CN104468302A - Token processing method and device and message handling cluster - Google Patents
Token processing method and device and message handling cluster Download PDFInfo
- Publication number
- CN104468302A CN104468302A CN201410550612.8A CN201410550612A CN104468302A CN 104468302 A CN104468302 A CN 104468302A CN 201410550612 A CN201410550612 A CN 201410550612A CN 104468302 A CN104468302 A CN 104468302A
- Authority
- CN
- China
- Prior art keywords
- token
- loop
- node
- message processing
- processing cluster
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 52
- 230000008569 process Effects 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims description 70
- 238000012544 monitoring process Methods 0.000 claims description 52
- 238000004804 winding Methods 0.000 claims description 21
- 238000002834 transmittance Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000006378 damage Effects 0.000 description 5
- 238000007726 management method Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 235000003642 hunger Nutrition 0.000 description 3
- 238000007630 basic procedure Methods 0.000 description 2
- 230000003139 buffering effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007717 exclusion Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000003892 spreading Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Landscapes
- Small-Scale Networks (AREA)
Abstract
The invention relates to the technical field of the network and provides a token processing method and device and a message handling cluster. The method comprises the steps that a token loop is built on an application layer of the message handling cluster; when nodes are added into or exit from the token loop, a token is generated or destroyed by the adoption of a simple majority voting algorithm; in the token conveying process, the timestamps when the nodes on the token loop transmit the token last time are monitored, if it is monitored that the difference between the timestamps when the nodes on the token loop transmit the token last time and the current time exceeds the set loop time, the route discovery process is sponsored, and the token loop is rebuilt. It is avoided that two or more token loops exist in the message handling cluster. In addition, the token loop time of the nodes is monitored, the problem that in the prior art, loss of tokens in the token loop can not be monitored is solved, and meanwhile the reliability of a distributed cluster system is greatly improved.
Description
Technical field
The invention belongs to networking technology area, particularly relate to a kind of processing method of token, device and Message Processing cluster.
Background technology
Along with the continuous expansion of network size, distributed system is widely used, and also brings the conforming problem of Distributed Coordination simultaneously.Consistency problem main manifestations is data consistency, logical consistency and temporal consistency, when solution consistency problem mainly guarantees that distributed system is accessed at the same time and controls a resource, can be harmonious, ensure the correctness (deadlock freedom and without hunger phenomenon) of access.Mutual exclusion is the key that distributed system solves consistency problem.The mutually exclusive algorithm that usual distributed system uses has: centralized algorithm, distributed algorithm and token ring algorithm.
Centralized algorithm selects a process as telegon, for coordinating entering of critical zone.
Feature: telegon only allows a process to enter critical zone at one time, therefore can ensure mutual exclusion.Obtain because request message is decision queue, there will not be " dying of hunger " phenomenon.But single telegon is its bottleneck.
The thought of distributed algorithm is: first set up a message to the process of wanting to enter critical zone, and this message comprises critical zone name, process name and timestamp to be entered.After a process receives message, have following operation: if do not want to enter critical zone not in critical zone, reply a message; If in critical zone, do not reply message, request message is put into queue, if not in critical zone, but want to enter critical zone, then compare the timestamp of the process of wanting to enter critical zone, the process that timestamp is little enters critical zone.
Feature: algorithm is complicated, easily occur " dying of hunger " phenomenon, and system is unhealthy and strong; But it indicates the feasibility of algorithm theoretically, practicable algorithm will be developed.
The thought of token ring algorithm is: whole system only has one piece of token, only has owner of a token just to have the qualification entering critical zone.When process i receives token from process i-1, it checks whether wants to enter critical zone, if so, then enters, and after it is released, passes token to process i+1; If do not want to enter, then directly token going down.
Feature: one is that token is easily lost, it is very difficult in fact detecting token lost; Two is easily occur process failure, and process failure compare be not easy recover.
Summary of the invention
Embodiments provide a kind of processing method of token, device and Message Processing cluster, be intended to the mutually exclusive algorithm that solution prior art provides, token lost is difficult to the problem monitored, process collapse is difficult to recovery.
