CN110505086A - A kind of method and device that distributed director is fault-tolerant - Google Patents
A kind of method and device that distributed director is fault-tolerant Download PDFInfo
- Publication number
- CN110505086A CN110505086A CN201910759325.0A CN201910759325A CN110505086A CN 110505086 A CN110505086 A CN 110505086A CN 201910759325 A CN201910759325 A CN 201910759325A CN 110505086 A CN110505086 A CN 110505086A
- Authority
- CN
- China
- Prior art keywords
- controller
- interchanger
- path
- load
- flow load
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0663—Performing the actions predefined by failover planning, e.g. switching to standby network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
- H04L45/123—Evaluation of link metrics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The present invention relates to a kind of method and devices that distributed director is fault-tolerant, method includes the following steps: each controller state in monitoring distributed system;In response to monitoring that controller breaks down, the flow load of each of the links in current network topology is acquired;Based on shortest path, transfer path as the minimal path flow load of each interchanger of weight calculation and is determined using the flow load of each interchanger in the control domain of failed controller to the link in the path of normal controller;According to the size order of the minimal path flow load of each interchanger, interchanger is handed over to corresponding normal controller according to corresponding transfer path respectively.When the fault-tolerant method of distributed director of the invention solves controller generation Single Point of Faliure, ensuring that the switch network for not causing overweight flow failed controller to be managed in the case where burden link hands over to normal controller and continue to manage, guarantee the normal use of network system.
Description
Technical field
The present invention relates to technical field of communication network.The invention further relates to a kind of methods that distributed director is fault-tolerant
And device.
Background technique
It is fast changing with big data, the rapid development of artificial intelligence field and Internet service, to data center
The demand of network is more more and more intense.The high stable and high-performance of network are also unable to satisfy real-time business demand, Ren Menyue at present
More to pay attention to the flexibility and agility of network.Under such circumstances, the customized network of software comes into being, forwarding and control
The decoupling framework for making separation, can just improve the problem of traditional network flexibility difference and expansible hardly possible.
SDN network carries out centralized management to data center network by SDN controller, and network management personnel can pass through
SDN controller obtains the global network information of data center, and SDN controller is write soft by programmable business support
Part application control lower layer forwarding device, greatly simplifies network management difficulty, effectively improves network flexibility.
Simultaneously, the controller of centralization also becomes a big hidden danger of network stabilization, once single-point occurs for controller
Failure, the interchanger which is managed will be unable to be forwarded according to the service logic of controller, will lead to a large amount of number
According to packet loss, network is caused to can not work normally.
It is, therefore, desirable to provide a kind of solution to the problems described above, so that when Single Point of Faliure occurs for SDN controller, it will
Its switch network managed, which hands over to normal controller, to be continued to manage, and guarantees the normal work of network service, while also not
Overweight flow is caused to bear link.
Summary of the invention
On the one hand, the present invention is based on above-mentioned purposes to propose a kind of method that distributed director is fault-tolerant, and this method includes
Following steps:
Monitor each controller state in distributed system;
In response to monitoring that controller breaks down, the flow load of each of the links in current network topology is acquired;
Based on shortest path, with each interchanger in the control domain of failed controller to the chain in the path of normal controller
The flow load on road is the minimal path flow load of each interchanger of weight calculation and determines transfer path;
According to the size order of the minimal path flow load of each interchanger, will be exchanged according to corresponding path of transferring respectively
Machine is handed over to corresponding normal controller.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein respectively being controlled in monitoring distributed system
Device state further comprises: monitoring the topology information of the control domain of each controller and is stored in database.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein monitor the control domain of each controller
Topology information is simultaneously stored in database and further comprises: in response to monitoring that network topological information changes, according to changing
Interchanger active reporting information updating database in topology information.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein in response to monitoring controller
Failure, the flow load for acquiring each of the links in current network topology further comprises:
Malfunction monitoring is carried out between adjacent three controllers, comprising:
The timing of first controller sends test information to second controller, third controller, and record is sent in the database
Moment;
Test information is not received in given time in response to second controller, and second controller is inquired to third controller
Whether test information is received;
