CN102624596B - Reliability optimal tree-shaped core topological solving method of P2P live broadcast covering network - Google Patents
Reliability optimal tree-shaped core topological solving method of P2P live broadcast covering network Download PDFInfo
- Publication number
- CN102624596B CN102624596B CN201210126532.0A CN201210126532A CN102624596B CN 102624596 B CN102624596 B CN 102624596B CN 201210126532 A CN201210126532 A CN 201210126532A CN 102624596 B CN102624596 B CN 102624596B
- Authority
- CN
- China
- Prior art keywords
- reliability
- tree
- node
- topology
- degree
- 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.)
- Expired - Fee Related
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention aims at providing a reliability optimal tree-shaped core topological solving method of a P2P live broadcast covering network, which comprises the steps of: converting out-degree and reliability of a backbone node of the P2P live broadcast covering network into weighting reliability through hierarchical out-degree weighting, constructing a tree-shaped core topology by layers by using the weighting reliability as a sequencing basis; optimizing a hierarchical weight set by using an evolutionary programming algorithm, and solving by using the sum of accumulated reliability of all nodes in an evolution process as an evolution index to obtain a tree topology with optimal reliability. According to the reliability optimal tree-shaped core topological solving method of the P2P live broadcast covering network, provided by the invention, the stability of the P2P live broadcast covering network is improved, and the live broadcast time delay is reduced.
Description
Technical field
The present invention relates to network service, more particularly, relate to a kind of optimum tree-shaped core topology method for solving of reliability that adopts the live nerve of a covering of P2P of tree-shaped or tree-like/netted mixed topology.
Background technology
According to investigations, on internet, 70% flow is P2P application at present, in P2P framework, shared resource can directly exchange, do not need transfer and the participation of centralized servers, P2P system has good fault-tolerant ability to the unstable and variable connectedness of overlay network simultaneously, and keeps good reliability.
Current, the live nerve of a covering of P2P is the main ways of distribution of large scale network TV, and the topological structure of the live nerve of a covering of P2P can be divided into four kinds of patterns substantially: (1) imitates merely the tree of IP multicast tree, and has occurred the complementation distribution of many trees.(2) in network, each node, centered by oneself, initiatively goes to draw data under demand data drives, and forms like this network structure that there is no clear and definite topological structure.(3) robustness of the high efficiency of tree and data-driven is combined, form mixed structure.(4) node administration nerve of a covering and each self-organizing of Data dissemination nerve of a covering, distribute with node administration vectoring information.
The live nerve of a covering multicast of early stage P2P adopts tree conventionally, as NICE, and ESM system.The Data dissemination of this type of tree adopts the mode pushing away, and overhead is little, and time delay is shorter, and Bullet utilizes many modes that tree is complementary, has improved the speed of distribution.AnySee sets up many Data dissemination trees earlier in netted node administration topology, and between many distribution trees, carries out the complementation optimization of transfer resource.But the structural maintenance difficulty because node dynamic causes, causes the large-scale application of tree nerve of a covering to be restricted.
DONet (Data-driven Overlay Network) realizes the distribution of data by building pure fully connected topology, without building complicated control structure.Real time flow medium Play System CoolStreaming based on DONet agreement, its outstanding result of broadcast, lower delay have obtained confirming and affirming in actual motion.Other are able to the system of large-scale application, as PPLive, and PPStream, UUSee, GridMedia, Sopcast, TVants etc. have also been used similar method in data dispatch.DONet data dispatch mainly adopts the mode of drawing, and needs exchange buffering mapping graph (BM) continually between node for this reason, causes its extra bandwidth consumption larger, and time delay is also longer.
In order to overcome the deficiency of pull-mode, there is again the mixing distribution mode of push-and-pull combination, in the process of drawing, set up local tree, the mode then pushing away.In the mode pushing away, for avoiding leaf node uploading bandwidth to be not used, often adopt many subflows (Substream), the method for simultaneously transmitting on many trees.