On the one hand, provide a kind of processing method of token, described method comprises:
Token loop is set up in the application layer of Message Processing cluster;
When node adds or exit token loop, adopt simple majority Voting Algorithm to produce or destroy token;
In the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
Further, described simple majority Voting Algorithm generation or destruction token comprise:
Number of nodes in presetting message Processing Cluster to be N, N be more than or equal to 1 integer;
When monitoring the token loop that to be added by node and be made up of the N number of node in Message Processing cluster or exiting described token loop, the number of nodes P in token loop described in dynamic calculation;
When P be greater than (N+2)/2 give up tail round value time, if without token in token loop, then the token that automatically generation is new;
If have token in token loop, then maintain former token;
When P be less than or equal to (N+2)/2 give up tail round value time, if having token in token loop, then destroy this token.
Further, described when P be less than or equal to (N+2)/2 give up tail round value time, also comprise:
If without token in token loop, then system is hung up, and waits for that new node adds in described Message Processing cluster.
Further, the node in described Message Processing cluster comprises polling station node, ormal station node and monitoring station node.
On the other hand, provide a kind of processing unit of token, described device comprises:
Token loop sets up unit, for setting up token loop in the application layer of Message Processing cluster;
Token produces or destroys unit, during for adding when node or exiting token loop, adopts simple majority Voting Algorithm to produce or destroys token;
Token winding time supervision unit, for in the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
Further, described token generation or destruction unit comprise:
Number of nodes presetting module, for the number of nodes in presetting message Processing Cluster to be N, N be more than or equal to 1 integer;
Number of nodes statistical module, for when monitoring the token loop that to be added by node and be made up of the N number of node in Message Processing cluster or exiting described token loop, the number of nodes P in token loop described in dynamic calculation;
Token generation module, for be greater than as P (N+2)/2 give up tail round value time, if without token in token loop, then the token that automatically generation is new;
Token maintains module, if for having token in token loop, then maintain former token;
Token destroy module, for be less than or equal to as P (N+2)/2 house tail round value time, if having token in token loop, then destroy this token.
Further, described token generation or destruction unit also comprise:
Node adds wait module, if in token loop without token, then system hang up, wait for that new node adds in described Message Processing cluster.
Further, described device is built in the respective nodes in Message Processing cluster, and the node in described Message Processing cluster comprises polling station node, ormal station node and monitoring station node.
Again on the one hand, provide a kind of Message Processing cluster, described Message Processing cluster comprises polling station node, ormal station node and monitoring station node, is built-in with the processing unit of token as above in the respective nodes in described Message Processing cluster.
In the embodiment of the present invention, after the application layer of Message Processing cluster sets up token loop, when node adds or exit this token loop, adopt simple majority Voting Algorithm to produce or destroy token, avoiding in Message Processing cluster the token loop occurring more than 2 or 2; In addition, to the token winding time of each node monitor whether lose with Timeliness coverage token, and when token lost, rebuild token loop, solve prior art, while the loss of token such as cannot to supervise at the difficult problem in token loop, substantially increase the reliability of distributed cluster system.
Accompanying drawing explanation
Fig. 1 is the realization flow figure of the processing method of the token that the embodiment of the present invention one provides;
Fig. 2 is the realization flow figure of the route finding process that the embodiment of the present invention one provides;
Fig. 3 is the realization flow figure of the route synchronization process that the embodiment of the present invention one provides;
Fig. 4 is the realization flow figure of the token checking process that the embodiment of the present invention one provides;
Fig. 5 is the realization flow figure of the token production process that the embodiment of the present invention one provides;
Fig. 6 is the realization flow figure of the alternative space process that the embodiment of the present invention one provides;
Fig. 7 is the structured flowchart of the processing unit of the token that the embodiment of the present invention two provides.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, 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, be not intended to limit the present invention.
In embodiments of the present invention, token loop is set up in the application layer of Message Processing cluster; When node adds or exit token loop, adopt simple majority Voting Algorithm to produce or destroy token; In the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
Below in conjunction with specific embodiment, realization of the present invention is described in detail:
Embodiment one
Fig. 1 shows the realization flow of the processing method of the token that the embodiment of the present invention one provides, and details are as follows:
In step S101, set up token loop in the application layer of Message Processing cluster.
In step s 102, when node adds or exit token loop, adopt simple majority Voting Algorithm to produce or destroy token.
