CN100370780C - Method for automatic adjusting business path for flexible packet ring network - Google Patents

Method for automatic adjusting business path for flexible packet ring network Download PDF

Info

Publication number
CN100370780C
CN100370780C CNB2004100009657A CN200410000965A CN100370780C CN 100370780 C CN100370780 C CN 100370780C CN B2004100009657 A CNB2004100009657 A CN B2004100009657A CN 200410000965 A CN200410000965 A CN 200410000965A CN 100370780 C CN100370780 C CN 100370780C
Authority
CN
China
Prior art keywords
node
business
message
sending direction
ssr
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
Application number
CNB2004100009657A
Other languages
Chinese (zh)
Other versions
CN1642128A (en
Inventor
江健
郑�勋
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.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CNB2004100009657A priority Critical patent/CN100370780C/en
Publication of CN1642128A publication Critical patent/CN1642128A/en
Application granted granted Critical
Publication of CN100370780C publication Critical patent/CN100370780C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The present invention provides a method for automatically adjusting a service path on a resilient packet ring network. When the congestion of the ring network occurs, nodes generating congestion can automatically acquire congestion information, and the nodes on the RPR network can automatically adjust the transmitting direction of a service by using a service adjusting protocol (RSSP) to eliminate congestion condition; thus, the utilization rate of a loop circuit is improved. When service condition on the loop circuit is changed, namely that after the congestion condition is changed, the nodes in the ring network adjusts the transmitting direction of the loop circuit service again according to the new condition of the current loop circuit to realize the purpose of automatically adjusting a service path by the nodes on the RPR ring network. By using the method, the present invention can limberly and effectively use the bandwidth of the loop circuit, avoid the limitation of manual configuration and further simplify service management. The RSSP protocol used by the present invention occurs in a form of an RPR protocol control frame; the RSSP protocol is easy to comprehend, and is compatible with the RPR protocol.