Except carry out tree-shaped and cancellated mixing in Data dissemination, use, also have the mixed mode that uses different structure in node administration and Data dissemination.And in AnySee2, adopt tree to carry out monitoring nodes and management, and adopt network structure to carry out the method for transmission of flow media data, be called TCMM (Tree-Control-Mesh-Media).
In sum, in order to utilize the high efficiency of tree topology structure and the advantage of low time delay, avoid again because node frequently adds the topology that exits and cause to safeguard the defect of high cost, the topological structure that needs a kind of optimization, make it there is good robustness and reliability, can improve again the stability of nerve of a covering and reduce live time delay.
Summary of the invention
The invention provides the optimum tree-shaped core topology method for solving of reliability of the live nerve of a covering of a kind of P2P, utilize the high efficiency of tree topology structure and the advantage of low time delay, while avoiding adopting tree topology, because of node, frequently added the problem that the topology causing is safeguarded high cost that exits simultaneously.
The optimum tree-shaped core topology method for solving of reliability that the invention provides the live nerve of a covering of a kind of P2P, the optimum tree-shaped core topology method for solving of described reliability comprises:
(1) in the live nerve of a covering of described P2P, according to out-degree and reliability, choose backbone node;
(2) according to the length of the quantity of backbone node and average out-degree calculating out-degree weight array;
(3), according to the length of out-degree weight array described in step (2), produce at random one group of out-degree weight array, using one group of out-degree weight array of described random generation as initial population;
(4) for each individuality in described initial population, construct corresponding tree topology, and calculate the accumulation reliability sum of all nodes of every tree topology;
(5) using the initial population described in the described initial population of step (4) or step (6) as Advanced group species, each individuality of described Advanced group species copied and made a variation, forming variation population; According to each individuality of described variation population, construct corresponding tree topology, and calculate the accumulation reliability sum of all nodes of every tree topology.
(6) described Advanced group species is carried out to sort descending, the initial population of selecting preceding half individuality of sequence to evolve as a new round with the described accumulation reliability sum that makes a variation node corresponding to all individualities of population;
(7) initial population of evolving for the described new round producing in step (6), whether meet the judgement of evolution termination condition, if do not meet described evolution termination condition, repeated execution of steps (5) is to step (7), until meet evolution termination condition;
(8) in step (7), meet the initial population of the described new round evolution of evolution termination condition, be the optimum tree-shaped core topology of reliability.
Preferably, the length of described out-degree weight array is the depth capacity of described tree topology.
Preferably, the maximum of accumulation reliability sum that described satisfied evolution termination condition is all nodes is, the variation of mean value is less than threshold value or described satisfied evolution termination condition is that evolutionary generation reaches the upper limit.
Preferably, the computational methods of the accumulation reliability of described node, comprising:
(1) the accumulation reliability of the root node of described tree topology is himself reliability;
(2) from root node described in step (1), range travels through all child nodes, and each node is calculated to its accumulation reliability by step (3);
(3) self reliability of described node is multiplied by the accumulation reliability of its parent node as the accumulation reliability of described node.
Preferably, in the live nerve of a covering of described P2P, according to out-degree and reliability, choose in backbone node, described out-degree is greater than Streaming Media speed, and the node that described reliability is greater than reliability thresholds is chosen for backbone node.
Preferably, described reliability thresholds realizes by setting fixed value.
Preferably, described reliability thresholds realizes by the ratio of specifying backbone node quantity to account in total number of nodes.
Preferably, according to the quantity of the described backbone node of choosing and average out-degree, pass through formula:
Calculate the depth capacity of described tree topology structure, the quantity of the backbone node that wherein c chooses described in representing, the average out-degree of the backbone node of choosing described in d represents, l represents the degree of depth of described tree topology structure.