In embodiments of the present invention, the logical loops being split into two or more in order to avoid Message Processing cluster carries out " fissure " phenomenon of serializing work, therefore adopt the Voting Algorithm of " (N+2)/2 simple majority " automatically produce and destroy token, the algorithm of simple majority Voting Algorithm is as follows:
Number of nodes in presetting message Processing Cluster is N, N be more than or equal to 1 integer, when monitoring the token loop that to be added by node and be made up of the N number of node in Message Processing cluster or exiting described token loop, number of nodes P in token loop described in dynamic calculation, when P be greater than (N+2)/2 give up tail round value time, if without token in token loop, then automatically produce new token; If have token in token loop, then maintain former token; When P be less than or equal to (N+2)/2 give up tail round value time, if having token in token loop, then destroy this token, if token loop in without token, then system hang-up, wait for that new node adds in described Message Processing cluster.
According to above simple majority Voting Algorithm, odd number platform main frame (node) redundancy ability is also eager to excel than the even number platform main frame of odd number+1, so generally adopt odd number platform main frame to form Message Processing cluster when actual deployment.Can disaster tolerance 1 mechanical disorder for 3 Message Processing clusters, 5 Message Processing clusters can hold 1 fault or 2 simultaneous faultss, and by that analogy, this simple majority Voting Algorithm only holds disaster and the fault of minority machine node, but unit, improves reliability relatively.Below the failure rate of Message Processing cluster is calculated respectively for 3,5,7.
Prerequisite:
1), the probability broken down of all node host is roughly the same, and namely failure rate is q.
2), suppose that main frame occurred 1 fault in 100 days, then the failure rate q calculating this main frame is 1%, the like.
3), suppose that the failure rate q of All hosts is no more than 10%, namely occur in 10 days being less than 1 fault.
1 failure rate Q (1)=q
3 failure rate Q (3)=3q
2(1-q)+q
3=3q
2-2q
3
5 failure rate Q (5)=10q
3(1-q)
2+ 5q
4(1-q)+q
5=10q
3-15q
4+ 6q
5
7 failure rate Q (7)=35q
4(1-q)
3+ 21q
5(1-q)
2+ 7q
6(1-q)+q
7=35q
4-84q
5+ 70q
6-20q
7
3 principals and subordinates arbitrate failure rate Q (second month in a season)=2q
2(1-q)+q
3=2q
2-q
3
Table 1 shows the failure rate of Message Processing cluster:
Q(1) | Q(3) | Q(5) | Q(7) | Q (second month in a season) |
0.1 | 0.028 | 0.00856 | 0.002728 | 0.018 |
0.01 | 0.000298 | 9.8506E-06 | 3.4167E-07 | 0.000198 |
0.001 | 0.000002998 | 9.98501E-09 | 3.49161E-11 | 0.000001998 |
Table 1
From above-mentioned Message Processing clustering fault rate, failure rate declines very fast, and MPU global reliability can rise to 99.97% ~ 99.999999%.
Wherein, all nodes in Message Processing cluster have unified planning coding, and static configuration has type of site.The flow process that function describes and dissimilar website participates in of the mode that dissimilar website produces, the quantity of dissimilar website, dissimilar website refers to the table 2 in step S102, as shown in Table 2, the generation of token is coordinated by polling station, ormal station and monitoring station with destruction work.
In step s 103, in the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
In embodiments of the present invention, the transitive dependency of token is in setting up unique, stable, efficient token loop between each node.The management of token loop is comprised to the combination process such as election, node dynamically add, node dynamically exits of the basic procedures such as node discovery, route synchronization, token inspection, token generation, alternative space, winding supervision and monitoring station.Each node in Message Processing cluster is divided into monitoring station, ormal station, polling station by the management responsibility of token.
Table 2 shows the Functional division of each node
Table 2
Token loop management algorithm be based upon unified planning coding carries out to all nodes in message Processing Cluster basis on and static configuration type of site.
The foundation of token loop completes on route discovery, these two basic handling flow processs of route synchronization, and the foundation of token loop, the adding of node, the exiting of node, monitoring station election etc. all follow route discovery, route synchronization, these two basic procedures.
Table 3 shows the route discovery of token loop, the promoter of route synchronization, initiates the description of the content of condition and execution:
Table 3
Concrete, route discovery flow process as shown in Figure 2, comprises the following steps:
There is abnormal node, not receiving and be token and received the order of Systematical control rerouting in each route discovery winding time in <1>, new entrant, discovery node, then initiates route discovery instruction.
<2>, promoter attempt connecting the available succession of searching according to increasing progressively successively according to cluster global configuration from its adjacent node.
If it is all unsuccessful that <3> attempts connection, this node is hung up, and waits for that other nodes add.
