CN103916822A - Rapid redirection method and system for mobile application layer multicast - Google Patents
Rapid redirection method and system for mobile application layer multicast Download PDFInfo
- Publication number
- CN103916822A CN103916822A CN201410144152.9A CN201410144152A CN103916822A CN 103916822 A CN103916822 A CN 103916822A CN 201410144152 A CN201410144152 A CN 201410144152A CN 103916822 A CN103916822 A CN 103916822A
- Authority
- CN
- China
- Prior art keywords
- node
- multicast
- redirected
- father node
- backup
- 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
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a rapid redirection method and system for mobile application layer multicast. The rapid redirection method comprises the following steps that S1, after multicast nodes are added to a multicast system, and backup father nodes are set for the multicast nodes and are registered in family tables of the multicast nodes; S2, the multicast nodes in the multicast system are detected, when one multicast node will have a fault, a child node of the multicast node is informed, and rapid redirection is carried out through the backup father node of the child node; S3, after redirection is accomplished, updating is carried out on the family tables of the affected multicast nodes so as to ensure effectiveness of the backup father nodes of the multicast nodes. According to the rapid redirection method and system for mobile application layer multicast, before one multicast node has the fault, the child node of the multicast node is informed, so that the child node carries out rapid redirection through the backup father node.
Description
Technical field
The present invention relates to being redirected of multicast node, relate in particular to a kind of quick reorientation method and system of mobile application layer multicast.
Background technology
In recent years, along with developing rapidly of mobile communication technology and smart mobile phone, the demand of mobile flow medium service and day rapid an increasing.The more domestic Internet video company that is devoted to provide high definition, smooth professional video service for user, as Baidu, Sohu, excellent cruel etc., expands its streaming media service in mobile network one after another.Meanwhile, application layer multicast alleviates streaming media server load pressure and optimizes the P2P technology of streaming media service as a kind of, and the application under mobile environment is the focus of research in recent years always.
Because application layer multicast is compared IP multicast and is had the not enough problem of inborn stability, and the height dynamic of mobile environment has more been aggravated this problem, cause multicast node in multi-cast system to occur frequently extremely leaving, and cause its child nodes in multi-cast system to occur the problem that multicast services is interrupted, so that affect multicast user's experience.For the research of multi-cast system stability under mobile environment, and after multi-cast system breaks down, the child nodes of malfunctioning node fast redirected research is the difficult point of this research field always.
For after solving multi-cast system and breaking down, the child nodes of malfunctioning node can be carried out redirected problem fast, the previous research direction of order is in multi-cast system, detect and recover fault by heartbeat detection mechanism is applied to, and keeps synchronous between multicast node simultaneously.Whether its thought is to be detected and had multicast node to break down extremely to leave by timed sending heartbeat packet, can solve redirected problem on a small scale and in metastable multi-cast system.But under mobile environment, due to the frequent disconnection of multicast node, and originally the experience impact in heartbeat detection cycle of heartbeat detection mechanism, heartbeat detection mechanism is doomed not can solve the problem that is redirected fast.Another one research direction is to utilize redundant link mode, after fault occurs, utilize the link of redundancy to be redirected fast, but it is confined to the multicast tree fault of network level, do not consider the degree restriction of node in overlay network, so under mobile environment, multicast node is in the limited situation of mobile terminal itself, utilizing redundant link to solve the redirected problem of malfunctioning node, is also worthless.
Summary of the invention
Main purpose of the present invention is to propose application layer multicast system fail-over policy under a kind of mobile environment, solve multicast node height dynamic under mobile environment and bring multicast services disruption, and guarantee in the time that abnormal disconnection appears in multicast node, can be redirected fast, recover multicast services, promote multicast user and experience.
The technical solution adopted for the present invention to solve the technical problems is:
Quick reorientation method based on father node back mechanism in a kind of mobile application layer multicast is provided, comprises the following steps:
S1, add after multi-cast system in multicast node, for it arranges backup father node, and be registered in family's table of this multicast node;
S2, each multicast node in multi-cast system is detected, when a multicast node will break down therein, notify its child nodes, so that child nodes is redirected fast by backup father node;
S3, be redirected after, affected multicast node is carried out to the renewal of family's table, to guarantee the validity of multicast node backup father node.