Preferably, the method for the corresponding tree topology of described structure, comprising:
601, according to succession, obtain the element of out-degree weight array, each element is carried out to step 602 to the processing of step 603, until all nodes join in tree;
602, according to described array element, calculate all weighting reliabilities that do not add the node of in-tree, and carry out sequence from high to low;
603, according to described node weighting reliability ordering result, the preceding node of sequence is joined in tree one by one.In if tree, there is no node, only get a node as root, otherwise node is added on the leaf node of original tree, until use up the out-degree of all leaf nodes or there is no node to be added.
The optimum tree-shaped core topology method for solving of reliability provided by the invention has improved the applicability of tree topology in live overlay topology structure, brings into play the advantage of tree topology high efficiency, low time delay, reduces again the cost that its topology is safeguarded as far as possible.The optimum tree-shaped core topology method for solving of reliability provided by the invention forms tree-shaped core topology for the live nerve of a covering overall situation or local backbone node, can improve the stability of nerve of a covering, reduces live time delay.With the comparison of random configuration tree topology, method provided by the invention can both significantly be improved in the reliability of tree topology and time delay; With the comparison of out-degree preferential construction tree topology, method reliability provided by the invention is obviously improved, and time delay approaches; With the comparison of simple node self reliability preferential construction tree topology, method reliability provided by the invention increases, and time delay significantly improves.
Accompanying drawing explanation
In order to make the present invention be convenient to understand, specific embodiments of the invention are described now by reference to the accompanying drawings.
Fig. 1 shows the topology diagram of the live nerve of a covering of P2P used in the present invention.
Fig. 2 shows the flow chart of the concrete implementation step of one embodiment of the present invention.
Fig. 3 shows the step schematic diagram of computing node accumulation reliability sum.
Fig. 4 shows the program code schematic diagram of individual (out-degree weighting array) variation in evolutionary process.
Fig. 5 shows the program code schematic diagram of the construction process that builds overlay tree structure.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
Fig. 1 shows the topology diagram of the live nerve of a covering of P2P used in the present invention.Wherein, 101 represent backbone node, and 102 represent ordinary node, and 103 represent the limit of tree-shaped core topology, and 104 represent the limit of mesh topology.The mode that the topology of the live nerve of a covering distribution of P2P adopts tree topology and mesh topology to mix conventionally.Because the efficient high and low time delay of tree topology a little, also has because the excessive shortcoming of topological maintenance cost that node dynamically causes.Tree topology is suitable for the stable node in live nerve of a covering, and mesh topology is suitable for more highly dynamic fringe nodes.The deficiency of safeguarding in order to overcome tree topology as far as possible, improves the applicability of its advantage performance, should select partially stabilized backbone node, and take reliability optimization as primary goal in tree topology structure, reduces time delay simultaneously.
The topological structure principle of the topology diagram of the live nerve of a covering of P2P used in the present invention is: according to the out-degree of each node of historical statistics (uploading bandwidth) and reliability, select out-degree and reliability in all nodes meet setting threshold node be backbone node, other nodes are ordinary node.
Wherein, described backbone node participates in structure tree topology, also participates in mesh topology structure.Backbone node is the integral multiple of Streaming Media speed for the out-degree (uploading bandwidth) of tree topology transmission, and remaining bandwidth is for the transmission of mesh topology.
Described ordinary node and described backbone node participate in constructing mesh topology jointly, and wherein backbone node is that mesh topology is only contributed the remaining bandwidth outside Streaming Media speed integral multiple.
Fig. 2 shows the flow chart of the concrete implementation step of one embodiment of the present invention.Step 201 represent to be selected backbone node, according to historical statistics, selects out-degree (being uploading bandwidth) to be greater than Streaming Media speed, and the node that reliability is greater than setting threshold is backbone node.Wherein, the reliability thresholds of described setting can realize by following two kinds of methods: the first, and setting fixed numbers is described threshold value, reliability is greater than the fixed numbers of described setting, is backbone node; The second, setting relative value is described threshold value, by specifying the ratio of backbone node quantity in total number of nodes to determine.