Description

A kind of in the online method of adjusting service path automatically of Resilient Packet Ring
Technical field
The present invention relates to RPR network (RPR) technical field, be meant a kind of method of on the RPR looped network, adjusting service path automatically especially.
Background technology
RPR network (RPR) is a kind of based on ethernet frame format, data-oriented a kind of looped network technology; this technology has been utilized the statistic multiplexing of service quality (QOS) characteristic, the Ethernet of ATM, the protective feature of succinct and synchronous digital network, data services that can be efficient, safe.
The RPR looped network adopts the mode of dicyclo work, and in order to distinguish two rings, one is referred to as interior ring, and another is referred to as outer shroud.RPR and transmission medium are irrelevant, therefore can expand based on the RPR looped network of RPR technology, and the RPR technology adopt is distributed management, congested control and protection mechanism, and therefore the RPR looped network based on the RPR technology possesses the ability that sends data according to the grade of service.
RPR provides the auto discovery mechanism of loop node, can monitor the state of loop node, and generates loop topology and routing iinformation thus, is that business to be sent is selected optimal path automatically according to the topology information of loop.According to the RPR standard, professional transmitting-receiving path is to select according to the principle of path optimum, promptly, in the RPR looped network, mail to the business of another node,, need through M node if press the transmission of outer shroud direction if need through N node by interior ring direction transmission from a node, if N<M, then this business will be transmitted by interior ring direction, if N>M, then this business will be transmitted by the outer shroud direction.
In order to narrate conveniently, stipulate that at this clockwise direction is an outer shroud in all illustrated RPR looped networks of this paper, be interior ring counterclockwise, and outer shroud represent that with solid line interior ring dots.
Figure 1 shows that the RPR looped network schematic diagram of forming by 5 nodes.This RPR loop is made of A, B, C, D, these 5 nodes of E, and the data service relation is arranged between the AC point.Loop topology information according to the generation of RPR agreement, the A node finds that it mails to the data of C node, transmitting from the outer shroud direction need be through two node of E, D, transmitting from interior ring direction only need be through a B node, therefore, can select interior ring direction to transmit by the data that the A node sends to the C node this moment, and the data that sent to the A node by the C node then select the outer shroud direction to transmit.
Suppose on this loop except from the business of C->A, the business that also has C->B, and each 600M of business two, according to the routing policy of RPR, two business finally all can be present in outer shroud so, because loop bandwidth is 1G, so will produce congested on the outer shroud, be subjected to the intervention of loop fair algorithm, the business of C->A and C->B all can not reach 600M, and the bandwidth on the ring is in idle condition in this moment.
The defective of said method is: may exist a certain loop owing to data payload causes the loop overload greatly on the RPR looped network, and another loop is not because load gently also reaches the situation of flowing full amount.To occur on the RPR looped network like this, a loop can not be finished professional transmission well, and meanwhile the looped network bandwidth of another loop is in the phenomenon of idle condition.
In order to address the above problem, the applicant number is " 03137714.9 " in Chinese patent application, denomination of invention is in the patent of invention of " a kind of realize method that the data service transmission path is selected ", provide a kind of user of permission that service route is carried out the method for static configuration, thereby solved the irrationality of RPR technology to a certain extent with the routing of hop-by-hop mode.Referring to Fig. 1, the user makes the business of C->A send from interior environment-development according to the path of C->D->E->A by static configuration, and C->B still sends from outer shroud, and such two business can both reach 600M, and the service route after the configuration as shown in Figure 2.
The service traffics of supposing C->B this moment have been reduced to 200M, the service traffics of C->A still are 600M, and the burst service of 600M has appearred between D->E, because the business of C->A is to transmit from interior ring according to the mode of static configuration, it can not change the path at any time, therefore, will occur on the interior ring of RPR looped network congested.In fact, if the business of C->A is transmitted according to the order of C->B->A at this moment, ring occurs congested in can avoiding.
The defective of said method is: when the path of service route is during by user's static configuration, it can't change the service rate on the loop make timely reaction.If variation has taken place the service distribution situation on the loop, in order to make full use of loop bandwidth, the user need reconfigure the transmission path of each bar business.Business on the RPR loop mostly is paroxysmal business, and this method by the manual static configuration service path has significant limitation, and is unpractical.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of method of on the RPR looped network, adjusting service path automatically, make the node on the RPR looped network adjust professional sending direction automatically according to distribution situation professional on the current looped network.
For achieving the above object, technical scheme of the present invention is achieved in that
A kind of in the online method of adjusting service path automatically of Resilient Packet Ring, this method may further comprise the steps:
A, when traffic congestion occurring on the looped network, produce congested node and will obtain congestion information automatically, and start a business reorganization cycle, simultaneously along the opposite direction that produces congestion condition place loop, send business reorganization message SSR message to adjacent downstream node;
B, the node that receives SSR message judge whether the business that self sends is subjected to the influence of congestion condition, if be not affected, execution in step c then, otherwise after switching the sending direction of this node business, judge whether this node still is subjected to the influence of congestion condition, and whether received SSR message is to return message, and described recurrence message is meant by node and sends, and gets back to the SSR message of this node along looped network after one week again;
If still be subjected to the influence of congestion condition behind the sending direction of this node business of switching, and the SSR message of being received is to return message, then the business recovery with this node is to switch preceding state, and after other node broadcasts business reorganization failure SSC message in looped network, execution in step d;
If still be subjected to the influence of congestion condition behind the sending direction of this node business of switching, but the SSR message of being received is not to return message, then the business recovery with this node is to switch preceding state, and after adjacent downstream node is transmitted SSR message, repeated execution of steps b;
If no longer be subjected to the influence of congestion condition behind the sending direction of this node business of switching, and the SSR message of being received is to return message, then this node keeps the sending direction of current business constant, and after other node broadcasts business reorganization success SSS message in looped network, execution in step e;
If no longer be subjected to the influence of congestion condition behind the sending direction of this node business of switching, but the SSR message of being received is not to return message, then this node keeps the sending direction of current business constant, and after adjacent downstream node is transmitted SSR message, repeated execution of steps b;
C, receive SSR message node along the opposite direction that produces congestion condition place loop, after adjacent downstream node sends SSR message, execution in step b;
D, receive the node of SSC message, the professional sending direction of this node is reverted to sending direction before switching, finish this business reorganization cycle, all nodes enter init state;
E, receive the node of SSS message, keep the professional current sending direction of this node, finish this business reorganization cycle, all nodes enter init state.
Preferably, this method further comprises: when looped network was in stable state, each node was lined up respectively for the business that this node is sent to other node in advance, and each lines up corresponding destination node, and each is lined up to monitor.
Preferably, step b is described to judge that the method whether business that self sends is subjected to the congestion condition influence is: judge with produce that congested node is corresponding and line up in whether professional the existence is arranged, and should business in this business reorganization cycle, also not have processed.
Preferably, described step a further comprises: produce congested node, whether another loop of at first judging RPR network RPR looped network allows to switch, if, then enter a business reorganization cycle again, and along the opposite direction that produces congestion condition place loop, send SSR message to adjacent downstream node, otherwise it is constant to keep the professional sending direction of this node.
Preferably, the congested node of the described generation of step a is after adjacent downstream node sends SSR message, further comprise: the PWI timer that starts this intranodal, if this node is after agreement waits for that the PWI timer then at interval, business in the loop is not still adjusted successfully, or do not receive receiveing the response of other node, then the business recovery of all nodes is to switch preceding sending direction in the loop, and finishes this adjustment cycle.
Preferably, the node of the described SSR of the receiving message of step b judges that the business that self sends is subjected to the influence of congestion condition, and after switching the sending direction of this node business, further comprise: what start this intranodal switches abnormal end FSI timer, behind the FSI timer expiry, carry out subsequent step again.
Preferably, the timing length of described FSI timer realizes the stabilization time of loop fair algorithm for the RPR agreement.
Preferably, the timing length of described PWI timer is to determine according to professional what in the number of node in the current loop and the current loop.
Preferably, described definite method is: the timing length of PWI timer equals interstitial content and multiply by the FSI time, multiply by the professional number that this node is influenced by congestion condition again.
Preferably, before congested other node broadcasts SSS message of node in looped network of the described generation of step b, further comprise: start the SSS timer of this intranodal, after this SSS timer expiry, if congestion condition does not appear in this node again, then carry out subsequent step again; Before this SSS timer expiry, if congestion condition appears again in this intranodal, wait for that then the PWI timer then after, other node broadcasts SSC message in looped network, and execution in step d.
Preferably, the timing length of described SSS timer is the multiple that the FSI timer is fixed time, and is generally the time span of 2 FSI.
Preferably, this method further comprises: when node received the check field FCS message that is triggered by the user, all nodes all reverted to the professional sending direction that is determined by the default routing algorithm of RPR with the professional sending direction of self.
Preferably, the method for the sending direction of described this node of node switches business is:
The node that carries out switching services with self all professional sending directions all switch, or
The node that carries out switching services to self all professional sending directions at first judge, will be through the preferred SSP source node of business reorganization, and switch in the path sending direction of long business, if after switching, still there is congestion condition in this node on the congested direction of generation, then again will be through the SSP source node, and the sending direction of path vice-minister's business is switched, by that analogy, till having switched all business, or work as node till congestion condition disappears on the congested direction of generation through the SSP source node.
Applied business object queue (SDQ of the present invention, Service Destination Queue) formation comes the current service distribution situation of description node, when appearance in the looped network is congested, produce congested node and can obtain this congestion information automatically, and applied business is adjusted agreement (RSSP, RPR Service SwitchProtocol) makes the node on the RPR looped network adjust professional sending direction automatically, eliminate congestion condition, thereby improved the utilance of loop.Service conditions on loop changes, be after congestion condition changes, node in the looped network can carry out the adjustment of the professional sending direction of loop once more according to current loop new situation, thereby has realized that the node on the RPR looped network adjusts service path this purpose automatically.Use the present invention, can be flexibly, effectively utilize loop bandwidth, avoided the limitation of manual configuration, and then simplified management to business, the applied RSSP agreement of the present invention is to occur with the form of RPR agreement control frame, its easy to understand, and good compatibility is arranged with the RPR agreement.
Description of drawings
Figure 1 shows that the RPR looped network schematic diagram of forming by 5 nodes;
Figure 2 shows that business shown in Figure 1 is carried out schematic diagram after the static configuration;
Figure 3 shows that a SDQ schematic diagram on the RPR node;
Figure 4 shows that the network layer schematic diagram of RSSP agreement work;
Figure 5 shows that the collaborative work schematic diagram of RSSP agreement and loop service channel key-course;
Figure 6 shows that the message packet form of RSSP agreement;
Figure 7 shows that the schematic diagram of state machine on the individual node;
Figure 8 shows that the schematic diagram of using one embodiment of the invention;
Figure 9 shows that the node 3 of using the embodiment of the invention sends the schematic diagram of SSR message to node 2;
Figure 10 shows that the schematic diagram of the node 2 of the application embodiment of the invention to node 1 forwarding SSR message;
The PSM4 that Figure 11 shows that the node 1 of using the embodiment of the invention adjusts the schematic diagram behind the professional sending direction;
The PSM5 that Figure 12 shows that the node 1 of using the embodiment of the invention adjusts the schematic diagram behind the professional sending direction;
Figure 13 shows that the node 3 of using the embodiment of the invention sends the schematic diagram of broadcast.
Embodiment
For making technical scheme of the present invention clearer, again the present invention is described in further details below in conjunction with drawings and the specific embodiments.
After the RPR looped network was set up, topological agreement in the RPR agreement and loop choice mechanism can be determined the shortest path of certain node to specific node traffic, and the RPR agreement can detect the traffic congestion that occurs on the loop and report congestion information.The present invention realizes on the basis of above-mentioned these two mechanism.
Thinking of the present invention is: occurred the switching services condition on looped network, as when traffic congestion having occurred, produce congested node and can obtain this information automatically, and applied business is adjusted agreement (RSSP, RPRService Switch Protocol) make the node on the looped network adjust its professional sending direction, eliminate congestion condition, thereby improve the utilance of loop.Service conditions on loop changes, be after congestion condition changes, node in the looped network carries out the adjustment of the professional sending direction of loop once more according to current loop new situation, thereby has realized that the node on the RPR looped network adjusts professional sending direction this purpose automatically.
To adjust professional sending direction according to current loop service conditions, must be described loop service distribution situation.In the present invention, applied business object queue (SDQ, Service DestinationQueue) realizes the description to the service distribution situation, and the specific descriptions method is as follows:
Before business enters the RPR loop, each node is the business that this node is sent to other node line up respectively (only limit to influenced by fair algorithm business), promptly according to the destination address difference that enters the loop business, respectively with each business setting in different formations, and make a destination node on the corresponding loop of each formation, and monitor the situation of each formation, as whether business, service transmission rate etc. are arranged in the formation.In current RPR loop agreement, support 255 nodes on the loop at most, therefore under the optimal situation, on each node, should there be 255 SDQ, wherein, 254 SDQ are used to monitor the business of this node to other node, and a remaining SDQ is used for the queuing of broadcasting service.The SDQ resource can be dynamically application, it also can be static configuration, owing on a node of loop, exist the situation of a plurality of business that are sent to other node also few simultaneously, certainly, except the broadcasting service, because broadcasting service queuing separately, therefore, dynamically application SDQ can conserve system resources, but the static better performances that SDQ is set.In the present embodiment, set SDQ resource is dynamically application, when promptly having only this node to have certain node professional, just creates the SDQ of this business correspondence on this node.
Figure 3 shows that a SDQ schematic diagram on the RPR node.Business among the SDQ realizes transmitting on the inner and outer ring of RPR by the data path of medium access control (MAC) after entering loop.The corresponding destination address of each SDQ in node, all broadcasting services are arranged on separately among the SDQ, and the service conditions in each this node of nodes records in each formation is not empty if the business in the formation is arranged, and the business that has specific node on this node at present then is described.This information will be used for judging whether and need switching this business in the RSSP agreement.The present invention uses the RSSP agreement and realizes that node on the RPR looped network adjusts the sending direction of himself service path automatically, and following mask body is introduced the RSSP agreement.
1, RSSP agreement place network layer.
Figure 4 shows that the network layer schematic diagram of RSSP agreement work.RSSP is operated in the mac layer of RPR protocol stack.Its receives the information from MAC data path layer, and as the RSSP message packet of congestion information and other node etc., and after handling according to the information that receives, professional sending direction is switched in the control MAC data path layer.
Figure 5 shows that the collaborative work schematic diagram of RSSP agreement and loop service channel key-course.The RSSP agreement is accepted user's indication, as the user to its configuration parameter or send mandatory order etc., and the RPR protocol message that reports by MAC data path layer, simultaneously, obtain the SDQ state by inquiry primitive, send the instruction that the ring traffic direction has been gone up in control to MAC data path layer.
2, RSSP agreement term and explanation.
Preferred (SSP, Service Switch Preferred) condition of business reorganization: be a kind of service condition in the RSSP agreement, when a node detection arrives the traffic congestion of certain direction, then claim this node on this loop direction, the SSP condition to occur.
The SSP source node: the node in (SAC, Service Adjust Cycle) cycle in initialization business reorganization cycle is called the SSP source node, and the information of SSP source node will be filled in the control message of RSSP agreement.
Influenced by the SSP source point: when some professional transmit path of certain node passes through the congested loop of SSP source node, then to claim this joint business influenced by the SSP source node.
The announcement direction: the SSP source node is the announcement direction to the direction of the opposite direction transmission protocol massages of the place loop that the SSP condition takes place.
The SSP source side to: the loop direction that the SSP condition takes place be the SSP source side to, the opposite direction of promptly announcing direction be the SSP source side to.
Return message: after the protocol massages that the SSP source node sends is got back to this node along loop after one week, then claim the recurrence message of this message for certain message.
Business reorganization failure indication: at an intranodal, the state machine of RSSP agreement sends to the message of message dispatcher, and this message does not have to make the two ends of this node all not have the SSP condition to take place after showing last professional switching of this state machine.
Business reorganization is successfully indicated: at an intranodal, the state machine of RSSP agreement sends to the message of message dispatcher, and this message makes the two ends of this node all not have the SSP condition to take place after showing last professional switching of this state machine.
SAC (the Service Adjust Cycle) cycle: detect the SSP condition and start protocol state machine by SSP source node each time and begin, to PWI (Protocol Wait Interval) timer expiry, perhaps (success or failure) switched and finished to all node traffics of loop, perhaps switches the process that abnormal end (FSI) experienced and be called a SAC cycle.The protocol massages in each SAC cycle all has own unique sequence number.Unless the state machine of intranodal is in free time (IDLE) state, otherwise state machine is not accepted the message in other cycle.
3, RSSP protocol message message.
The RSSP agreement is a distributed agreement.On each node, all move an example of agreement.The agreement of different nodes is by protocol message message switching information and co-ordination.RSSP protocol message message occurs with the form of RPR agreement control frame.Figure 6 shows that RSSP protocol message message format.The message packet of RSSP agreement comprises RPR head, payload and check field (FCS) three parts.In the RPR head, comprise life span (TTL), basic control information (baseControl), destination address (DA), source address (sa), initial life span (ttlBase), expansion control information (extendedControl) and an error checking and information (hen); The sequence number (RSSP Seq) and the RSSP control message type of message (RSSP Type) that comprise control information type (controlType), control information version (controlVersion), RSSP message in payload part.Wherein, having only RSSP Seq and RSSP Type is new field, and all the other fields are all identical with the RSSP consensus standard.Each SAC cycle of above-mentioned RSSP Seq is unique, and the particular type of RSSP control message message is as shown in table 1:
RSSP TYPE Type of message
0X01 SSR message
0X02 SSC message
0X03 SSS message
0X04 FSC message
Other Keep
Table 1
Business reorganization request (SSR, Service Steering Request) message: this message sends to adjacent downstream node by the upstream node on the announcement direction.The business reorganization failure of this message explanation upstream node needs downstream node to carry out business reorganization.This message is clean culture, and each node is when transmitting it, and da field is the address of next node, and TTL is 1.Sa field is SSP address of node, source.
Business reorganization failure (SSC, Service Steering Cancel) message: this message is by SSP source node other node broadcasts in loop, it illustrates business reorganization failure in this SAC cycle, and the business recovery of notifying each node to do self is to switch preceding sending direction.
Business reorganization success (SSS, Service Steering Succeed) message: this message is by SSP source node other node broadcasts in loop, and it illustrates business reorganization success in this SAC cycle, notifies each node to keep current professional sending direction.
Force to adjust failure (FSC, Forced Steering Cancel) message: this message is triggered by the user.The RSSP agreement allows user's Compulsory Removal current by the adjusted professional sending direction of RSSP agreement, reverts to the professional sending direction by the default routing algorithm decision of RPR.This message is by the node that takes orders other node broadcasts in loop.
4, the timer of RSSP protocol definition.
Protocol stateful inspection (PSC, Protocol State Checking) timer: this timer is the basic time cycle of RSSP agreement, and each state machine of agreement just detects state one time every the time interval of PSC timer length.
Stable (FSI, the Fairness Stable Interval) timer at interval of fair algorithm: this timer is the stabilization time of fair algorithm.After having adjusted professional sending direction, need wait for a period of time, so that the fair algorithm of RPR is stable.In this RSSP agreement, it is set to the needed stabilization time of fair algorithm that concrete RPR agreement realizes loop, and the concrete time is disposed by the user, and each state machine of each node is identical.
Agreement is waited for (PWI, Protocol Wait Interval) timer at interval: each SSP source node is being initiated a SAC week after date, initialization PWI timer.If the PWI timer time at interval in business still do not adjust successfully, perhaps the SSP source node can not receive receiveing the response of other node, thinks that then this business reorganization fails.What of the number of current loop node and current business this timer determined according to by agreement.In the present embodiment, the time of this timer is: interstitial content multiply by the FSI time, multiply by the number that our station is subjected to the business of SSP condition influence again.
SSS (Service Steering Succeed) timer: after the SSP source node is checked through the disappearance of SSP condition, start this timer.If loop has kept the SSS time interval not occur the SSP condition again, then Congestion Avoidance success.The time of this timer is the multiple of FSI.Recommendation is 2 FSI.
Specific implementation step of the present invention is as follows:
Step 1, when traffic congestion occurring on the looped network, produce congested node and will obtain this information automatically, and judge at first whether another loop of RPR looped network allows to switch, if, then enter a business reorganization cycle, start the PWI timer of this intranodal, and, send SSR message to adjacent downstream node along the announcement direction, otherwise the professional sending direction of keeping this node is constant, does not promptly carry out the adjustment of service path;
Step 2, the node that receives SSR message judges whether the business that self sends is subjected to the SSP condition effect, in promptly judging self and producing that congested node is corresponding and line up whether there be professional the existence, and should business in this business reorganization cycle, also not have processed, if be not affected, then execution in step 3, otherwise after switching the sending direction of this node business, start the FSI timer of this intranodal, behind the FSI timer expiry, judge again whether this node still is subjected to the SSP condition effect, and whether received SSR message is to return message
If this node still is subjected to the SSP condition effect after switching professional sending direction, and the SSR message of being received is to return message, then the business recovery with this node is to switch preceding state, and after other node broadcasts SSC message in looped network, execution in step 4;
If this node still is subjected to the SSP condition effect after switching professional sending direction, but the SSR message of being received is not to return message, then the business recovery with this node is to switch preceding state, and after adjacent downstream node is transmitted SSR message, repeated execution of steps 2;
If this node no longer is subjected to the SSP condition effect after switching professional sending direction, and the SSR message of being received is to return message, then this node keeps the sending direction of current business constant, and start the SSS timer of this intranodal, after this SSS timer expiry, if congestion condition does not appear in this node again, the then node broadcasts of other in looped network SSS message, and execution in step 5; Before this SSS timer expiry, if congestion condition has appearred again in this intranodal, think that then congestion condition is not avoided, behind the wait PWI timer expiry, other node broadcasts SSC message in looped network, and execution in step 4;
If this node no longer is subjected to the SSP condition effect after switching professional sending direction, but the SSR message of being received is not to return message, then this node keeps the sending direction of current business constant, and after adjacent downstream node is transmitted SSR message, repeated execution of steps 2;
Step 3, the node that receives SSR message be along the announcement direction, after adjacent downstream node sends SSR message, and execution in step 2;
Step 4 receives the node of SSC message, and the professional sending direction of this node is reverted to sending direction before switching, finishes this business reorganization cycle, and all nodes enter init state;
Step 5 receives the node of SSS message, keeps the professional current sending direction of this node, finishes this business reorganization cycle, and all nodes enter init state.
Each carries out the selection of the node of switching services according to the user, can use simple mode or the sending direction of this node business is switched or recovered to switch to complex patterns.So-called simple mode is meant, the node that carries out switching services with self all professional sending directions all switch; So-called complex patterns is meant, the node that carries out switching services to self all professional sending directions at first judge, will be through the SSP source node, and switch in the path sending direction of long business, if after switching, still there is congestion condition in this node on the congested direction of generation, then again will be through the SSP source node, and the sending direction of path vice-minister's business is switched, by that analogy, till having switched all business, or work as node till congestion condition disappears on the congested direction of generation through the SSP source node.
In the present invention, the message dispatcher of each node by the RSSP protocol definition (MS, MessageScheduler) and finite-state automata (FSM) finish the automatic adjustment of service path.Wherein, finite-state automata, after be called state machine, be divided into active state machine (ASM) and passive state machine (PSM) two classes.Respectively it is introduced below.
All there are two message dispatcher on each node, are respectively applied for the RSSP protocol message of reception, and, transmit to the finite-state automata that this locality is suitable according to the difference of type of message from two loops.Its concrete operations are as follows:
If what message dispatcher received is SSR message, then judge whether to be subjected to the passive state machine of the SSP condition influence that this SSR message comprises, promptly judge the local current business that whether has through the SSP source node, and at this SAC in the cycle, the pairing passive state machine of this business was not also handled this SSR message, if have, then the sending direction of business is switched, and waited for that the passive state machine returns indication by previously selected simple mode of user or complex patterns.
If the passive state machine returns business reorganization and successfully indicates, then this scheduler restPoses;
If the passive state machine returns business reorganization failure indication, and this SSR message be not the counterparty to the recurrence message, then this SSR message is transmitted to adjacent downstream node along the announcement direction of this SSR message, and this scheduler is recovered initial state;
If the passive state machine returns business reorganization failure indication, and this SSR message be the counterparty to the recurrence message, then local active state machine all node broadcasts SSC message in looped network of order, and initialization scheduler finish the adjustment operation in this SAC cycle.
If message dispatcher receives is other message except that SSR message, then transmitted by the state machine of this messages affect to all this message.
All there are two active state machines on each node, correspond respectively to two different sending directions; The number of passive state machine is relevant with the interstitial content on the loop on each node.If N node arranged on the loop, N-1 passive state machine just arranged, the respectively corresponding destination node of each passive state machine.
Figure 7 shows that the schematic diagram of state machine on the individual node.Active state machine (ASM) is a software entity of initiating and finish the SAC cycle in the RSSP agreement.For the node of an operate as normal, there are two professional sending directions, therefore there is the example of two active state machines to handle two business on the sending direction respectively.For example, when certain node find one of this node receive occur on the loop congested after, corresponding active state machine will be initiated a SAC cycle, send protocol message to other node, select professional sending direction with the state machine that drives other node according to the regulation of RSSP agreement; Simultaneously, the active state machine also needs to receive the notification message based on the RPR agreement from other node, thereby whether decision has successfully carried out the adjustment of loop traffic direction.
Comprise following information on the active state machine: the node number of this state machine correspondence, this state machine corresponding service receive loop, SSS timer state, PWI timer state, the SSP condition whether occurs on this state machines, the preceding state of this SAC periodic sequence number and this active state machine.
Passive state machine (PSM) is the software entity of specific implementation traffic direction adjustment (ServiceSteering) in the RSSP agreement.The passive state machine is by the message dispatcher of this node or by the active state machine driving work of other node.Each passive state machine is managed the service path of a loop node.On the RPR loop, any one node all may send professional to other node, therefore for a loop that N node arranged, N-1 passive state machine is arranged all on each node.The passive state machine is switched driving message receiving, during as SSR message, the SSP condition influence that comprised by this SSR message according to own corresponding professional transmission loop at present determines whether adjusting professional sending direction.
Comprise following information on the passive state machine: the sending direction of current business purpose node ID or MAC Address, the current business determined by the RPR of lower floor topology agreement, the service path of the current SSP of existence condition referred to professional sending direction that sending direction (not considering bidirectional broadcast), a last SAC cycle determine, whether have the SSP condition on the direction before switching, whether exist on the direction after switching the current state of SSP condition, this SAC periodic sequence number, FSI timer state and this passive state machine.
Under the driving of service condition, the state of state machine moves on each node in RSSP protocol message and loop, carries out different actions simultaneously, thereby finishes the dynamic adjustment process of traffic direction.The state that the active state machine may be in has: idle (IDLE) state, handoff failure are waited for (SSR_W) state and handover success wait (SSR_S) state; The state that the passive state machine may be in has: IDLE state, switching (SWITCH) state and SSR_S state.SSR_W is after this node business reorganization failure, a wait state after other node sends SSR message.If disappear in SSR_W state time point discovery congestion condition, then further enter the SSR_S state, this SSR_S state is the shake for fear of the loop business.
Table 2 is depicted as the state transition table of state machine.
OK State Condition Action NextState
1 IDLE 1. this state machine is that passive state machine 2. receives 3. current service paths of state machine of upstream SSR message and is subjected on SSP condition influence 4. another loops of SSR announcement the SSP condition influence is arranged also 1. return business reorganization failure indication to the message dispatcher of correspondence SSR_W
2 IDLE 1. this state machine is that passive state machine 2. receives upstream SSR message 3. service paths and is subjected to announcement to influence no SSP condition on 4. another loops 1. the corresponding traffic direction of this state machine is switched 2. record current business switch status, the current transmission loop 3. initialization FSI timers of update mode machine SWITCH
3 IDLE 1. the current state machine is that the SSP condition appears in the operation loop that local active state machine 2. detects this state machine correspondence 1. be directed downwards the trip node along announcement and send SSR message 2. initialization PWI timers SSR_W
4 IDLE 1. receive the difference that FSC message 2. current transmission loops calculate with the RPR agreement 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop of update mode machine IDLE
5 IDLE Other All timers of initialization (FSI, PWI, SSS) IDLE
6 SWITCH 1. local passive state machine enters the SWITCH attitude 1. do not have IDLE
7 SWITCH 1. there is the SSP condition in the current transmission loop of overtime 3. state machines of passive state machine 2.FSI 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop 3. of update mode machine returns under business reorganization failure indication 4. specific (special) requirements the user to the message dispatcher of correspondence, can initialization PWI timer SSR_W
8 SWITCH 1. the current transmission loop of overtime 3. state machines of passive state machine 2.FSI does not exist SSP condition 4. operation loops with the SSP condition not occurring on the different local state machine of the current transmission loop of this state machine Do not have IDLE
9 SWITCH 1. the current transmission loop of overtime 3. state machines of passive state machine 2.FSI does not exist SSP condition 4. operation loops with the SSP condition occurring on the different local state machine of the current transmission loop of this state machine 1. return under business reorganization failure indication 2. the specific (special) requirements initialization PWI timer the user to the message dispatcher of correspondence SSR_W
10 SWITCH 1. passive state machine 2. receives the SSS message that the SSR source node is sent 1. do not have IDLE
11 SWITCH 1. passive state machine 2. receives the SSC message that the SSR source node is sent 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop of update mode machine IDLE
12 SWITCH 1. receive the difference that FSC message 2. current transmission loops calculate with the RPR agreement 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop of update mode machine IDLE
13 SWITCH Other 1. do not have SWITCH
14 SSR_W 1. local active state machine 2. is that the SSR message (illustrate message on announcement direction around ring one week) in SSR source 4.PWI not overtime in the SSP source side to receiving with oneself to there being SSP condition 3. in the SSP source side 1. return business reorganization failure indication to the message dispatcher of correspondence SSR_W
15 SSR_W 1. local active state machine 2. is not overtime at the SSP condition disappearance 3.PWI that the SSP source side makes progress 1. start the SSS timer SSR_S
16 SSR_W 1. local active state machine 2.PWI overtime 3. in the SSP source side to still being the SSP condition 1. broadcast SSC message IDLE
17 SSR_W 1. the current sender of overtime 3. state machines of passive state machine 2.PWI is to the sending direction difference of a same SSP in the cycle 1. with this state machine corresponding service directional switching record current business switch status, the current transmission loop of update mode machine IDLE
18 SSR_W 1. the current sender of overtime 3. state machines of passive state machine 2.PWI is identical to the sending direction of a same SSP in the cycle 1. do not have IDLE
19 SSR_W 1. passive state machine 2.PWI not overtime 3. receives the SSS message of SSR source node 1. do not have IDLE
20 SSR_W 1. passive state machine 2.PWI not overtime 3. receives the current sender of 4. state machines of SSC message of SSR source node to the sending direction difference of a same SSP in the cycle 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop of update mode machine IDLE
21 SSR_W 1. to receive the current sender of 4. state machines of SSC message of SSR source node identical to the sending direction of a same SSP in the cycle for passive state machine 2.PWI not overtime 3. 1. do not have IDLE
22 SSR_W 1. receive the difference that FSC message 2. current transmission loops calculate with the RPR agreement 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop of update mode machine IDLE
23 SSR_W Other 1. do not have SSR_W
24 SSR_S 1. the not overtime 3.SSS not overtime 4. of local active state machine 2.PWI the SSP source side to receive with oneself be the SSR source the SSR message (illustrate message the announcement direction on around the ring one week) 1. remove the SSS timer SSR_W
25 SSR S 1. the not overtime 3.SSS not overtime 4. of local active state machine 2.PWI in the SSP source side to the SSP condition occurring 1. remove the SSS timer SSR_W
26 SSR_S 1. local active state machine 2.SSS overtime 3. disappears in the SSP source side SSP condition that makes progress 1. broadcast SSS message IDLE
27 SSR_S 1. passive state machine (the passive state machine generally can not enter the SSR_S state, enters this state, and then explanation has made mistakes, and directly switch to it SSR_W state this moment) 1. do not have SSR_W
28 SSR_S 1. receive the difference that FSC message 2. current transmission loops calculate with the RPR agreement 1. with this state machine corresponding service directional switching 2. record current business switch status, the current transmission loop of update mode machine IDLE
29 SSR_S Other 1. do not have SSR_S
Table 2
The example that switches with the application complex patterns specifies below.
Figure 8 shows that the schematic diagram of using one embodiment of the invention.There are 5 nodes on this loop, are respectively node 1, on node 1, have 4 business, have 1 business on the node 3 to node 5.Wherein, each the bar business on the node 1 is 100M, and node 3 is 200M to the business of node 4, and loop bandwidth is 500M, and all in the outer shroud direction, promptly illustrates clockwise direction.At this moment, do not have congestedly on the loop, the state of node 1 each state machine is as shown in table 3, and the state of node 3 each state machines is as shown in table 4:
Node 1 state machine
PSM2(DA=2) PSM3(DA=3) PSM4(DA=4) PSM5(DA=5) The active state machine
State IDLE State IDLE State IDLE State IDLE State IDLE
Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL
RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud The SSP condition NO
RSSP agreement sending direction Outer shroud RSSP agreement sending direction Outer shroud RSSP agreement sending direction Outer shroud RSSP agreement sending direction Outer shroud
Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud
Table 3
Node 3 state machines
PSM4(DA=4) ASM0 (active state machine)
State IDLE State IDLE
Periodic sequence number NULL Periodic sequence number NULL
RPR agreement sending direction Outer shroud The SSP condition NO
RSSP agreement sending direction Outer shroud
Last one-period sending direction Outer shroud
Table 4
Suppose to increase to 450M in node 3 to the business between the node 4 this moment, so the RPR agreement on the looped network has detected congested on the outer shroud of node 3.At this moment, the active state machine initialization SAC cycle of node 3 outer shroud directions is SAC1, and encircles direction along the announcement direction promptly, sends the SSR message to node 2, and the active state machine of node 3 enters the SSR_W state simultaneously.Referring to Fig. 9, node 3 is in the SSS message that node 2 is sent out, and destination address (DA) is a node 2, and source address (sa) is a node 3, periodic sequence number (SQ) be designated SAC1.At this moment, the state of node 3 each state machines is as shown in table 5:
Node 3 state machines
PSM4(DA=4) ASM0 (active state machine)
State IDLE State SSR_W
Periodic sequence number NULL Periodic sequence number SAC1
RPR agreement sending direction Outer shroud The SSP condition YES
RSSP agreement sending direction Outer shroud The SSP source side to Outer shroud
Last one-period sending direction Outer shroud PWI is overtime NO
Table 5
Node 2 is after the SSR message of receiving from node 1, and the SDQ formation of finding oneself all is empty, that is to say, node 2 itself is not subjected to the business of the SSP condition influence that comprised in the SSR message, therefore, node 2 keeps current state constant, and this SSR message is transmitted to next node.Each state machine of node 3 remains unchanged in this process.This moment looped network state as shown in figure 10, node 2 is in the SSS message that node 1 is sent out, destination address (DA) is a node 1, source address (sa) is a node 3, SQ is designated SAC1.
After SSR message arrives node 1, the message dispatcher of node 1 is according to the service conditions information of self current all SDQ formation, judging self has two business to be subjected to the SSP condition effect that comprises in this SSR message, the business that is node 1 to the business of node 4 and node 1 to node 5 is affected, so, message dispatcher is at first selected the longest professional pairing passive state machine in path from affected business, and this SSR message passed to this passive state machine, promptly handle by PSM5.After PSM5 receives this message,, whether allow to switch as another loop according to the condition of switching, judge self can switch after, the sending direction that change is own corresponding professional, and updates self information, initialization FSI timer then, this moment the looped network business state as shown in figure 11.The state of node 1 each state machine is as shown in table 6:
Node 1 state machine
PSM2(DA=2) PSM3(DA=3) PSM4(DA=4) PSM3(DA=5) The active state machine
State IDLE State IDLE State IDLE State SWITCH State IDLE
Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number SAC1 Periodic sequence number NULL
RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud The SSP condition NO
RSSP agreement sending direction Outer shroud RSSP agreement sending direction Outer shroud RSSP agreement sending direction Outer shroud RSSP agreement sending direction Interior ring
Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud
The SSP source side is to the SSP bar NO
Part
Current direction SSP condition NO
FSI is overtime NO
Switch sign YES
Table 6
After FSI was overtime, there was not the SSP condition in the professional sending direction after PSM5 finds self to switch, but still there is the SSP condition in the direction before switching, so, return business reorganization failure indication and give MS.After MS receives the feedback of PSM5, from affected business, select path vice-minister's professional pairing passive state machine, and this SSR message is passed to this passive state machine, promptly handle by PSM4.After PSM4 receives this message, according to the condition of switching, judge self can switch after, so the own corresponding professional sending direction of change, and updates self information, initialization FSI timer then, this moment the looped network business state as shown in figure 12.The state of node 1 each state machine is as shown in table 7:
Node 1 state machine
PSM2(DA=2) PSM3(DA=3) PSM4(DA=4) PSM5(DA=5) The active state machine
State IDLE State IDLE State SWITCH State SWITCH State IDLE
Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number SAC1 Periodic sequence number SAC1 Periodic sequence number NULL
RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud The SSP condition NO
RSSP agreement sending direction Outer shroud RSSP agreement sending direction Outer shroud RSSP agreement sending direction Interior ring RSSP agreement sending direction Interior ring
Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud
The SSP source side is to the SSP condition NO The SSP source side is to the SSP condition NO
Current direction SSP condition NO Current direction SSP affair part NO
FSI is overtime NO FSI is overtime YES
Switch sign YES Switch sign YES
Table 7
Wait for that at the PSM4 of node 1 node 3 detects congestion cleared in the overtime process of FSI, and this moment, PWI did not have also overtimely, so node 3 enters the SSR_S state, and started the SSS timer.The sending direction of each bar business all continues to have in the looped network at this moment.The state of node 3 each state machines is as shown in table 8:
Node 3 state machines
PSM4(DA=4) ASM0 (active state machine)
State IDLE State SSR_S
Periodic sequence number NULL Periodic sequence number SAC1
RPR agreement sending direction Outer shroud The SSP condition NO
RSSP agreement sending direction Outer shroud PWI is overtime NO
Last one-period sending direction Outer shroud SSS is overtime NO
Table 8
After SSS was overtime, the SSP condition did not appear in node 3 again, so all state machines in the node 3 all revert to the IDLE state, and all the node broadcasts SSS message in looped network.Referring to Figure 13, the source address that node 3 is sent in the broadcast is a node 3, and SQ is designated SAC1.This moment, the state of node 3 each state machines was as shown in table 9:
Node 3 state machines
PSM4(DA=4) ASM0 (active state machine)
State IDLE State IDLE
Periodic sequence number NULL Periodic sequence number NULL
RPR agreement sending direction Outer shroud The SSP condition NO
RSSP agreement sending direction Outer shroud
Last one-period sending direction Outer shroud
Table 9
After node 1 was being received SSS message, MS was transmitted to PSM4 and PSM5 with SSS, and the state of PSM4 and PSM5 self is set to IDLE.The state of node 1 each state machine is as shown in table 10:
Node 1 passive state machine
PSM2(DA=2) PSM3(DA=3) PSM4(DA=4) PSM5(DA=5) The active state machine
State IDLE State IDLE State IDLE State IDLE State IDLE
Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL Periodic sequence number NULL
RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud RPR agreement sending direction Outer shroud The SSP condition NO
RSSP Outer shroud RSSP Outer shroud RSSP Interior ring RSSP Interior ring
The agreement sending direction The agreement sending direction The agreement sending direction The agreement sending direction
Last one-period sending direction Outer shroud Last one-period sending direction Outer shroud Last one-period sending direction Interior ring Last one-period sending direction Interior ring
The SSP source node NO The SSP source node NO
Table 10
So far, the business on the loop has reached a new stable state, and the business in the node 1,3 can both be with the wide transmission of filled band.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being made, is equal to and replaces and improve, and all should be included within protection scope of the present invention.

Claims (13)

1. one kind in the online method of adjusting service path automatically of Resilient Packet Ring, it is characterized in that this method may further comprise the steps:
A, when traffic congestion occurring on the looped network, produce congested node and will obtain congestion information automatically, and start a business reorganization cycle, simultaneously along the opposite direction that produces congestion condition place loop, send business reorganization request SSR message to adjacent downstream node;
B, the node that receives SSR message judge whether the business that self sends is subjected to the influence of congestion condition, if be not affected, execution in step c then, otherwise after switching the sending direction of this node business, judge whether this node still is subjected to the influence of congestion condition, and whether received SSR message is to return message, and described recurrence message is meant by node and sends, and gets back to the SSR message of this node along looped network after one week again;
If still be subjected to the influence of congestion condition behind the sending direction of this node business of switching, and the SSR message of being received is to return message, then the business recovery with this node is to switch preceding state, and after other node broadcasts business reorganization failure SSC message in looped network, execution in step d;
If still be subjected to the influence of congestion condition behind the sending direction of this node business of switching, but the SSR message of being received is not to return message, then the business recovery with this node is to switch preceding state, and after adjacent downstream node is transmitted SSR message, repeated execution of steps b;
If no longer be subjected to the influence of congestion condition behind the sending direction of this node business of switching, and the SSR message of being received is to return message, then this node keeps the sending direction of current business constant, and after other node broadcasts business reorganization success SSS message in looped network, execution in step e;
If no longer be subjected to the influence of congestion condition behind the sending direction of this node business of switching, but the SSR message of being received is not to return message, then this node keeps the sending direction of current business constant, and after adjacent downstream node is transmitted SSR message, repeated execution of steps b;
C, receive SSR message node along the opposite direction that produces congestion condition place loop, after adjacent downstream node sends SSR message, execution in step b;
D, receive the node of SSC message, the professional sending direction of this node is reverted to sending direction before switching, finish this business reorganization cycle, all nodes enter init state;
E, receive the node of SSS message, keep the professional current sending direction of this node, finish this business reorganization cycle, all nodes enter init state.
2. method according to claim 1, it is characterized in that this method further comprises: when looped network was in stable state, each node was lined up respectively for the business that this node is sent to other node in advance, and each lines up corresponding destination node, and each is lined up to monitor.
3. method according to claim 2, it is characterized in that, step b is described to judge that the method whether business that self sends is subjected to the congestion condition influence is: judge with produce that congested node is corresponding and line up in whether professional the existence is arranged, and should business in this business reorganization cycle, also not have processed.
4. method according to claim 1, it is characterized in that, described step a further comprises: produce congested node, whether another loop of at first judging RPR network RPR looped network allows to switch, if, then enter a business reorganization cycle again, and along the opposite direction that produces congestion condition place loop, send SSR message to adjacent downstream node, otherwise it is constant to keep the professional sending direction of this node.
5. method according to claim 1, it is characterized in that, the congested node of the described generation of step a is after adjacent downstream node sends SSR message, further comprise: the agreement that starts this intranodal is waited for PWI timer at interval, if this node the PWI timer then after, the business in the loop is not still adjusted successfully, or does not receive receiveing the response of other node, then the business recovery of all nodes is to switch preceding sending direction in the loop, and finishes this adjustment cycle.
6. method according to claim 5, it is characterized in that, the node of the described SSR of the receiving message of step b judges that the business that self sends is subjected to the influence of congestion condition, and after switching the sending direction of this node business, further comprise: what start this intranodal switches abnormal end FSI timer, behind the FSI timer expiry, carry out subsequent step again.
7. method according to claim 6 is characterized in that, the timing length of described FSI timer realizes the stabilization time of loop fair algorithm for the RPR agreement.
8. method according to claim 6 is characterized in that, the timing length of described PWI timer is according to professional how much determining in the number of node in the current loop and the current loop.
9. method according to claim 8 is characterized in that, described definite method is: the timing length of PWI timer equals interstitial content and multiply by the FSI time, multiply by the professional number that this node is influenced by congestion condition again.
10. method according to claim 6, it is characterized in that, before congested other node broadcasts SSS message of node in looped network of the described generation of step b, further comprise: the SSS timer that starts this intranodal, after this SSS timer expiry, if congestion condition does not appear in this node again, then carry out subsequent step again; Before this SSS timer expiry, if congestion condition appears again in this intranodal, wait for that then the PWI timer then after, other node broadcasts SSC message in looped network, and execution in step d.
11. method according to claim 10 is characterized in that, the timing length of described SSS timer is the multiple that the FSI timer is fixed time, and is generally the time span of 2 FSI.
12. method according to claim 1, it is characterized in that, this method further comprises: when node received the check field FCS message that is triggered by the user, all nodes all reverted to the professional sending direction that is determined by the default routing algorithm of RPR with the professional sending direction of self.
13. method according to claim 1 is characterized in that, the method for the sending direction of described this node of node switches business is:
The node that carries out switching services with self all professional sending directions all switch, or
The node that carries out switching services to self all professional sending directions at first judge, will be through the preferred SSP source node of business reorganization, and switch in the path sending direction of long business, if after switching, still there is congestion condition in this node on the congested direction of generation, then again will be through the SSP source node, and the sending direction of path vice-minister's business is switched, by that analogy, till having switched all business, or work as node till congestion condition disappears on the congested direction of generation through the SSP source node.
CNB2004100009657A 2004-01-17 2004-01-17 Method for automatic adjusting business path for flexible packet ring network Expired - Fee Related CN100370780C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2004100009657A CN100370780C (en) 2004-01-17 2004-01-17 Method for automatic adjusting business path for flexible packet ring network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2004100009657A CN100370780C (en) 2004-01-17 2004-01-17 Method for automatic adjusting business path for flexible packet ring network

Publications (2)

Publication Number Publication Date
CN1642128A CN1642128A (en) 2005-07-20
CN100370780C true CN100370780C (en) 2008-02-20

Family

ID=34866957

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2004100009657A Expired - Fee Related CN100370780C (en) 2004-01-17 2004-01-17 Method for automatic adjusting business path for flexible packet ring network

Country Status (1)

Country Link
CN (1) CN100370780C (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100493019C (en) * 2005-10-29 2009-05-27 华为技术有限公司 Method and device for implementing ring configuration in ring network
CN101304380B (en) * 2008-06-30 2011-04-13 杭州华三通信技术有限公司 Traffic transmission method of elastic packet loop and elastic packet loop node
CN101582857B (en) * 2009-06-30 2011-06-22 福建星网锐捷网络有限公司 Data upper ring forwarding method based on elastic grouping ring and device
CN102726006B (en) * 2011-11-23 2015-03-18 华为技术有限公司 Ethernet ring protection switching method, node and system
CN102754390B (en) 2012-04-28 2015-06-17 华为技术有限公司 Method, network device and system for ethernet ring protection switching

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1189269A (en) * 1995-06-26 1998-07-29 艾利森电话股份有限公司 Self-healing network
JP2000196524A (en) * 1998-12-23 2000-07-14 Alcatel Method and device for optimizing availability of low priority channel in transmitting optical fiber transoceanic ms-sp ring in error exsting state
CN1314749A (en) * 2000-03-17 2001-09-26 朗迅科技公司 Method and device for restoring information signal of sending path in net type net
CN1099789C (en) * 1996-11-12 2003-01-22 株式会社东芝 Loop-type network system and controlling of path thereof
US6553034B2 (en) * 1999-03-15 2003-04-22 Tellabs Operations, Inc. Virtual path ring protection method and apparatus

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1189269A (en) * 1995-06-26 1998-07-29 艾利森电话股份有限公司 Self-healing network
CN1099789C (en) * 1996-11-12 2003-01-22 株式会社东芝 Loop-type network system and controlling of path thereof
JP2000196524A (en) * 1998-12-23 2000-07-14 Alcatel Method and device for optimizing availability of low priority channel in transmitting optical fiber transoceanic ms-sp ring in error exsting state
US6553034B2 (en) * 1999-03-15 2003-04-22 Tellabs Operations, Inc. Virtual path ring protection method and apparatus
CN1314749A (en) * 2000-03-17 2001-09-26 朗迅科技公司 Method and device for restoring information signal of sending path in net type net

Also Published As

Publication number Publication date
CN1642128A (en) 2005-07-20

Similar Documents

Publication Publication Date Title
US20190356580A1 (en) Resilient data communications with physical layer link aggregation, extended failure detection and load balancing
CN100525234C (en) Systems and methods for alleviating client over-subscription in ring networks
US6952396B1 (en) Enhanced dual counter rotating ring network control system
AU607571B2 (en) Distributed load sharing
JP4034782B2 (en) Ring connection device and data transfer control method
US7746768B2 (en) System and method for supporting SDH/SONET APS on ethernet
CN100571165C (en) A kind of method of finding automatically based on topological structure of multi-service transmission looped network
JP4167072B2 (en) Selective protection against ring topology
JP5633469B2 (en) NETWORK, FAILURE RECOVERY METHOD, AND NODE DEVICE
US20060182033A1 (en) Fast multicast path switching
CN101060485B (en) Topology changed messages processing method and processing device
CN109547339B (en) Method and system for preventing network from forming loop in wireless router networking
JP2006129071A (en) Node device
CN104301226A (en) Designated routing bridge (DRB) election method, equipment and system
CN106685783A (en) Ring network protection method and system
CN100370780C (en) Method for automatic adjusting business path for flexible packet ring network
CN102857316A (en) Method and system for realizing source looped network protection
CN101262399B (en) A cross-loop RPR two point failure processing method and system
CN100496023C (en) Method for transmitting link-state information
US20070223373A1 (en) Communication control apparatus and communication control method
US7391720B1 (en) Local span mesh restoration (LSMR)
CN100359880C (en) Method for implementing transmission and protection for service in MPLS ring net
CN100464528C (en) Method and system for preventing circuit loop after failure recovery
JP2009044562A (en) Multi-ring rpr node device
Cisco Configuring IP Enhanced IGRP

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
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20080220

Termination date: 20120117