In method of the present invention, step S1 is choosing backup when father node, mainly according to the time delay in multicast node adition process and degree constraint, in the brotgher of node of its father node or its grandfather's node, preferentially chooses, and specifically comprises:
S11, according to the family table of father node, obtain the nearest ancestor node of node and family's table;
S12, according to the family table of obtained ancestor node, send Pin to the brotgher of node of its ancestor node and father node
gmessage is carried out acquisition time stamp;
S13, according to obtained timestamp, get the node of time delay minimum as the backup father node of node.
In method of the present invention, in step S2, the multicast node that will break down is being notified after its child nodes, and its child nodes sends and suppresses alarm information to child node, and is shown to search backup father node and sent redirect message by family; In the time that the backup father node of child nodes is received redirect message, if the child nodes number connecting does not reach its degree constrained, share the forwarding task of data, otherwise consign to its father node or other child nodes complete redirected.
In method of the present invention, the method also comprises step:
S4, the strategy that is redirected fast of multi-cast system is evaluated.
5, method according to claim 1, is characterized in that, step S4 specifically comprises:
S41, before multi-cast system breaks down, the topological structure of multicast tree is carried out to vectorization, node go out the dimension of the number of degrees after as this knot vector;
S42, node complete be redirected after, again the topological structure of whole multicast tree is carried out to vectorization;
S43, calculate before and after fault according to calculating formula of similarity, the similarity of the topological structure of multicast tree, thus judgement is redirected tactful quality.
The present invention also provides the quick Redirectional system based on father node back mechanism in a kind of mobile application layer multicast, comprising:
Backup father node is chosen module, for adding after multi-cast system in multicast node, for this multicast node arranges backup father node, and is registered in family's table of this multicast node;
Backup father node redirection module, detects for each multicast node to multi-cast system, when a multicast node will break down therein, notifies its child nodes, so that child nodes is redirected fast by backup father node; After being redirected, affected multicast node is carried out to the renewal of family's table, to guarantee the validity of multicast node backup father node.
In system of the present invention, backup father node is chosen module choosing backup when father node, mainly according to the time delay in multicast node adition process and degree constraint, in the brotgher of node of its father node or its grandfather's node, preferentially chooses;
Backup father node is chosen module specifically for according to family's table of father node, obtains the nearest ancestor node of node and family's table; According to family's table of obtained ancestor node, send Pin to the brotgher of node of its ancestor node and father node
gmessage is carried out acquisition time stamp; According to obtained timestamp, get the node of time delay minimum as the backup father node of node.
In system of the present invention, this system also comprises:
Be redirected tactical comment module, evaluate for the strategy that is redirected fast to multi-cast system.
In system of the present invention, described redirected tactical comment module, specifically for before multi-cast system breaks down, is carried out vectorization to the topological structure of multicast tree, node go out the dimension of the number of degrees after as this knot vector; Node complete be redirected after, again the topological structure of whole multicast tree is carried out to vectorization; Calculate before and after fault according to calculating formula of similarity, the similarity of the topological structure of multicast tree, thus judgement is redirected tactful quality.
The beneficial effect that the present invention produces is: the strategy that proposed in addition should have to whole multi-cast system the feature that overhead is little, thereby meets the limited needs of mobile terminal itself.
In the present invention, choosing of backup father node, is independent of the adition process of multicast node, can finely carry out functional expansion to multi-cast system, and backup father node Algorithms of Selecting also can back up the selected of father node according to different strategies, has good flexibility.While being redirected, utilize initiatively accident analysis to find the multicast node that is about to break down, by gas, family shows, and can notify its child nodes, so that child nodes is redirected fast by backup father node in advance; After being redirected, upgrade the backup father node information of affected node, to guarantee to back up the validity of father node.
Further, counterweight directional strategy is evaluated, the similarity of the proposition multicast tree of innovation judge break down before and recover after fault, be redirected the impact of strategy on multi-cast system topologies change.