Test information is not received yet in given time in response to third controller, inquires recorded sending instant and meter
Calculate the time difference with current time;
It is more than threshold value in response to the time difference, determines that the first controller breaks down.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein in response to monitoring controller
Failure, the flow load for acquiring each of the links in current network topology further comprises:
In response to monitoring that controller breaks down, compare in the control domain of failed controller switch requirement whether be more than
The summation of the idle control load of normal controller;
The idle control load of normal controller is less than in response to switch requirement in the control domain of failed controller
Summation acquires the flow load of each of the links in current network topology.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein in response to monitoring controller
Failure, the flow load for acquiring each of the links in current network topology further comprises:
It is more than the total of the idle control load of normal controller in response to switch requirement in the control domain of failed controller
With the warning of sending increase controller.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein based on shortest path with Fault Control
The flow load of each interchanger to the link in the path of normal controller in the control domain of device is each interchanger of weight calculation
Minimal path flow load and determine transfer path further comprise:
Interchanger is calculated to the different paths of a number of other controllers based on shortest path;
The flow for calculating separately all links in each path based on shortest path using the flow load of link as weight is negative
The sum of carry;
Using the smallest path of the sum of flow load in each path as selected transfer path, to transfer the flow in path
Minimal path flow load of the sum of the load as interchanger.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein with the minimal path runoff of each interchanger
Interchanger is handed over to corresponding normal controller according to corresponding transfer path respectively and further comprises by the sequence for measuring load:
Judge whether the idle control load of controller to be handed over to meets the demand of interchanger to be transferred;
The demand for meeting interchanger to be transferred in response to the idle control load of controller to be handed over to, will be wait transfer
Interchanger be handed over to controller to be handed over to.
The embodiment of the fault-tolerant method of distributed director according to the present invention, wherein with the minimal path runoff of each interchanger
Interchanger is handed over to corresponding normal controller according to corresponding transfer path respectively and further comprises by the sequence for measuring load:
The demand of interchanger to be transferred is unsatisfactory in response to the idle control load of controller to be handed over to, it will be wait move
Controller extremely is handed over to exclude computer capacity and recalculate the minimal path flow load of interchanger to be transferred.
On the other hand, the invention also provides a kind of device that distributed director is fault-tolerant, which includes:
At least one processor;With
Memory, the memory are stored with the program code that processor can be run, which runs by processor
Shi Zhihang method as described above.
By adopting the above technical scheme, the present invention at least has the following beneficial effects: that distributed director of the invention is fault-tolerant
Method when solving SDN controller Single Point of Faliure occurs, interchanger which is managed can not normal operation lead
The problem of causing network can not work normally, this method have been played in the case where Single Point of Faliure occurs for SDN controller to failure control
Device processed carries out fault-tolerant effect, and the switch network that failed controller is managed, which hands over to normal controller, to be continued to manage, and
And consider that the flow load of link is completed interchanger and transferred, guarantee the normal use of network, while ensuring not cause link
The flow burden of weight.
The present invention provides the various aspects of embodiment, should not be taken to limit protection scope of the present invention.According to retouching herein
The technology stated is contemplated that other embodiment, this is to those skilled in the art in research the following drawings and specific
It will be apparent after embodiment, and these embodiments intention is comprised in scope of the present application.
Below with reference to the accompanying drawings it is explained in greater detail and describes the embodiment of the present invention, but they should not be construed as this
The limitation of invention.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be in the prior art and embodiment description
Required attached drawing is briefly described, and the component in attached drawing is not drawn necessarily to scale, and can be omitted relevant
Element, or ratio may be amplified in some cases, it is described herein novel special to emphasize and to be clearly shown
Sign.In addition, as it is known in the art, structural order can be arranged differently.
Fig. 1 shows the schematic block diagram of the embodiment of the fault-tolerant method of distributed director according to the present invention.
Specific embodiment
Although the present invention can be implemented in a variety of manners, it is shown in the accompanying drawings and will be described below some show
Example property and non-limiting embodiment, it is understood that, the disclosure will be considered as that example of the invention is not intended this
Invention is limited to illustrated specific embodiment.