<4>, as connect succession success, upgrade routing table self.
The route discovery instruction that <5>, promoter send, passes to succession.
<6>, succession compare promoter ID in route discovery instruction and self ID.
If <7> promoter ID> self ID, route discovery instruction stops, and initiates new route discovery instruction.
If <8> promoter ID< self ID, life cycle in route discovery instruction is subtracted 1, if be still greater than 0, then continue to transmit route discovery instruction by step <2>-<8GreatT.Gre aT.GT, otherwise stop transmitting.
If <9> promoter ID=self ID, illustrates that a circle has been transmitted in route discovery instruction, route instruction stops, and upgrade its routing table, this node is cluster monitoring station.
Route synchronization flow process as shown in Figure 3, comprises the following steps:
<1>, route discovery instruction normal termination and instruction sequence number are the maximum sequence number that present node is issued, and initiate route synchronization instruction by MPU monitoring station.
After <2>, succession receive route synchronization instruction, the relation of discrimination instruction promoter ID and self ID.
If <3> promoter ID> self ID, stops the transmission of this route synchronization, again initiates route synchronization instruction by this node.
If <4> promoter ID< self ID, upgrade self routing table and monitoring station information, life cycle in route synchronization instruction is subtracted 1, if be still greater than 0, continue to transmit by step <2>->LEssT.LTss T.LT5>, otherwise stop transmitting.
If <5> promoter ID=self ID, illustrates that a circle has been transmitted in route synchronization instruction, instruction stops.
The maintenance that table 4 shows token loop describes:
Table 4
Token work management comprises token inspection, token generation, alternative space, token winding time monitoring and token performance monitoring.
Table 5 shows the interaction content in token work management flow process:
Table 5
Wherein, token check process as shown in Figure 4, comprises the following steps:
<1>, initiate token by monitoring station and check instruction, and pass to its succession.
<2>, succession receive token and check instruction, the relation of discrimination instruction promoter ID and self ID.
If <3> promoter ID> self ID, stop the transmission that this token checks instruction, again initiate token by this node and check instruction.
If <4> promoter ID< self ID, the token mark in instruction is checked according to self-operating status update token, and the life cycle in instruction is subtracted 1, if be still greater than 0, by step <1>-<5GreatT.Gre aT.GT) continue to transmit instruction, otherwise stop transmitting.
If <5> promoter ID=self ID, illustrate and check that a circle has been transmitted in instruction, instruction stops.
<6>, monitoring station check the token mark in instruction by the token that receives, and initiation time and the time receiving token for the last time, differentiate in Message Processing cluster whether there is token.
Wherein, token produces flow process as shown in Figure 5, comprises the following steps:
<1>, monitoring station initiate token check process, and differentiate that token loop does not exist token.
<2>, judge that in routing table, effective number of nodes exceedes half.
<3>, monitoring station initiate the inquiry of maximum sequence number to all polling stations.
<4>, monitoring station Receive sequence number inquiry result, maximum sequence number M is got in all response result, and increase certain radix on this basis, form start sequence number N=M+ all nodes * maximum token step-length of new token.
<5>, monitoring station produce new token, and issue token update notification to all polling stations.
Wherein, alternative space flow process as shown in Figure 6, comprises the following steps:
<1>, node are after receiving the token of transmission, and mark the machine being had token is set to true, and response simultaneously returns.
When <2>, node have token, carry out serializing to being in the instruction sending no more than token step-length in buffering area, multicast transmission is carried out in combination simultaneously.
<3>, when the instruction of sequence of tokens reach token step-length or send buffering area send instructions without waiting, then upgrade token information and pass to its succession, this node enters armed state, and record enters the timestamp of armed state.
When <4>, alternative space are to succession, repeatedly transmit unsuccessfully if occur, this succession is changed into ' N ' down state in the state of the routing table of the machine, and trigger a route discovery instruction, count whether reach minimum working host quantity according to the available section upgraded in rear routing table again, if be less than this value token to stop transmitting, and send alarm.
<5>, node are in armed state, and the service order monitor the stand-by period and whether exceed the winding time of looking ahead, if exceed, the service order that stops looking ahead enters suspended state, triggers a route discovery instruction simultaneously.