Accompanying drawing explanation
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing:
Fig. 1 is that the embodiment of the present invention moves the quick reorientation method flow chart based on father node back mechanism in application layer multicast;
Fig. 2 is the flow chart of embodiment of the present invention backup father node choosing;
Fig. 3 is embodiment of the present invention backup father node redirection mechanism schematic diagram;
Fig. 4 is that embodiment of the present invention backup father node is redirected flow process figure;
Fig. 5 is that the embodiment of the present invention moves the quick Redirectional system structural representation based on father node back mechanism in application layer multicast.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
The present invention proposes application layer multicast system fail-over policy under a kind of mobile environment.The active fault recovery scheme that this strategy utilization proposes and backup father node are redirected algorithm, thereby reach redirected object fast.The proposition that the present invention also innovates utilizes the similarity of multicast tree to quantize to be redirected the impact that algorithm changes multicast topology, thereby judgement is redirected tactful quality.Experimental result shows, the proposed by the invention strategy that is redirected fast has the advantage that time delay is low, and less to multicast topology variable effect after being redirected.
As shown in Figure 1, the embodiment of the present invention moves the quick reorientation method based on father node back mechanism in application layer multicast, comprises the following steps:
S1, in the time that multicast node adds multi-cast system, for it arranges backup father node, and be registered in the family table of this multicast node;
S2, each multicast node in multi-cast system is detected, when a multicast node will break down therein, notify its child nodes, so that child nodes is redirected fast by backup father node;
S3, be redirected after, affected multicast node is carried out to the renewal of family's table, to guarantee the validity of multicast node backup father node.
Wherein step S1 is choosing backup when father node, mainly according to the time delay in multicast node adition process and degree constraint, in the brotgher of node of its father node or its grandfather's node, preferentially chooses, and specifically comprises:
S11, according to the family table of father node, obtain the nearest ancestor node of node and family's table;
S12, according to the family table of obtained ancestor node, send Pin to the brotgher of node of its ancestor node and father node
gmessage is carried out acquisition time stamp;
S13, according to obtained timestamp, get the node of time delay minimum as the backup father node of node.
Further, in step S2, the multicast node that will break down is being notified after its child nodes, and its child nodes sends and suppresses alarm information to child node, and is shown to search backup father node and sent redirect message by family; In the time that the backup father node of child nodes is received redirect message, if the child nodes number connecting does not reach its degree constrained, share the forwarding task of data, otherwise consign to its father node or other child nodes complete redirected.
In a preferred embodiment of the present invention, the method also comprises step: S4, the strategy that is redirected fast of multi-cast system is evaluated.
Step S4 specifically comprises:
S41, before multi-cast system breaks down, the topological structure of multicast tree is carried out to vectorization, node go out the dimension of the number of degrees after as this knot vector;
S42, node complete be redirected after, again the topological structure of whole multicast tree is carried out to vectorization;
S43, calculate before and after fault according to calculating formula of similarity, the similarity of the topological structure of multicast tree, thus judgement is redirected tactful quality.
In a preferred embodiment of the present invention, the present invention moves in application layer multicast in the quick reorientation method based on father node back mechanism, at the design and the backup father node Algorithms of Selecting that back up family's table that choosing of father node comprise node.
The design of family's table, except preserving the information of father node and child node, also preserves backup father node information, as shown in table 1 below.
Table 1 is with family's table of backup father node
The backup father node Algorithms of Selecting of the embodiment of the present invention, referring to Fig. 2, as follows for its specific algorithm process of multicast node N:
(1) node N obtains family's table information by GetFTInfoFromNode, and judge whether its father node F is Server node, if, because can not breaking down, Server node extremely disconnects, so the backup father node of node N is set to Server, be registered in family's table of node N, algorithm completes, otherwise, carry out (2).
(2) node N obtains the nearest ancestor node A of node N by GetAncestorFromParent, and obtains family's table information by GetFTInfoFromNode.
(3) node A travels through its child nodes, and all nodes except the father node F of the child nodes that can not allow node to add and node N in its family's table are sent to Pin to node N
gmessage is stabbed with acquisition time.
(4) get the node of timestamp minimum, the node that time delay is minimum is as the backup father node of node N, and is registered in family's table of node N, and algorithm completes.
Backup father node Algorithms of Selecting also can back up the selected of father node according to different strategies, is not limited to above-mentioned algorithm, therefore has good flexibility.In backup father node Algorithms of Selecting, add algorithmic procedure owing to being independent of multicast node, and because family's table additional information is few, so concerning whole system, expense is less, can well be embedded into any multi-cast system, is certainly also applicable in mobile application layer multicast system.And, due to its independence, so can be according to different because different backup father node Algorithms of Selectings is usually set.
The judgement of multicast node fault is an Active recovery process, give warning in advance and avoid the out-of-service that do not have data to cause because of buffering area by active alarm mechanism, by oppositely suppressing warning strategies, reduce the quantity of repetition and false alarm simultaneously, get rid of and disturb for fault location.
Referring to Fig. 3, shown in backup father node redirection mechanism schematic diagram, due to multicast node v
3break down, its subtree multicast node v
6, v
7, v
8, v
9, v
10, v
14all will break down, now detect by active alarm mechanism and be that fault recovery is prepared, concrete steps are:
(1) as multicast node v
6, v
7detect when its buffering area reaches critical point, to its subtree v
8, v
9, v
10, v
14send and suppress alarm information.
(2) as node v
8, v
9, v
10, v
14receive and suppress after alarm information, will not carry out follow-up father node and be redirected work.
Reduce due to source of trouble node v through two steps above
3multiple nodes of causing of abnormal disconnection be redirected, reduce the expense of whole system.
Redirected algorithm of the present invention is that for malfunctioning node carries out redirect operation, referring to Fig. 4, concrete steps are on the basis of multicast node fault judgement:
(1) node N detects when its father node F will break down, and to backup father node, B sends REJOIN_REQUEST message;
(2) backup father node B receives after the REJOIN_REQUEST message from node N, if child nodes number does not reach the upper limit, send REJOIN_RESPONSE message to node N, otherwise the REJOIN_REQUEST message of node is forwarded to and in its child nodes, completes this by child nodes and operate, after having added equally, send REJOIN_RESPONSE message to node N, and carry out data retransmission preparation.
(3) node N receives after REJOIN_RESPONSE, monitors and waits for reception data, thereby completing redirect operation.
Referring to Fig. 3, v
3after breaking down, v
6, v
7to backup father node v
1send redirect request, due to v
1can be carried as v
6, v
7the expense of forwarding data, after being redirected, v
1for v
6, v
7forwarding data.
Be redirected algorithm quality, depend on the one hand the timely detection of fault, depend on the other hand choosing of backup father node.With respect to traditional heartbeat detection mechanism, by the active detecting to fault and quick redirect operation, thereby in the time that multi-cast system breaks down, can recover in time.
The renewal operation of family table is after completing and being redirected, to malfunctioning node, referring to the v in Fig. 3
6, v
7, v
8, v
9, v
10, v
14family table upgrade, operating process and backup father node Algorithms of Selecting are similar.
The renewal of family's table is mainly can be after fault occurs, and family's table of the malfunctioning node that upgrades in time, prevents from backing up father node Problem of Failure, and meanwhile, can solve because current network conditions changes the backup father node causing is not optimum problem yet.
In the time that counterweight directional strategy is passed judgment on, mainly carry out the vectorization of multicast tree and the similarity of multicast tree and calculate two parts.
The vectorization procedure of multicast tree is: for containing N(N>=2) multicast tree of individual node, according to start level traversal from root node, if child's number of node is followed successively by k
1, k
2... k
n, vector (k
1, k
2... k
n) be called the vector after this multicast tree vectorization, and definition is called the vectorization of multicast tree to vectorial process by multicast tree.
Referring to Fig. 3, before fault occurs, due to s, v
1, v
2..v
14child nodes number be 2,2,1..0, corresponding vector is (2,2,1,2,1,1,1,2,0,0,0,0,0,1,0).
For 0 be interpreted as after vectorization, due to leaving of leaf node in multicast tree, owing to there not being child nodes, so can there is not redirection process, concerning whole multi-cast system, leaving for the impact that in multicast tree, other node produces of leaf node ignored.When only containing root node in multicast tree, in multi-cast system, only have server, so now the corresponding vector of multicast tree is null vector o.