Step 202 represents to calculate out-degree weight array length, and the length of described out-degree weight array is the depth capacity of described tree topology, and the depth capacity of described tree topology structure adopts the mode of taking the logarithm, and formula is as follows:
Wherein, the quantity of the backbone node of choosing described in c represents, the average out-degree of the backbone node of choosing described in d represents, l represents the degree of depth of described tree topology structure.
Step 203 represents random generation
individuality, each individuality is that length is the out-degree weight array of l, wherein,
represent pre-set parameter.
Step 204 represents to form generation out-degree weight array population, and the scale of described out-degree weight array population is
, before described out-degree weight array population can derive from and evolve, in step 203, produce
individual out-degree weight array, can also derive from the selected more excellent individuality of step 208 in evolutionary process.Wherein, for deriving from, produce in step 203 before evolving
individual out-degree weight array, the employing step as shown in Figure 3 of calling of each individuality in population is calculated to its node accumulation reliability sum.
Step 205 represent to produce variation out-degree weight array population, and by the row variation of going forward side by side of each the individual replicate portion in the population in step 204, and the individual employing step as shown in Figure 3 of calling is calculated its node accumulation reliability sum to making a variation.The specific implementation of described variation is with reference to figure 4, and Fig. 4 shows the program code schematic diagram of individual (out-degree weighting array) variation in evolutionary process.
Step 206 represents to select to produce out-degree weight population of future generation, and all individualities before and after described variation are carried out to descending sort according to node accumulation reliability sum, selects
the preceding individuality of individual sequence is as population of new generation.
Step 207 represents to judge whether to meet the condition of evolving and finishing, according to the node accumulation reliability sum r of optimum individual
m(T),
the mean value r of the node accumulation reliability sum of the preceding individuality of individual sequence
a(T) and evolutionary generation g, whether judgement is evolved and is finished.The formula of evolution termination condition is as follows, and set up simultaneously (1) (2) or set up (3), evolves and finish:
(3) g≥g
m
Wherein,
setup parameter, r '
m(T), r '
a(T) be optimum and the mean value of the node accumulation reliability sum that previous generation (be also before evolutionary process in) is corresponding.
(1) (2) set up the maximum of the accumulation reliability sum that represents all nodes simultaneously, the variation of mean value is less than threshold value; (3) set up and represent that evolutionary generation reaches the upper limit.
Step 208 represent to obtain the optimum tree-shaped core topology of reliability, meets the corresponding tree of the optimum individual obtaining after evolution termination condition and be exactly the optimum tree-shaped core topology of reliability of the live nerve of a covering of P2P solving.
Fig. 3 shows the step schematic diagram of computing node accumulation reliability sum.Be in a preferred embodiment of the present invention, according to individual (the out-degree weighting array) structure of evolving, to set and calculate the node accumulation reliability sum of this tree.
Step 301 represents structure tree topology.According to out-degree, strengthen array, carry out layered structure tree structure, the specific implementation of described construction process is with reference to figure 5, and Fig. 5 shows the program code schematic diagram of the construction process that builds overlay tree structure.
Step 302 represents to calculate the node accumulation reliability sum of tree.According to the tree topology of structure in step 301, calculate all node accumulation reliability sums of this tree, each node accumulation reliability is calculated by following formula:
r
a(p)＝r(p)*∏
u∈path(p)r(u)
Wherein, path (p) represents all nodes except p the path from root node to node p.Described overlay tree reliability is calculated by following formula:
r(T)＝∑
p∈Tr
a(p)
Fig. 4 shows the program code schematic diagram of individual (out-degree weighting array) variation in evolutionary process.Wherein, input w represents out-degree weighting weight array, evolves individual; Input g represents current evolutionary generation; Input r represents the overlay tree that builds with the w rank in population; g
mrepresent maximum evolutionary generation;
represent population scale; δ represents the parameter of setting.
F
grepresent the evolutionary generation factor, by algebraic expression f
g=1-g/g
mobtain, wherein, g
mfor maximum evolutionary generation.
F
rrepresent the individual rank factor, unifier numerical expression f
r=r/
obtain,
for population scale.