Concrete, the supervision of token refers to and monitors the operation conditions of token in token loop, comprises token winding time monitoring and these two parts of token performance monitoring.Token winding time monitoring refers to that each node in Message Processing cluster all records the timestamp spreading out of token last time, monitoring station timing differentiates whether the time difference of this timestamp and current time exceeds the winding time of setting, if exceed, then need again to initiate route finding process, rebuild token loop.
Token performance monitoring is the time delay and throughput of being enclosed by monitoring station monitoring alternative space one, and monitoring result is transmitted with token.
By the supervision of monitoring station to the token winding time, token can be monitored whether lose, if lost, then rebuild token loop, produce new token, solve prior art, while the loss of token such as cannot to supervise at the difficult problem in token loop, substantially increase the reliability of distributed cluster system.
The present embodiment, with reference to token-ring network principle, after the application layer of Message Processing cluster sets up token loop, when node adds or exit this token loop, adopt simple majority Voting Algorithm to produce or destroy token, avoiding in Message Processing cluster the token loop occurring more than 2 or 2; In addition, to the token winding time of each node monitor whether lose with Timeliness coverage token, and when token lost, rebuild token loop, solve prior art, while the loss of token such as cannot to supervise at the difficult problem in token loop, substantially increase the reliability of distributed cluster system.
One of ordinary skill in the art will appreciate that all or part of step realized in the various embodiments described above method is that the hardware that can carry out instruction relevant by program has come, corresponding program can be stored in a computer read/write memory medium, described storage medium, as ROM/RAM, disk or CD etc.
Embodiment two
Fig. 7 shows the concrete structure block diagram of the processing unit of the token that the embodiment of the present invention two provides, and for convenience of explanation, illustrate only the part relevant to the embodiment of the present invention.Message Processing cluster comprises polling station node, ormal station node and monitoring station node, be built-in with the processing unit of described token in respective nodes in described Message Processing cluster, the processing unit 7 of this token comprises: token loop sets up unit 71, token produces or destroy unit 72 and token winding time supervision unit 73.
Wherein, token loop sets up unit 71, for setting up token loop in the application layer of Message Processing cluster;
Token produces or destroys unit 72, during for adding when node or exiting token loop, adopts simple majority Voting Algorithm to produce or destroys token;
Token winding time supervision unit 73, for in the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
Concrete, described token produces or destroys unit 72 and comprises:
Number of nodes presetting module is N for the number of nodes in presetting message Processing Cluster;
Number of nodes statistical module, for when monitoring the token loop that to be added by node and be made up of the N number of node in Message Processing cluster or exiting described token loop, the number of nodes P in token loop described in dynamic calculation;
Token generation module, for be greater than as P (N+2)/2 give up tail round value time, if without token in token loop, then the token that automatically generation is new;
Token maintains module, if for having token in token loop, then maintain former token;
Token destroy module, for be less than or equal to as P (N+2)/2 house tail round value time, if having token in token loop, then destroy this token.
Further, described token generation or destruction unit 72 also comprise:
Node adds wait module, if in token loop without token, then system hang up, wait for that new node adds in described Message Processing cluster.
Further, described device 7 is built in the respective nodes in Message Processing cluster, and the node in described Message Processing cluster comprises polling station node, ormal station node and monitoring station node.
The processing unit of the token that the embodiment of the present invention provides can be applied in the embodiment of the method one of aforementioned correspondence, and details, see the description of above-described embodiment one, do not repeat them here.
It should be noted that in the processing unit embodiment of above-mentioned token, included unit is carry out dividing according to function logic, but is not limited to above-mentioned division, as long as can realize corresponding function; In addition, the concrete title of each functional unit, also just for the ease of mutual differentiation, is not limited to protection scope of the present invention.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (9)
1. a processing method for token, is characterized in that, described method comprises:
Token loop is set up in the application layer of Message Processing cluster;
When node adds or exit token loop, adopt simple majority Voting Algorithm to produce or destroy token;
In the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
2. the method for claim 1, is characterized in that, described simple majority Voting Algorithm produces or destroys token and comprises:
Number of nodes in presetting message Processing Cluster to be N, N be more than or equal to 1 integer;
When monitoring the token loop that to be added by node and be made up of the N number of node in Message Processing cluster or exiting described token loop, the number of nodes P in token loop described in dynamic calculation;
When P be greater than (N+2)/2 give up tail round value time, if without token in token loop, then the token that automatically generation is new;
If have token in token loop, then maintain former token;
When P be less than or equal to (N+2)/2 give up tail round value time, if having token in token loop, then destroy this token.