Fig. 1 shows the schematic block diagram of the embodiment of the fault-tolerant method of distributed director according to the present invention.Such as figure
Shown, this method at least includes the following steps:
S1: each controller state in monitoring distributed system;
S2: in response to monitoring that controller breaks down, the flow load of each of the links in current network topology is acquired;
S3: it is based on shortest path, with each interchanger in the control domain of failed controller into the path of normal controller
Link flow load be each interchanger of weight calculation minimal path flow load and determine transfer path;
S4:, respectively will according to corresponding path of transferring according to the size order of the minimal path flow load of each interchanger
Interchanger is handed over to corresponding normal controller.
According to the method for the present invention, step S1 first will monitor the state of each controller in distributed system to send out in time
The failure of existing controller simultaneously carries out fault-tolerant processing.The interchanger managed by failed controller is between other normal controllers
In the presence of the selection that many kinds are transferred, and such as theoretically, interchanger can transfer any normal controller to, and exchange
Machine can transfer a certain controller to by different paths, it is therefore desirable to which to the transfer object of interchanger, (i.e. band is handed over to
Controller) and transfer path targetedly selected.Selection gist of the invention is according to link each in network topology
Flow load and selected based on shortest path interchanger transfer object and transfer path.For this purpose, when having monitored control
After device breaks down, the flow load of each of the links in the acquisition current network topology of step S2 is executed.For example, the flow of link
Being fully loaded with is 10,000,000, and the flow load of link a is 6,000,000, and residual flow load is 4,000,000.The flow load of each link has been determined
Afterwards, step S3 is based on shortest path with each interchanger in the control domain of failed controller to the chain in the path of normal controller
The flow load on road is the minimal path flow load of each interchanger described in weight calculation and determines transfer path.That is,
In order to the interchanger that is managed to failed controller select one it is suitable transfer object (i.e. band be handed over to controller), according to
The embodiment of the present invention, based on shortest path and using the flow load of link each in path as weight, such as aforementioned link a's
In the case where flow load, the weight of link a is 6.The above both sides factor is comprehensively considered in all normal controllers
Minimal path flow load is calculated for interchanger, select optimal transfer object based on this and transfers path.It is each determining
Behind the transfer path of interchanger, step S4 is with the sequence of the minimal path flow load of identified each interchanger respectively according to phase
Interchanger is handed over to corresponding normal controller by the transfer path answered.So far, ensure not cause link as far as possible it is overweight
In the case that flow is born, the transfer of the interchanger managed failed controller is completed, ensure that distributed system just
It is often used.
Will be discussed below further embodiment of the present invention, it should be noted that it is mentioned that the step of number do not having
In the case where having specified otherwise, be only used for it is convenient explicitly indicate the step, do not limit the sequence of the step.
In some embodiments of the fault-tolerant method of distributed director of the invention, step S1 is monitored in distributed system
Each controller state further comprises: S11: monitoring the topology information of the control domain of each controller and is stored in database.Namely
It says, in these embodiments, in order to realize the fault-tolerant of controller, has not only monitored the state of controller itself, while also monitoring
The topology information of the control domain of each controller (i.e. the topological view of the interchanger of controller and its management), for example including each chain
The information such as link information, the flow entry on road, and the topology information monitored is saved in the database to transfer exchange
It is used when machine.
In several embodiments of the fault-tolerant method of distributed director of the invention, step S11 monitors each controller
The topology information of control domain is simultaneously stored in database and further comprises: S12: in response to monitoring that network topological information changes,
According to the topology information in the information updating database of changed interchanger active reporting.In general, in most of distributions
In system, the control domain topology information of each controller is obtained by way of timing scan, poll, these information are abstracted
Change and handles and be persisted in database.It can thus ensure that once Single Point of Faliure occurs for some controller, it can be according to number
According to the newest network information before the acquisition of information failure in library, foundation is provided for the transfer of control domain.However, there are a kind of comparisons
The upgrading Network topological information that special situation, i.e. controller just will acquire is into database, its network managed immediately
Topology changes, and the controller is before next controller actively obtains and stores the timing cycle of network topological information
Failure has occurred, the topology information in database is just inconsistent with real network topology information at this time.Therefore, implementation of the invention
Example provides the monitoring mechanism of step S12, as long as monitoring that network topological information changes, changed interchanger
Active transmission message (such as in the form of packet-in message) reports the variation, subsequent controller to controller
The topology information to timely update in database, avoid topology information when breaking down in database just with real network topology
The inconsistent problem of information.
In one or more embodiments of the fault-tolerant method of distributed director of the invention, step S2 is in response to monitoring
It breaks down to controller, the flow load for acquiring each of the links in current network topology further comprises: being controlled at adjacent three
Malfunction monitoring is carried out between device processed, comprising:
The timing of S21: the first controller sends test information to second controller, third controller, records in the database
Sending instant;
S22: not receiving test information in given time in response to second controller, and second controller is to third controller
Whether inquiry receives test information;
S23: test information is not received yet in given time in response to third controller, inquires recorded sending instant
And the time difference of calculating and current time;
S24: being more than threshold value in response to the time difference, determines that the first controller breaks down.
Specifically, for example the first controller A, second controller B, third controller C are adjacent, step S21 first is controlled
Device A timing sends test information, such as Hello information to second controller B, third controller C, while recording in the database
At the time of this time sending Hello information.If second controller B, third controller C can formulate reception arrive this
Hello then the first controller A is worked normally, and receiving time is stored in database.If second controller B fails
Given time receives test information Hello, then whether step S22 executes second controller B inquires it to third controller C
Receive test information Hello.If discovery third controller C is also not received by test information Hello at this time, then recognize
Be likely to occur failure for the first controller A, the sending instant for needing the further query steps S21 of step S23 to be recorded with work as
The time difference at preceding moment.If the time difference has been more than pre-set threshold value, step S24 is based on this and judges the first control
Device A breaks down, and the interchanger in the control domain to first controller A is needed to transfer.So far, in these embodiments,
Controller failure is detected by two-way converting technology, and have passed through two kinds of validation testings, improves fault detection accuracy rate.
In some embodiments of the fault-tolerant method of distributed director of the invention, step S2 is in response to monitoring to control
Device breaks down, and the flow load for acquiring each of the links in current network topology further comprises:
S25: in response to monitoring that controller breaks down, whether compare in the control domain of failed controller switch requirement
More than the summation of the idle control load of normal controller;
S26: the idle control for being less than normal controller in response to switch requirement in the control domain of failed controller is negative
The summation of load acquires the flow load of each of the links in current network topology.
After having monitored that controller breaks down, progress controller load is first had in these embodiments and whether can
Bear the judgement of transfering work.That is, step S25 in the case where being determined that failure has occurred in controller, compares first
It whether is more than that the idle of normal controller controls the summation loaded compared with switch requirement in the control domain of failed controller, with judgement
Whether other normal controllers in distributed system are enough to undertake the pipe for the interchanger for managing the controller to break down
The work of reason.If switch requirement is less than the total of the idle control load of normal controller in the control domain of failed controller
With, then it is assumed that the normal controller in system undertakes interchanger to be transferred enough at present, therefore in order to determine further move
Cross-channel diameter acquires the flow load of each of the links in current network topology to execute subsequent step.
In some embodiments of the fault-tolerant method of distributed director of the invention, step S2 is in response to monitoring to control
Device breaks down, and the flow load for acquiring each of the links in current network topology further comprises:
S27: being more than the idle control load of normal controller in response to switch requirement in the control domain of failed controller
Summation, issue increase controller warning.
I.e. if the idle control that switch requirement has been over normal controller in the control domain of failed controller is born
The summation of load, then it is assumed that the normal controller in system does not undertake interchanger to be transferred enough at present, and system needs to increase at this time
Add additional controller that could complete to undertake the control domain of failed controller with additional controller to the fault-tolerant of failed controller
The work for continuing management of interior interchanger.In this case, it is therefore necessary to issue the warning for increasing controller so as to the people that works
Member is artificial or increases necessary controller by modes such as other auto-programmings to guarantee the normal operation of network.
In several embodiments of the fault-tolerant method of distributed director of the invention, step S3 is based on shortest path with event
The flow load for hindering each interchanger to the link in the path of normal controller in the control domain of controller is that weight calculation is each
The minimal path flow load of interchanger simultaneously determines that transferring path further comprises:
S31: interchanger is calculated to the different paths of a number of other controllers based on shortest path;
S32: the stream of all links in each path is calculated separately based on shortest path using the flow load of link as weight
The sum of amount load;
S33: using the smallest path of the sum of flow load is as selected transfer path in each path, to transfer path
Minimal path flow load of the sum of the flow load as interchanger.
If the step S3 having been described in previous embodiment is in order to which the interchanger managed to failed controller selects one
It is suitable to transfer object (i.e. band be handed over to controller), it is power based on shortest path and with the flow load of link each in path
Weight, the factor for comprehensively considering these two aspects calculates minimal path flow load in all normal controllers for interchanger, with this
Based on select optimal transfer object and transfer path.In some embodiments, step S3 can be further by three steps
It realizes, wherein step S31 first calculates interchanger to the different paths of a number of other controllers based on shortest path.Theoretically,
Interchanger can transfer a certain controller to by different paths, therefore in those embodiments of the invention, selection exchange
The machine path of shortest path as the interchanger to the controller into the different paths of a certain controller.With this criterion calculation
Shortest path of the interchanger to different controllers.Later step S32 is based on shortest path by weight of the flow load of link
Calculate separately the sum of the flow load of all links in each path.Theoretically, interchanger can transfer any normal control to
Device processed, it is therefore desirable to further judge for interchanger to be handed over to which controller is only optimal selection, therefore of the invention
Using the flow load of link determined by step S2 as each path determined by weight calculation step S31 in these embodiments
The sum of flow load.For example, interchanger contains link a, link b, link into the path of different controllers, in first path
C, wherein link a flow load is 6,000,000, link a flow load is 3,000,000, link a flow load is 2,000,000, then the flow in the path
The sum of load 6+3+2=11;Link a, link d are contained in second path, wherein link d flow load is 9,000,000, then the path
Flow load 6+9=15;Link a, link e are contained in third path, wherein link e flow load is 7,000,000, then the path
Flow load 6+7=13.And so on calculate interchanger to each path of different controllers the sum of flow load.Then exist
Step S33 is compared the sum of the flow load in each path, using in each path in the sum of flow load the smallest path as
Selected transfer path, using the sum of flow load for transferring path as the minimal path flow load of interchanger.On such as
It states in example, due to the sum of the flow load of first path minimum, selected first path, which is used as, transfers path, and the first via
The corresponding controller of diameter also just becomes the transfer object (controller i.e. to be handed over to) of the interchanger, with the transfer path
Minimal path flow load of the sum of the flow load 11 as the interchanger.
In one or more embodiments of the fault-tolerant method of distributed director of the invention, step S4 is with each interchanger
The sequence of minimal path flow load transfer path according to corresponding respectively interchanger be handed over to corresponding normal controller
Further comprise:
S41: judge whether the idle control load of controller to be handed over to meets the demand of interchanger to be transferred;
S42: meeting the demand of interchanger to be transferred in response to the idle control load of controller to be handed over to, will be to
The interchanger of transfer is handed over to controller to be handed over to.
After determining the transfer path of each interchanger, step S4 is with the minimal path flow load of identified each interchanger
Sequence transfer path according to corresponding respectively interchanger be handed over to corresponding normal controller.It wherein further comprise judgement
Whether controller (i.e. transfer object) to be handed over to can undertake the management work of interchanger to be transferred, therefore step S41
Judge whether the idle control load of controller to be handed over to meets the demand of interchanger to be transferred.If meet demand,
Then follow the steps the process that interchanger to be transferred is handed over to controller to be handed over to of S42.
In the fault-tolerant further embodiment of a method of distributed director of the invention, step S4 with each interchanger most
Interchanger is handed over to corresponding normal controller into one according to corresponding transfer path respectively by the sequence of small path flow load
Step includes:
S43: being unsatisfactory for the demand of interchanger to be transferred in response to the idle control load of controller to be handed over to, will
Controller to be handed over to excludes computer capacity and recalculates the minimal path flow load of interchanger to be transferred.
After determining the transfer path of each interchanger, judge whether controller to be handed over to (i.e. transfer object) can hold
When the management work of the interchanger of magnanimous transfer, if it find that the idle control load of controller to be handed over to is unsatisfactory for wait move
The demand of the interchanger of friendship, that is, controller to be handed over to cannot undertake the management work of the interchanger, then just needing should
Controller excludes candidate range and returns to the minimal path flow load that step S3 recalculates interchanger to be transferred, i.e., heavy
New selection transfers path and transfers object.
On the other hand, the invention also provides a kind of device that distributed director is fault-tolerant, which includes: at least one
Processor;And memory, the memory are stored with the program code that processor can be run, which runs by processor
Any foregoing method of Shi Zhihang.
Computer readable storage medium (such as memory) as described herein can be volatile memory or non-volatile
Memory, or may include both volatile memory and nonvolatile memory.As an example and not restrictive, it is non-easy
The property lost memory may include read-only memory (ROM), programming ROM (PROM), electrically programmable ROM (EPROM), electrically-erasable
Programming ROM (EEPROM) or flash memory.Volatile memory may include random access memory (RAM), which can
To serve as external cache.As an example and not restrictive, RAM can be obtained in a variety of forms, such as synchronous
RAM (DRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate SDRAM (DDR SDRAM), enhancing SDRAM
(ESDRAM), synchronization link DRAM (SLDRAM) and directly Rambus RAM (DRRAM).The storage of disclosed aspect is set
The standby memory for being intended to including but not limited to these and other suitable type.
By adopting the above technical scheme, the present invention at least has the following beneficial effects: that distributed director of the invention is fault-tolerant
Method when solving SDN controller Single Point of Faliure occurs, interchanger which is managed can not normal operation lead
Cause network the problem of can not work normally, in the case where Single Point of Faliure occurs for SDN controller, played to failed controller into
The fault-tolerant effect of row, the switch network that failed controller is managed, which hands over to normal controller, to be continued to manage, and is considered
The flow load of link is completed interchanger and is transferred, and guarantees the normal use of network, while ensuring not cause link overweight stream
Amount burden.
It should be understood that technically in feasible situation, above in relation to technical characteristic cited by different embodiments
It can be combined with each other, to form the additional embodiment in the scope of the invention.In addition, particular example as described herein and embodiment
It is non-limiting, and structure explained above, step and sequence can be revised accordingly without departing from the present invention
Protection scope.
In this application, the use of disjunctive is intended to include conjunction.Fixed or indefinite article use is not intended to
Indicate radix.Specifically, be intended to indicate that the reference of "the" object or " one " and "one" object can in multiple such objects
One of energy.However, removing and non-clearly limiting although element disclosed by the embodiments of the present invention can be described or be required in the form of individual
It is made as odd number, it is understood that be multiple.In addition it is possible to use conjunction "or" conveys simultaneous feature, rather than
Repelling programs.In other words, conjunction "or" is understood to include "and/or".Term " includes " is inclusiveness and has
Range identical with "comprising".
Above-described embodiment, especially any " preferably " embodiment are the possibility examples of embodiment, and just to clear
Chu understands the principle of the present invention and proposes.It, can without departing substantially from the spirit and principle of technique described herein
To make many change and modification to above-described embodiment.All modifications are intended to be included in the scope of the present disclosure.
Claims (10)
1. a kind of method that distributed director is fault-tolerant, which is characterized in that the described method comprises the following steps:
Monitor each controller state in distributed system;
In response to monitoring that controller breaks down, the flow load of each of the links in current network topology is acquired;
Based on shortest path, with each interchanger in the control domain of the failed controller to the chain in the path of normal controller
The flow load on road is the minimal path flow load of each interchanger described in weight calculation and determines transfer path;
It, respectively will according to corresponding path of transferring according to the size order of the minimal path flow load of each interchanger
The interchanger is handed over to corresponding normal controller.
2. the method according to claim 1, wherein each controller state is into one in the monitoring distributed system
Step includes:
It monitors the topology information of the control domain of each controller and is stored in database.
3. according to the method described in claim 2, it is characterized in that, the topology information of the control domain of each controller of the monitoring simultaneously
Deposit database further comprises:
In response to monitoring that network topological information changes, according to the information updating institute of changed interchanger active reporting
State the topology information in database.
4. the method according to claim 1, wherein described in response to monitoring that controller breaks down, acquisition
The flow load of each of the links further comprises in current network topology:
Malfunction monitoring is carried out between adjacent three controllers, comprising:
The timing of first controller sends test information to second controller, third controller, records sending instant in the database;
The test information is not received in given time in response to the second controller, and the second controller is to described
Whether the inquiry of three controllers receives the test information;
The test information is not received yet in given time in response to third controller, inquires the sending instant recorded
And the time difference of calculating and current time;
It is more than threshold value in response to the time difference, determines that first controller breaks down.
5. the method according to claim 1, wherein described in response to monitoring that controller breaks down, acquisition
The flow load of each of the links further comprises in current network topology:
In response to monitoring that controller breaks down, whether the demand for comparing interchanger in the control domain of failed controller is more than just
The summation of the idle control load of normal controller;
The summation of the idle control load of normal controller is less than in response to switch requirement in the control domain of failed controller,
Acquire the flow load of each of the links in current network topology.
6. according to the method described in claim 5, it is characterized in that, described in response to monitoring that controller breaks down, acquisition
The flow load of each of the links further comprises in current network topology:
It is more than the summation of the idle control load of normal controller, hair in response to switch requirement in the control domain of failed controller
Increase the warning of controller out.
7. the method according to claim 1, wherein it is described based on shortest path with the control of the failed controller
The flow load of each interchanger to the link in the path of normal controller in domain processed is each interchanger described in weight calculation
Minimal path flow load simultaneously determines that transferring path further comprises:
Interchanger is calculated to the different paths of a number of other controllers based on shortest path;
Calculated separately using the flow load of link as weight based on shortest path all links in each path flow load it
With;
Using the smallest path of the sum of flow load in each path as selected transfer path, with the transfer path
Minimal path flow load of the sum of the flow load as the interchanger.
8. the method according to claim 1, wherein the minimal path flow with each interchanger
The interchanger is handed over to corresponding normal controller according to corresponding transfer path respectively by the sequence of load:
Judge whether the idle control load of controller to be handed over to meets the demand of interchanger to be transferred;
The demand for meeting the interchanger to be transferred in response to the idle control load of the controller to be handed over to, by institute
It states interchanger to be transferred and is handed over to the controller to be handed over to.
9. according to the method described in claim 8, it is characterized in that, the minimal path flow with each interchanger
The interchanger is handed over to corresponding normal controller according to corresponding transfer path respectively by the sequence of load:
The demand of the interchanger to be transferred is unsatisfactory in response to the idle control load of the controller to be handed over to, it will
The controller to be handed over to excludes computer capacity and recalculates the minimal path flow of the interchanger to be transferred
Load.
10. a kind of device that distributed director is fault-tolerant, which is characterized in that described device includes:
At least one processor;With
Memory, the memory are stored with the program code that processor can be run, and said program code is run by processor
Shi Zhihang method as claimed in any one of claims 1-9 wherein.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910759325.0A CN110505086B (en) | 2019-08-16 | 2019-08-16 | Fault-tolerant method and device for distributed controller |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910759325.0A CN110505086B (en) | 2019-08-16 | 2019-08-16 | Fault-tolerant method and device for distributed controller |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110505086A true CN110505086A (en) | 2019-11-26 |
CN110505086B CN110505086B (en) | 2023-01-06 |
Family
ID=68587550
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910759325.0A Active CN110505086B (en) | 2019-08-16 | 2019-08-16 | Fault-tolerant method and device for distributed controller |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110505086B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111131464A (en) * | 2019-12-26 | 2020-05-08 | 安徽朋德信息科技有限公司 | Distributed terminal management system for laboratory of colleges and universities |
CN111211854A (en) * | 2019-12-26 | 2020-05-29 | 北京控制工程研究所 | Distributed fault-tolerant avionics control system and method based on time deterministic network |
CN115933565A (en) * | 2022-12-23 | 2023-04-07 | 广东职业技术学院 | AGV task exchange method, device, system and medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105119840A (en) * | 2015-09-01 | 2015-12-02 | 华东师范大学 | Data center network load balancing method based on SDN and employing fat-tree topological structure |
CN106559254A (en) * | 2015-12-29 | 2017-04-05 | 国网智能电网研究院 | SDN multiple-domain networks device and implementation method based on both-end mouth switch |
CN107819695A (en) * | 2017-10-19 | 2018-03-20 | 西安电子科技大学 | A kind of distributed AC servo system SiteServer LBS and method based on SDN |
CN108270669A (en) * | 2016-12-30 | 2018-07-10 | 中兴通讯股份有限公司 | Business recovery device, master controller, the system and method for SDN network |
-
2019
- 2019-08-16 CN CN201910759325.0A patent/CN110505086B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105119840A (en) * | 2015-09-01 | 2015-12-02 | 华东师范大学 | Data center network load balancing method based on SDN and employing fat-tree topological structure |
CN106559254A (en) * | 2015-12-29 | 2017-04-05 | 国网智能电网研究院 | SDN multiple-domain networks device and implementation method based on both-end mouth switch |
CN108270669A (en) * | 2016-12-30 | 2018-07-10 | 中兴通讯股份有限公司 | Business recovery device, master controller, the system and method for SDN network |
CN107819695A (en) * | 2017-10-19 | 2018-03-20 | 西安电子科技大学 | A kind of distributed AC servo system SiteServer LBS and method based on SDN |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111131464A (en) * | 2019-12-26 | 2020-05-08 | 安徽朋德信息科技有限公司 | Distributed terminal management system for laboratory of colleges and universities |
CN111211854A (en) * | 2019-12-26 | 2020-05-29 | 北京控制工程研究所 | Distributed fault-tolerant avionics control system and method based on time deterministic network |
CN115933565A (en) * | 2022-12-23 | 2023-04-07 | 广东职业技术学院 | AGV task exchange method, device, system and medium |
CN115933565B (en) * | 2022-12-23 | 2023-10-20 | 广东职业技术学院 | AGV task exchange method, device, system and medium |
Also Published As
Publication number | Publication date |
---|---|
CN110505086B (en) | 2023-01-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110505086A (en) | A kind of method and device that distributed director is fault-tolerant | |
US20200287976A1 (en) | Adaptive routing of branch traffic in software-defined wide area network (sdwan) deployments | |
CN106095589B (en) | A kind of method, apparatus and system for distributing subregion | |
CN102415054B (en) | Node device and communication method | |
CN106713487A (en) | Data synchronization method and device | |
US20150055453A1 (en) | Detour path calculation method in emergency | |
US9825462B2 (en) | Power network system, control method for power network system, and control program for power network system | |
CN109495300A (en) | A kind of reliable SDN virtual network mapping algorithm | |
CN109783214A (en) | Task schedule control system | |
CN102204188A (en) | Routing computation method and host node device in virtual network element | |
CN108696581A (en) | Caching method, device, computer equipment and the storage medium of distributed information | |
CN106598789A (en) | Container service disaster recovery method and device, production site and disaster recovery backup site | |
CN105468442A (en) | Application and information movement in a cloud environment | |
CN106068626A (en) | Load balancing in distributed network management framework | |
CN113259148B (en) | Alarm correlation detection method, system and medium based on federal learning | |
CN106953926A (en) | A kind of method for routing and device | |
CN109039795A (en) | A kind of Cloud Server resource monitoring method and system | |
CN104426968B (en) | Data managing method and device | |
US10474644B2 (en) | Systems and methods for optimizing selection of a replication data node in a distributed file system | |
CN105893176B (en) | A kind of management method and device of network store system | |
CN110119314A (en) | A kind of server calls method, apparatus, server and storage medium | |
WO2016180436A1 (en) | Pattern-based data collection for a distributed stream data processing system | |
CN101207518B (en) | Asynchronization maintenance system facing to distributed resource node | |
CN113890850B (en) | Route disaster recovery system and method | |
Chandrakala et al. | Improved data availability and fault tolerance in MANET by replication |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |