2021-07-16 中移智行网络科技有限公司 Edge cloud cooperation method and edge cloud cooperation system
CN112328174A (en) * 2020-10-29 2021-02-05 北京京东尚科信息技术有限公司 Data pushing method, device, system and medium for cloud computing platform
CN112584439A (en) * 2020-11-27 2021-03-30 重庆邮电大学 Caching method in edge calculation
CN113518111A (en) * 2021-05-07 2021-10-19 西安交通大学 Fountain code driven multi-path side unit cooperative caching method
CN113572813A (en) * 2021-06-22 2021-10-29 复旦大学 Data backup method based on network coding

Similar Documents

Publication Publication Date Title
CN102684817A (en) Digital fountain code based cloud storage system structure and service providing method
CN101977242A (en) Layered distributed cloud computing architecture and service delivery method
Lv et al. Interaction of edge-cloud computing based on SDN and NFV for next generation IoT
Fan et al. Workload allocation in hierarchical cloudlet networks
CN111475274B (en) Cloud collaborative multi-task scheduling method and device
CN102857548B (en) A kind of mobile cloud computing resources distributes method rationally
WO2023024219A1 (en) Joint optimization method and system for delay and spectrum occupancy in cloud-edge collaborative network
EP2972854B1 (en) Distributed data center technology
CN103001892B (en) Based on network resource allocation method and the system of cloud computing
Si et al. QoS-aware dynamic resource management in heterogeneous mobile cloud computing networks
CN109831524A (en) A kind of load balance process method and device
CN115499859B (en) NWDAF-based method for managing and deciding computing resources
KR20180124419A (en) Application excution system based on distributed cloud, apparatus and control method thereof using the system
Jahromi et al. Online VNF placement and chaining for value-added services in content delivery networks
Maiti et al. Efficient data collection for IoT services in edge computing environment
Song et al. Machine learning enabling traffic-aware dynamic slicing for 5G optical transport networks
Hu et al. Joint load balancing and offloading optimization in multiple parked vehicle‐assisted edge computing
KR20170014804A (en) Virtual machine provisioning system and method for cloud service
Hosseini et al. Energy-efficient scheduling based on task prioritization in mobile fog computing
Badshah et al. Use of regional computing to minimize the social big data effects
Lu et al. Resource provisioning in collaborative fog computing for multiple delay‐sensitive users
Hung et al. A new technique for optimizing resource allocation and data distribution in mobile cloud computing
CN115550335A (en) Distributed mobile WEB3D rendering system and method
Happ et al. On the impact of clustering for IoT analytics and message broker placement across cloud and edge
Jasim et al. Efficient load migration scheme for fog networks

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
DD01 Delivery of document by public notice

Addressee: Wang Qifan

Document name: Notification of Publication of the Application for Invention

DD01 Delivery of document by public notice

Addressee: Wang Zheng

Document name: Notification that Application Deemed to be Withdrawn

C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20120919