3. method as claimed in claim 1 or 2, is characterized in that, described when P be less than or equal to (N+2)/2 house tail round value time, also comprise:
If without token in token loop, then system is hung up, and waits for that new node adds in described Message Processing cluster.
4. method as claimed in claim 1 or 2, it is characterized in that, the node in described Message Processing cluster comprises polling station node, ormal station node and monitoring station node.
5. a processing unit for token, is characterized in that, described device comprises:
Token loop sets up unit, for setting up token loop in the application layer of Message Processing cluster;
Token produces or destroys unit, during for adding when node or exiting token loop, adopts simple majority Voting Algorithm to produce or destroys token;
Token winding time supervision unit, for in the transmittance process of token, each node last in monitoring token loop spreads out of the timestamp of token, if the time difference monitoring the timestamp and current time that node last spreads out of token exceeds the winding time of setting, then initiate route finding process, rebuild token loop.
6. device as claimed in claim 5, is characterized in that, described token produces or destroys unit and comprises:
Number of nodes presetting module, for the number of nodes in presetting message Processing Cluster to be N, N be more than or equal to 1 integer;
Number of nodes statistical module, for when monitoring the token loop that to be added by node and be made up of the N number of node in Message Processing cluster or exiting described token loop, the number of nodes P in token loop described in dynamic calculation;
Token generation module, for be greater than as P (N+2)/2 give up tail round value time, if without token in token loop, then the token that automatically generation is new;
Token maintains module, if for having token in token loop, then maintain former token;
Token destroy module, for be less than or equal to as P (N+2)/2 house tail round value time, if having token in token loop, then destroy this token.
7. the device as described in claim 5 or 6, is characterized in that, described token produces or destroys unit and also comprises:
Node adds wait module, if in token loop without token, then system hang up, wait for that new node adds in described Message Processing cluster.
8. the device as described in claim 5 or 6, is characterized in that, described device is built in the respective nodes in Message Processing cluster, and the node in described Message Processing cluster comprises polling station node, ormal station node and monitoring station node.
9. a Message Processing cluster, it is characterized in that, described Message Processing cluster comprises polling station node, ormal station node and monitoring station node, is built-in with the processing unit of the token as described in any one of claim 5 to 8 in the respective nodes in described Message Processing cluster.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410550612.8A CN104468302B (en) | 2014-10-16 | 2014-10-16 | A kind of processing method and processing device of token |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410550612.8A CN104468302B (en) | 2014-10-16 | 2014-10-16 | A kind of processing method and processing device of token |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104468302A true CN104468302A (en) | 2015-03-25 |
CN104468302B CN104468302B (en) | 2018-03-30 |
Family
ID=52913709
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410550612.8A Expired - Fee Related CN104468302B (en) | 2014-10-16 | 2014-10-16 | A kind of processing method and processing device of token |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104468302B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107306289A (en) * | 2016-04-21 | 2017-10-31 | 中国移动通信集团重庆有限公司 | A kind of load-balancing method and equipment based on cloud computing |
CN108268386A (en) * | 2016-12-30 | 2018-07-10 | 英特尔公司 | Memory order in accelerating hardware |
CN109614403A (en) * | 2018-10-24 | 2019-04-12 | 北京三快在线科技有限公司 | The data consistency verification method and device of cluster service node |
CN112131012A (en) * | 2020-11-26 | 2020-12-25 | 九州华兴集成电路设计(北京)有限公司 | Token processing method, token processing device and computer readable storage medium |
US11593295B2 (en) | 2018-06-30 | 2023-02-28 | Intel Corporation | Apparatuses, methods, and systems for operations in a configurable spatial accelerator |
US11907713B2 (en) | 2019-12-28 | 2024-02-20 | Intel Corporation | Apparatuses, methods, and systems for fused operations using sign modification in a processing element of a configurable spatial accelerator |
US12086080B2 (en) | 2020-09-26 | 2024-09-10 | Intel Corporation | Apparatuses, methods, and systems for a configurable accelerator having dataflow execution circuits |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06112964A (en) * | 1992-09-30 | 1994-04-22 | Yokokawa Denshi Kiki Kk | Loop controller and its automatic changeover method in token loop lan |
CN1649319A (en) * | 2005-02-04 | 2005-08-03 | 西安电子科技大学 | Token transfer and manage method for Ad Hoc network |
CN1964353A (en) * | 2006-12-05 | 2007-05-16 | 上海大学 | A protocol stack facing wireless industrial control network |
CN103533661A (en) * | 2013-10-23 | 2014-01-22 | 中国舰船研究设计中心 | Wireless channel access control method based on virtual token |
-
2014
- 2014-10-16 CN CN201410550612.8A patent/CN104468302B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06112964A (en) * | 1992-09-30 | 1994-04-22 | Yokokawa Denshi Kiki Kk | Loop controller and its automatic changeover method in token loop lan |
CN1649319A (en) * | 2005-02-04 | 2005-08-03 | 西安电子科技大学 | Token transfer and manage method for Ad Hoc network |
CN1964353A (en) * | 2006-12-05 | 2007-05-16 | 上海大学 | A protocol stack facing wireless industrial control network |
CN103533661A (en) * | 2013-10-23 | 2014-01-22 | 中国舰船研究设计中心 | Wireless channel access control method based on virtual token |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107306289A (en) * | 2016-04-21 | 2017-10-31 | 中国移动通信集团重庆有限公司 | A kind of load-balancing method and equipment based on cloud computing |
CN107306289B (en) * | 2016-04-21 | 2020-11-27 | 中国移动通信集团重庆有限公司 | Load balancing method and device based on cloud computing |
CN108268386A (en) * | 2016-12-30 | 2018-07-10 | 英特尔公司 | Memory order in accelerating hardware |
US11593295B2 (en) | 2018-06-30 | 2023-02-28 | Intel Corporation | Apparatuses, methods, and systems for operations in a configurable spatial accelerator |
CN109614403A (en) * | 2018-10-24 | 2019-04-12 | 北京三快在线科技有限公司 | The data consistency verification method and device of cluster service node |
CN109614403B (en) * | 2018-10-24 | 2020-03-06 | 北京三快在线科技有限公司 | Data consistency checking method and device for cluster service node |
US11907713B2 (en) | 2019-12-28 | 2024-02-20 | Intel Corporation | Apparatuses, methods, and systems for fused operations using sign modification in a processing element of a configurable spatial accelerator |
US12086080B2 (en) | 2020-09-26 | 2024-09-10 | Intel Corporation | Apparatuses, methods, and systems for a configurable accelerator having dataflow execution circuits |
CN112131012A (en) * | 2020-11-26 | 2020-12-25 | 九州华兴集成电路设计(北京)有限公司 | Token processing method, token processing device and computer readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN104468302B (en) | 2018-03-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104468302A (en) | Token processing method and device and message handling cluster | |
US5923832A (en) | Method and apparatus for checkpointing in computer system | |
US8850141B2 (en) | System and method for mirroring data | |
WO2018072618A1 (en) | Method for allocating stream computing task and control server | |
CN103019889A (en) | Distributed file system and failure processing method thereof | |
CN108551397A (en) | The communication control method of network bridge device and application and more PLC master stations and more PLC slave stations | |
KR20110044858A (en) | Maintain data indetermination in data servers across data centers | |
CN105069152B (en) | data processing method and device | |
CN104469699B (en) | Cluster quorum method and more cluster coupled systems | |
CN107508694B (en) | Node management method and node equipment in cluster | |
CN102843259A (en) | Middleware self-management hot backup method and middleware self-management hot backup system in cluster | |
CN103634411A (en) | Real-time market data broadcasting system and real-time market data broadcasting method with state consistency | |
CN109391691A (en) | The restoration methods and relevant apparatus that NAS is serviced under a kind of single node failure | |
EP3427157B1 (en) | Cross-regional data transmission | |
Cheraghlou et al. | A novel fault-tolerant leach clustering protocol for wireless sensor networks | |
CN106817387A (en) | A kind of method of data synchronization, device and system | |
CN103516734A (en) | Data processing method, device and system | |
CN111309515B (en) | Disaster recovery control method, device and system | |
CN105812492B (en) | A kind of method of data synchronization and system | |
CN107071189A (en) | A kind of connection method of communication apparatus physical interface | |
CN104794026A (en) | Cluster instance and multi-data-source binding failover method | |
KR101358995B1 (en) | Method and system for managing high availability | |
CN109639640B (en) | Message sending method and device | |
CN101145955A (en) | Hot backup method, network management and network management system of network management software | |
CN113630445B (en) | Data storage method and device based on block chain network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180330 |