Fig. 5 shows the program code schematic diagram of the construction process that builds overlay tree structure.The root node of the overlay tree of described structure is source node R, and described source node R is also backbone node.Input p represents that scale is the node set p (not comprising source node R in the node of p set) of c.Input w represents out-degree weighting weight array, and each in population evolved individual.
Wherein, the individual w={ ω that evolves of each in population
1, ω
2..., ω
δ] in ω
irepresent the selection of i node layer (level of root node is 0 layer).
Selection is undertaken by node being carried out to out-degree weighting reliability ordering.The out-degree weighting Calculation of Reliability of node is calculated by following formula:
r
w(P)＝r(p)+(d(p)-1)*w
i
R (p) represents the reliability of node set p, and d (p) represents the out-degree of node set p.
Foregoing detailed description has been illustrated the various embodiment of system and/or process by embodiment and/or schematic diagram.With regard to these schematic diagrames and/or comprise with regard to one or more functions and/or operation, it will be understood by those skilled in the art that each function in these schematic diagrames or embodiment and/or operation can by various hardware, software, firmware or in fact its combination in any come individually and/or jointly realize.
Should be appreciated that, method described herein can combined with hardware or software, or in conjunction with both combination, realize in due course.Therefore, method of the present invention, can adopt the program code that is included in tangible mediums such as floppy disk, CD-ROM, hard disk drive or any other machinable medium (, instruction) form, wherein, in the situation that program code is carried out on programmable calculator, computing equipment generally includes processor, readable storage medium (comprising volatile memory and/or memory element), at least one input equipment and at least one output equipment of this processor.One or more programs can be for example, and by using API, the process of describing in conjunction with the present invention is realized or utilized to reusable control etc.Such program preferably realizes with high level procedural or Object-Oriented Programming Language, to communicate by letter with computer system.Yet if needed, this program can realize by assembler language or machine language.In any case, language can be compiler language or interpretative code, and realizes and combining with hardware.
Although illustrate and described the present invention with reference to its preferred embodiment particularly, those skilled in the art will appreciate that and can make the various changes in form and details and not depart from the scope of the present invention described in appended claims.Above combination specific embodiments of the invention are described in detail, but are not limitation of the present invention.Every foundation technical spirit of the present invention, to any simple modification made for any of the above embodiments, all still belongs to the scope of technical solution of the present invention.
Claims (5)
1. the optimum tree-shaped core topology method for solving of the reliability of the live nerve of a covering of P2P, the optimum tree-shaped core topology method for solving of described reliability comprises:
(1) in the live nerve of a covering of described P2P, according to out-degree and reliability, choose backbone node;
(2) quantity of the backbone node of choosing described in basis and average out-degree are calculated the length of out-degree weight array;
(3), according to the length of out-degree weight array described in step (2), produce at random one group of out-degree weight array, using one group of out-degree weight array of described random generation as initial population;
(4) for each individuality in described initial population, construct corresponding tree topology, and calculate the accumulation reliability sum of all nodes of every tree topology;
(5) using the initial population described in the described initial population of step (4) or step (6) as Advanced group species, each individuality of described Advanced group species copied and made a variation, forming variation population; According to each individuality of described variation population, construct corresponding tree topology, and calculate the accumulation reliability sum of all nodes of every tree topology;
(6) described Advanced group species is carried out to sort descending, the initial population of selecting preceding half individuality of sequence to evolve as a new round with the described accumulation reliability sum that makes a variation node corresponding to all individualities of population;
(7) initial population of evolving for the described new round producing in step (6), whether meet the judgement of evolution termination condition, if do not meet described evolution termination condition, repeated execution of steps (5) is to step (7), until meet evolution termination condition;
(8) in step (7), meet the initial population of the described new round evolution of evolution termination condition, be the optimum tree-shaped core topology of reliability,
It is characterized in that, the length of described out-degree weight array is the depth capacity of described tree topology, and according to the quantity of the backbone node of choosing described in certificate and average out-degree, passes through formula:
Calculate the depth capacity of described tree topology, wherein
the quantity of the backbone node of choosing described in expression,
the average out-degree of the backbone node of choosing described in expression,
the degree of depth that represents described tree topology;
The condition that described satisfied evolution finishes comprises the maximum of accumulation reliability sum of all nodes and the variation of mean value is all less than corresponding threshold value or evolutionary generation reaches any one in the upper limit, wherein, by parameter preset, the maximum of the node accumulation reliability sum corresponding with previous generation in evolutionary process and mean value multiply each other and obtain each self-corresponding threshold value respectively.
2. the optimum tree-shaped core topology method for solving of the reliability of the live nerve of a covering of P2P as claimed in claim 1, wherein, the computational methods of the accumulation reliability of described node, comprising:
201, the accumulation reliability of the root node of described tree topology is himself reliability;
202, from root node described in step 201, range travels through all child nodes, and each child nodes is calculated to its accumulation reliability by step 203;
203, self reliability of described child nodes is multiplied by the accumulation reliability of its parent node as the accumulation reliability of described child nodes.
3. the optimum tree-shaped core topology method for solving of the reliability of the live nerve of a covering of P2P as claimed in claim 1, wherein, in the live nerve of a covering of described P2P, according to out-degree and reliability, choose in backbone node, described out-degree is greater than Streaming Media speed, and the node that described reliability is greater than reliability thresholds is chosen for backbone node.
4. the optimum tree-shaped core topology method for solving of the reliability of the live nerve of a covering of P2P as claimed in claim 3, wherein, described reliability thresholds realizes by setting fixed value.
5. the optimum tree-shaped core topology method for solving of the reliability of the live nerve of a covering of P2P as claimed in claim 3, wherein, described reliability thresholds realizes by the ratio of specifying backbone node quantity to account in total number of nodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210126532.0A CN102624596B (en) | 2012-04-26 | 2012-04-26 | Reliability optimal tree-shaped core topological solving method of P2P live broadcast covering network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210126532.0A CN102624596B (en) | 2012-04-26 | 2012-04-26 | Reliability optimal tree-shaped core topological solving method of P2P live broadcast covering network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102624596A CN102624596A (en) | 2012-08-01 |
| CN102624596B true CN102624596B (en) | 2014-11-05 |
Family
ID=46564287
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210126532.0A Expired - Fee Related CN102624596B (en) | 2012-04-26 | 2012-04-26 | Reliability optimal tree-shaped core topological solving method of P2P live broadcast covering network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102624596B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104901818B (en) * | 2015-05-13 | 2019-01-01 | 浙江传媒学院 | A kind of IP multicast is blended in the application layer multicast realization method and system of netted nerve of a covering |
| CN104821964B (en) * | 2015-05-13 | 2019-01-01 | 浙江传媒学院 | Across channel assistance method and system based on IP multicast group |
| CN105049315B (en) * | 2015-08-07 | 2018-12-07 | 浙江大学 | A kind of virtual network improvement mapping method based on virtual network segmentation |
| CN105897910B (en) * | 2016-05-20 | 2019-10-11 | 广州华多网络科技有限公司 | Data transmission method and transmission device |
| CN109951725B (en) * | 2019-03-07 | 2021-06-15 | 武汉斗鱼鱼乐网络科技有限公司 | Recommendation method for live broadcast room and related equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102395146A (en) * | 2011-12-09 | 2012-03-28 | 中国科学院上海微系统与信息技术研究所 | Multiple-target monitoring oriented method for sensing topology construction in wireless sensor network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7593353B2 (en) * | 2004-03-13 | 2009-09-22 | International Business Machines Corporation | Methods and apparatus for content delivery via application level multicast with minimum communication delay |
-
2012
- 2012-04-26 CN CN201210126532.0A patent/CN102624596B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102395146A (en) * | 2011-12-09 | 2012-03-28 | 中国科学院上海微系统与信息技术研究所 | Multiple-target monitoring oriented method for sensing topology construction in wireless sensor network |
Non-Patent Citations (9)
| Title |
|---|
| "Anysee: Multicast-based Peer-to-Peer Media Streaming Service System";Chao Zhang等;《2005 Asia-Pacific Conference on Communications》;20051005;第274-274页 * |
| Chao Zhang等."Anysee: Multicast-based Peer-to-Peer Media Streaming Service System".《2005 Asia-Pacific Conference on Communications》.2005,第274-278页. * |
| 秦丰林，刘琚."P2P网络流媒体关键技术".《电子学报》.2011,第39卷(第4期), * |
| 秦丰林,刘琚."P2P网络流媒体关键技术".《电子学报》.2011,第39卷(第4期), * |
| 贾晓雯，翁建广."多网融合承载的P2P直播覆盖网优化关键问题分析".《计算机应用于软件》.2011,第28卷(第4期), * |
| 贾晓雯,翁建广."多网融合承载的P2P直播覆盖网优化关键问题分析".《计算机应用于软件》.2011,第28卷(第4期), * |
| 郭彤城，慕春棣."并行遗传算法在一类计算机网络可靠性优化问题中的应用".《系统工程理论与实践》.2003,(第1期), * |
| 郭彤城,慕春棣."并行遗传算法在一类计算机网络可靠性优化问题中的应用".《系统工程理论与实践》.2003,(第1期), * |
| 黄思嘉."基于CDN和P2P树网混合的流媒体直播系统设计".《中国优秀硕士学位论文全文数据库信息科技辑》.2012,(第1期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102624596A (en) | 2012-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hu et al. | Decentralized federated learning: A segmented gossip approach | |
| CN102624596B (en) | Reliability optimal tree-shaped core topological solving method of P2P live broadcast covering network | |
| CN105049536A (en) | Load balancing system and load balancing method in IaaS (Infrastructure As A Service) cloud environment | |
| CN101777990B (en) | Method for selecting multi-objective immune optimization multicast router path | |
| CN106227599B (en) | The method and system of scheduling of resource in a kind of cloud computing system | |
| CN102075352A (en) | Method and device for predicting network user behavior | |
| Lin et al. | Research on resource self-organizing model for cloud computing | |
| CN115907038A (en) | Multivariate control decision-making method based on federated split learning framework | |
| CN101616074B (en) | Multicast routing optimization method based on quantum evolution | |
| Candan et al. | A dynamic island model for adaptive operator selection | |
| CN104219319A (en) | Method for distributed network flow self-organizing scheduling | |
| CN103581329A (en) | Construction method for topological structure based on clustered peer-to-peer network streaming media direct broadcast system | |
| CN102546230B (en) | Overlay-network topological optimization method in P2P (Peer-To-Peer) streaming media system | |
| Tsipis et al. | Qoe-aware rendering service allocation in fog-assisted cloud gaming environments | |
| TW201734831A (en) | Method of assigning application to assigned service cluster and device | |
| Harutyunyan et al. | New heuristic for message broadcasting in networks | |
| Chmaj et al. | Heuristic algorithms for optimization of task allocation and result distribution in peer-to-peer computing systems | |
| CN117596122A (en) | General calculation fusion network architecture and resource adaptation method | |
| Toka et al. | Analysis of user-driven peer selection in peer-to-peer backup and storage systems | |
| Shen et al. | CCLBR: Congestion control-based load balanced routing in unstructured P2P systems | |
| CN101741749A (en) | Method for optimizing multi-object multicast routing based on immune clone | |
| Fan et al. | Dual-eigenrep: a reputation-based trust model for P2P file-sharing networks | |
| Li et al. | Consistency aware update schedule in multi-server Distributed Virtual Environments | |
| CN102882727B (en) | Monitoring area partition method for hierarchical monitoring network | |
| CN117687800B (en) | Cross-domain distributed computing method, system, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141105 Termination date: 20180426 |