The similarity of multicast tree, the i.e. similarity degree of multi-cast system topological structure.Suppose that whole multi-cast system is before node breaks down, reach a stable state, so after multi-cast system breaks down, after redirect operation, recover fault, the topological structure of multi-cast system so now to break down before more similar, so also just more stable, be redirected strategy better.
The similarity of multicast tree has to give a definition: if the identical corresponding vector of two multicast tree of nodes is A, B, definition cos<A, the value of B> is referred to as the similarity of two multicast tree, and there are 0<cos<A, B>≤1 for any two nodes same multicast trees.
The similarity of multicast tree is had to following expansion:
(1) for the different multicast tree of two nodes (k1 ≠ 0), corresponding vector is A, and B, as A=(k
1, k
2... k
m), B=(k
1', k
2' ... km ' ... k
n') (n>m), for more vectorial A, the similarity of B, by A=(k
1, k
2... k
m, 0,0 ... 0) expand to n dimension, and supplement in the back n-m the component that value is 0, before and after expansion, can not affect the similarity of two multicast tree.
(2) if two multicast tree topological structures are identical, the vectorial A of its correspondence, B should meet A=B, and cos<A, B>=1.
(3) for any two multicast tree, the vectorial A of its correspondence, B necessarily meets 0≤cos<A, B>≤1
The vectorization of the similarity of multicast tree based on multicast tree, general computational process is as follows:
(1) before multicast tree breaks down, carry out vectorization procedure one time, referring to Fig. 3, multicast node v
3before breaking down, after vectorization, result is A(2,2,1,2,1,1,1,2,0,0,0,0,0,1,0).
(2) after this fault is recovered, i.e. node v
6, v
7after completing and being redirected, carrying out a vectorization, vectorization result is B(2,3,1,1,2,1,1,0,0,1,0,0,0,0).
(3) the corresponding vector of topological structure of multicast tree after being redirected is expanded, end adds 0, and now corresponding vector is B(2, and 3,1,1,2,1,1,1,0,0,1,0,0,0,0,0).
(4) according to calculating formula of similarity cos<A, before and after B> calculates and breaks down, the similarity of multi-cast system topological structure.
The present invention moves the quick Redirectional system based on father node back mechanism in application layer multicast, as shown in Figure 5, comprising:
Backup father node is chosen module, the multicast node being responsible in multi-cast system arranges backup father node, prepares for being redirected, and is mainly used in adding after multi-cast system in multicast node, for this multicast node arranges backup father node, and be registered in family's table of this multicast node.In the present invention, backup father node chooses module and multi-cast system structure algorithm has independence, can complete according to different backup father node Algorithms of Selectings the registration of multicast node family table simultaneously.
Backup father node redirection module, detects for each multicast node to multi-cast system, when a multicast node will break down therein, notifies its child nodes, so that child nodes is redirected fast by backup father node; After completing and being redirected, carry out the renewal of family's table, to guarantee the validity of multicast node backup father node.In the time that multicast node is about to or has broken down, backup father node redirection module can be redirected fast according to the backup father node in family's table of its child nodes, thereby does not affect the Streaming Media experience of the child nodes of malfunctioning node.
In the embodiment of the present invention, backup father node is chosen design and the backup father node Algorithms of Selecting that module comprises multicast node family table.The design of multicast node family table, except preserving under normal circumstances the information of this node father node and child nodes, also subsidiary backup father node information.
Backup father node Algorithms of Selecting normally constrains in the brotgher of node of its father node or its grandfather's node and preferentially chooses according to time delay, degree in multicast node adition process, specifically comprises:
(1), according to family's table of father node, obtain the nearest ancestor node of node and family's table;
(2), according to family's table of obtained ancestor node, send Ping message to the child nodes (being the brotgher of node of father node, except the father node of multicast node and the node of degree constrained) of its ancestor node and ancestor node and carry out acquisition time stamp;
(3), according to obtained timestamp, get the node of time delay minimum as the backup father node of node.
In preferred embodiment of the present invention, backup father node is chosen module choosing backup when father node, mainly according to the time delay in multicast node adition process and degree constraint, in the brotgher of node of its father node or its grandfather's node, preferentially chooses.
Backup father node is chosen module specifically for according to family's table of father node, obtains the nearest ancestor node of node and family's table; According to family's table of obtained ancestor node, send Pin to the brotgher of node of its ancestor node and father node
gmessage is carried out acquisition time stamp; According to obtained timestamp, get the node of time delay minimum as the backup father node of node.
Fault detect of the present invention, in the time that multicast node will break down, detects in time and notifies to its child nodes, detects with respect to common heartbeat mechanism, and be an Active recovery process.
Be redirected algorithm in the time that multicast node breaks down, can recover in time multicast services, specifically comprise:
(1) child nodes of source of trouble node suppresses, alarm information, shown to search backup father node and sent redirect message by family except sending to child node.
(2), in the time that backup father node is received the redirect message from node, if the child nodes number connecting does not reach its degree constrained, share the forwarding task of data, otherwise consign to its father node or child nodes completes this step.
The renewal of family table is after multicast node completes and is redirected, and affected multicast node is carried out to the renewal of family's table, to guarantee the validity of backup father node of multicast node.
Be combined with by choosing module with backup father node, backup father node redirection module can, in the time that multi-cast system breaks down, be recovered multicast services in time, and has considered degree restricted problem and the system resources consumption problem of multicast node.
In a preferred embodiment of the present invention, as shown in Figure 5, this system also comprises redirected tactical comment module, evaluates for the strategy that is redirected fast to multi-cast system.Mainly quantize to be redirected the size of algorithm on the impact of multi-cast system topological structure according to the multicast tree similarity proposing, thereby judgement is redirected tactful quality.
This redirected tactical comment module, specifically for before multi-cast system breaks down, is carried out vectorization to the topological structure of multicast tree, node go out the dimension of the number of degrees after as this knot vector; Node complete be redirected after, again the topological structure of whole multicast tree is carried out to vectorization; Calculate before and after fault according to calculating formula of similarity, the similarity of the topological structure of multicast tree, thus judgement is redirected tactful quality.Wherein, the vectorization of multicast tree, refers to the topological structure of multi-cast system is carried out to vectorization.Specifically according to being that server node starts level traversal from root node, and node gone out to the dimension of the number of degrees after this knot vector.The similarity of multicast tree, refer to multi-cast system is broken down before and fault recovery after, the similarity degree of the topological structure of multi-cast system topological structure multi-cast system.
Be redirected tactical comment module and carry out Real-time Feedback as counterweight directional strategy, the proposition of innovation utilize the similarity of multicast tree to calculate the size of multi-cast system change in topology before and after fault occurs, and evaluate as an index counterweight directional strategy.This redirected tactical comment module, be easy to calculate, and expense is less concerning whole system, be suitable for equally in large-scale multi-cast system, in addition, this module also can be applicable on any multi-cast system, have independence, the similarity of multicast tree can be used as the evaluation index of multi-cast system.
To sum up, the present invention proposes a kind of mobile application layer multicast system recovery mechanism, can be effectively applied in limited mobile terminal, and can be good at solving the problem of frequently leaving that the height dynamic of multicast node is brought.For the recovery mechanism of multi-cast system, the present invention proposes the similarity concept of multicast tree, before fault occurs and after fault recovery, according to the similarity degree of the topological structure of multi-cast system, recovery mechanism is evaluated, there is good feasibility.
Should be understood that, for those of ordinary skills, can be improved according to the above description or convert, and all these improvement and conversion all should belong to the protection range of claims of the present invention.
Claims (9)
1. the quick reorientation method based on father node back mechanism in mobile application layer multicast, is characterized in that, comprises the following steps:
S1, add after multi-cast system in multicast node, for it arranges backup father node, and be registered in family's table of this multicast node;
S2, each multicast node in multi-cast system is detected, when a multicast node will break down therein, notify its child nodes, so that child nodes is redirected fast by backup father node;
S3, be redirected after, affected multicast node is carried out to the renewal of family's table, to guarantee the validity of multicast node backup father node.
2. method according to claim 1, is characterized in that, step S1 is choosing backup when father node, mainly according to the time delay in multicast node adition process and degree constraint, in the brotgher of node of its father node or its grandfather's node, preferentially chooses, and specifically comprises:
S11, according to the family table of father node, obtain the nearest ancestor node of node and family's table;
S12, according to the family table of obtained ancestor node, send Pin to the brotgher of node of its ancestor node and father node
gmessage is carried out acquisition time stamp;
S13, according to obtained timestamp, get the node of time delay minimum as the backup father node of node.
3. method according to claim 1, it is characterized in that, in step S2, the multicast node that will break down is being notified after its child nodes, its child nodes sends and suppresses alarm information to child node, and is shown to search backup father node and sent redirect message by family; In the time that the backup father node of child nodes is received redirect message, if the child nodes number connecting does not reach its degree constrained, share the forwarding task of data, otherwise consign to its father node or other child nodes complete redirected.
4. method according to claim 1, is characterized in that, the method also comprises step:
S4, the strategy that is redirected fast of multi-cast system is evaluated.
5. method according to claim 4, is characterized in that, step S4 specifically comprises:
S41, before multi-cast system breaks down, the topological structure of multicast tree is carried out to vectorization, node go out the dimension of the number of degrees after as this knot vector;
S42, node complete be redirected after, again the topological structure of whole multicast tree is carried out to vectorization;
S43, calculate before and after fault according to calculating formula of similarity, the similarity of the topological structure of multicast tree, thus judgement is redirected tactful quality.
6. the quick Redirectional system based on father node back mechanism in mobile application layer multicast, is characterized in that, comprising:
Backup father node is chosen module, for adding after multi-cast system in multicast node, for this multicast node arranges backup father node, and is registered in family's table of this multicast node;
Backup father node redirection module, detects for each multicast node to multi-cast system, when a multicast node will break down therein, notifies its child nodes, so that child nodes is redirected fast by backup father node; After being redirected, affected multicast node is carried out to the renewal of family's table, to guarantee the validity of multicast node backup father node.
7. system according to claim 6, it is characterized in that, backup father node is chosen module choosing backup when father node, mainly according to the time delay in multicast node adition process and degree constraint, in the brotgher of node of its father node or its grandfather's node, preferentially chooses;
Backup father node is chosen module specifically for according to family's table of father node, obtains the nearest ancestor node of node and family's table; According to family's table of obtained ancestor node, send Pin to the brotgher of node of its ancestor node and father node
gmessage is carried out acquisition time stamp; According to obtained timestamp, get the node of time delay minimum as the backup father node of node.
8. system according to claim 6, is characterized in that, this system also comprises:
Be redirected tactical comment module, evaluate for the strategy that is redirected fast to multi-cast system.
9. system according to claim 8, is characterized in that, described redirected tactical comment module, specifically for before multi-cast system breaks down, is carried out vectorization to the topological structure of multicast tree, node go out the dimension of the number of degrees after as this knot vector; Node complete be redirected after, again the topological structure of whole multicast tree is carried out to vectorization; Calculate before and after fault according to calculating formula of similarity, the similarity of the topological structure of multicast tree, thus judgement is redirected tactful quality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410144152.9A CN103916822B (en) | 2014-04-11 | 2014-04-11 | The quick reorientation method and system of Mobile solution layer multicast |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410144152.9A CN103916822B (en) | 2014-04-11 | 2014-04-11 | The quick reorientation method and system of Mobile solution layer multicast |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103916822A true CN103916822A (en) | 2014-07-09 |
| CN103916822B CN103916822B (en) | 2017-12-22 |
Family
ID=51042139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410144152.9A Expired - Fee Related CN103916822B (en) | 2014-04-11 | 2014-04-11 | The quick reorientation method and system of Mobile solution layer multicast |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103916822B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106412023A (en) * | 2016-09-05 | 2017-02-15 | 南京臻融软件科技有限公司 | Multi-source data distribution method based on distribution subscription mechanism |
| CN107566992A (en) * | 2017-08-28 | 2018-01-09 | 华中师范大学 | Mobile solution layer multicast tree structure and restoration methods based on node combination property |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1588862A (en) * | 2004-06-30 | 2005-03-02 | 中国科学院计算技术研究所 | Data restoring method for application layer group broadcast |
| CN101873219A (en) * | 2010-05-31 | 2010-10-27 | 中山大学深圳研究院 | System for live broadcast of video conference on basis of application-layer multicast services |
| CN101958845A (en) * | 2010-11-02 | 2011-01-26 | 武汉大学 | Method for recovering multicast routing tree by proactive reconstruction in overlay network environment |
| CN102694833A (en) * | 2011-03-22 | 2012-09-26 | 日电（中国）有限公司 | Multicast tree building method, and server and node in multicast system |
-
2014
- 2014-04-11 CN CN201410144152.9A patent/CN103916822B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1588862A (en) * | 2004-06-30 | 2005-03-02 | 中国科学院计算技术研究所 | Data restoring method for application layer group broadcast |
| CN101873219A (en) * | 2010-05-31 | 2010-10-27 | 中山大学深圳研究院 | System for live broadcast of video conference on basis of application-layer multicast services |
| CN101958845A (en) * | 2010-11-02 | 2011-01-26 | 武汉大学 | Method for recovering multicast routing tree by proactive reconstruction in overlay network environment |
| CN102694833A (en) * | 2011-03-22 | 2012-09-26 | 日电（中国）有限公司 | Multicast tree building method, and server and node in multicast system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106412023A (en) * | 2016-09-05 | 2017-02-15 | 南京臻融软件科技有限公司 | Multi-source data distribution method based on distribution subscription mechanism |
| CN107566992A (en) * | 2017-08-28 | 2018-01-09 | 华中师范大学 | Mobile solution layer multicast tree structure and restoration methods based on node combination property |
| CN107566992B (en) * | 2017-08-28 | 2020-02-18 | 华中师范大学 | Method for constructing and recovering multicast tree of mobile application layer based on comprehensive performance of nodes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103916822B (en) | 2017-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2017092400A1 (en) | Failure recovery method and device, controller, and software defined network | |
| CN106131919B (en) | A kind of synchronous method of non-stop layer self-organized network topology | |
| CN102137017A (en) | Working method and device used for virtual network unit | |
| WO2019101244A3 (en) | Consensus system downtime recovery | |
| WO2012167624A1 (en) | Fault processing method, aggregation node and optical network protection system | |
| US20140185429A1 (en) | Communication system, path switching method and communication device | |
| CN102833108A (en) | Processing method and equipment of location information of fault point | |
| CN102546427B (en) | OSPF (Open Shortest Path First) protocol-based graceful restart (GR) method and router | |
| CN102624635A (en) | Method and device for realizing graceful restart | |
| CN103825759A (en) | Method and system for acquiring network topological information | |
| CN102082710A (en) | Method and system for implementing detection on network topology fault of nodes in network | |
| CN105095008A (en) | Distributed task fault redundancy method suitable for cluster system | |
| CN109462533B (en) | Link switching method, link redundancy backup network and computer readable storage medium | |
| WO2016062166A1 (en) | Method, apparatus and system for network operations, administration and maintenance | |
| CN101192959B (en) | A recovery method for multicast service connection in automatic switching optical network | |
| CN105530115A (en) | Method and device for realizing operation management and maintenance function | |
| CN103916822A (en) | Rapid redirection method and system for mobile application layer multicast | |
| Duolikun et al. | Power consumption models for redundantly performing mobile-agents | |
| Du et al. | Application-oriented fault detection and recovery algorithm for wireless sensor and actor networks | |
| Imran et al. | Application-centric recovery algorithm for wireless sensor and actor networks | |
| CN103414591A (en) | Method and system for fast converging when port failure is recovered | |
| CN102647424B (en) | Data transmission method and data transmission device | |
| CN103647709A (en) | ARP form item establishing method and device | |
| WO2017152595A1 (en) | Method and device for responding to network topology change | |
| WO2016177211A1 (en) | Address resolution protocol (arp) table entry synchronization method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171222 Termination date: 20200411 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |