CN112702186B - Multi-controller deployment method and terminal in SD-WAN environment - Google Patents
Multi-controller deployment method and terminal in SD-WAN environment Download PDFInfo
- Publication number
- CN112702186B CN112702186B CN202011322406.3A CN202011322406A CN112702186B CN 112702186 B CN112702186 B CN 112702186B CN 202011322406 A CN202011322406 A CN 202011322406A CN 112702186 B CN112702186 B CN 112702186B
- Authority
- CN
- China
- Prior art keywords
- controller
- time delay
- wan
- delay
- switch
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 108020001568 subdomains Proteins 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 abstract description 20
- 238000005192 partition Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 241000287196 Asthenes Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
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
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression 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
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
Abstract
The invention provides a multi-controller deployment method and a terminal in an SD-WAN environment, which are used for acquiring a topological structure of an SD-WAN network, a first time delay between a switch and a controller in the SD-WAN network and a second time delay between controllers in the SD-WAN network; dividing the SD-WAN network into a first preset number of subfields with a first time delay and a second time delay which are minimum according to a topological structure; acquiring a third time delay and a preset load balancing constraint between the switch and the controller in each sub-domain; determining the position of the controller in each sub-domain which minimizes the first delay, the second delay and the third delay according to the load balancing constraint; the invention divides the software defined network into subdomains, and on the basis, the deployment position of the controller in each subdomain is determined by taking the time delay between the minimized controller and the switch as an optimization target, so that the time delay between the controller and the switch and the time delay between the switch are minimized, and the load is balanced, thereby ensuring the overall performance of the SD-WAN.
Description
Technical Field
The invention relates to the field of network deployment, in particular to a multi-controller deployment method and a terminal in an SD-WAN environment.
Background
A Software Defined wide area network (SD-WAN) is a feature of applying the Software Defined network (Software Defined Network, SDN) to manage the wide area network. Compared with the traditional wide area network, the software defined wide area network can effectively reduce network cost and greatly improve network interconnection degree. Software-defined wide area networks have the characteristics of large scope and wide coverage, and existing deployment strategies with single-controller centralized control limit the flexibility and expandability of the network to a great extent, so that multi-controller deployment occurs. How to implement a logically centralized and physically decentralized multi-controller deployment is critical to the multi-controller deployment problem. In the deployment of multiple controllers, it is particularly important how QoS (Quality of Service ) of SDN is guaranteed through selection of reasonable locations. The current mainstream solution is to divide the network into a plurality of disjoint sub-domains, each of which is controlled by a controller or set of controllers. And setting targets on performances such as time delay, load balancing, deployment cost, reliability and the like to optimize the deployment scheme.
In this case, since the SD-WAN control and forwarding are decoupled from each other, all functions of the network are realized by message exchange between the controller and the switch, and thus, delay between the controller and the switch is the most academic focus as an optimization objective. The time delay mainly comprises queuing time delay, transmission time delay, propagation time delay and processing time delay, and when the network is smooth, the queuing time delay can be ignored; the transmission delay is related to the packet and port rate, typically a fixed value; the processing delay is mainly determined by the performance and load of the controller, and the propagation delay is mainly determined by the distance between nodes. Thus, in SD-WAN, delay optimization in controller deployment issues typically only considers propagation delay. The scholars Heller et al (The controller placement problem 42 (4) (2012), 473-478) that were the earliest to address the controller deployment problem mainly used the mean latency and the maximum latency as optimization targets, and adopted greedy algorithms to achieve multi-controller deployment in SDN. Literature (International Conference on Algorithms and Architectures for Parallel Processing,2015, 213-225) proposes a multi-controller deployment algorithm named NCPSO that deploys a combination of node partition and particle swarm algorithms with minimal latency as an optimization objective. WANG et al (IEEE Transactions on Network and Service Management 15 (1) (2017), 344-355) propose an improved k-means algorithm with propagation delay and queuing delay as optimization targets, which iterates from one partition to a progressively increasing number of partitions to determine the controller position in each partition. None of the above algorithms achieves a lower average delay.
There are few related studies that target load balancing as an optimization. In SDN, some controllers may be in an overload or near-load state. The load condition of the controller is proportional to the number of switches associated with the controller and the number of flows that the switches can carry per second. Therefore, it is important to study how to balance the relationship among the controller load, the number of switches, and the number of streams that can be tolerated by the switches per second. Dixit et al (IEEE Symposium on Architectures for Networking and Communications Systems (ANCS 2014), 2014,17-27) propose for the first time a controller deployment mechanism in a dynamic environment, by which an overload controller can distribute its load to other idle controllers, thereby reducing the probability of controller failure due to overload.
In summary, aiming at the time delay optimization target in the deployment stage of the SDN controller, the existing scheme and the achievement are partially improved based on the clustering method. Although the method can solve some problems, the method has the defect that lower average time delay cannot be obtained, and the reason is that the randomness of the initial node leads to the clustering result and the randomness of the deployment position of the controller, and the randomness of the deployment position of the controller leads to the increase of the average time delay. Secondly, most of the existing schemes require the division number of subfields to be manually determined. In addition, under the multi-objective optimized controller deployment scheme, few documents consider the controller and switch time delay, the inter-controller time delay and the multi-controller deployment scheme under the load balancing multi-index.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: a multi-controller deployment method and a terminal in an SD-WAN environment are provided, and low-average-time-delay multi-controller deployment is realized.
In order to solve the technical problems, the invention adopts a technical scheme that:
a multi-controller deployment method in an SD-WAN environment comprises the following steps:
s1, acquiring a topological structure of an SD-WAN (secure digital-wide area network), a first time delay between a switch and a controller in the SD-WAN and a second time delay between controllers in the SD-WAN;
s2, dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure;
s3, acquiring a third time delay between the switch and the controller in each sub-domain and a preset load balancing constraint;
s4, determining the position of the controller in each sub-domain which minimizes the first time delay, the second time delay and the third time delay according to the load balancing constraint.
In order to solve the technical problems, the invention adopts another technical scheme that:
a multi-controller deployment terminal in an SD-WAN environment, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
s1, acquiring a topological structure of an SD-WAN (secure digital-wide area network), a first time delay between a switch and a controller in the SD-WAN and a second time delay between controllers in the SD-WAN;
s2, dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure;
s3, acquiring a third time delay between the switch and the controller in each sub-domain and a preset load balancing constraint;
s4, determining the position of the controller in each sub-domain which minimizes the first time delay, the second time delay and the third time delay according to the load balancing constraint.
The invention has the beneficial effects that: aiming at the problem of SD-WAN multi-controller deployment, taking the minimum propagation delay between controllers as an optimization target, and carrying out subdomain division on a software defined network; and meanwhile, the deployment position of the controller in each sub-domain is determined by taking the time delay between the minimum controller and the switch as an optimization target on the basis, so that the time delay between the controller and the switch and the time delay between the switch can be minimized under the condition of ensuring the load balance of multiple controllers as much as possible, and the overall performance of the SD-WAN is ensured.
Drawings
FIG. 1 is a flow chart of steps of a method for deploying multiple controllers in an SD-WAN environment according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a multi-controller deployment terminal in an SD-WAN environment according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating steps of a multi-controller deployment method applied to an actual scenario in an SD-WAN environment according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a single sub-domain of an embodiment of the present invention;
FIG. 5 is a schematic diagram of a controller and a switch in two subzones according to an embodiment of the present invention;
description of the reference numerals:
1. a multi-controller deployment terminal in an SD-WAN environment; 2. a processor; 3. a memory.
Detailed Description
In order to describe the technical contents, the achieved objects and effects of the present invention in detail, the following description will be made with reference to the embodiments in conjunction with the accompanying drawings.
Referring to fig. 1, a multi-controller deployment method in SD-WAN environment includes the steps of:
s1, acquiring a topological structure of an SD-WAN (secure digital-wide area network), a first time delay between a switch and a controller in the SD-WAN and a second time delay between controllers in the SD-WAN;
s2, dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure;
s3, acquiring a third time delay between the switch and the controller in each sub-domain and a preset load balancing constraint;
s4, determining the position of the controller in each sub-domain which minimizes the first time delay, the second time delay and the third time delay according to the load balancing constraint.
From the above description, the beneficial effects of the invention are as follows: aiming at the problem of SD-WAN multi-controller deployment, taking the minimum propagation delay between controllers as an optimization target, and carrying out subdomain division on a software defined network; and meanwhile, the deployment position of the controller in each sub-domain is determined by taking the time delay between the minimum controller and the switch as an optimization target on the basis, so that the time delay between the controller and the switch and the time delay between the switch can be minimized under the condition of ensuring the load balance of multiple controllers as much as possible, and the overall performance of the SD-WAN is ensured.
Further, the topology structure in S1 includes a node location of a node in the SD-WAN network, where the node includes a switch and a controller;
the step S2 is specifically as follows:
s21, taking each node in the SD-WAN network as an independent alternative subdomain, and calculating a first average bit value in the SD-WAN network at the moment;
s22, sampling all nodes in the topological structure to obtain a node sequence;
s23, acquiring a first node and a second node adjacent to the first node according to the node sequence, and acquiring a first alternative subdomain corresponding to the first node and a second alternative subdomain corresponding to the second node;
s24, merging the first node into the second alternative subdomain to obtain a third alternative subdomain, calculating a second average bit in the SD-WAN network at the moment, comparing the first average bit value with the second average bit value, if the first average bit value is smaller than the second average bit value, restoring the first node into the first alternative subdomain and returning to S22, otherwise, reserving the third alternative subdomain and returning to S22.
From the above description, it can be seen that the average bits are calculated, and whether to incorporate the nodes into the sub-domains to form a new sub-domain is determined by comparing the average bits, so as to automatically divide the SD-WAN network sub-domains and ensure the data transmission rate in the sub-domains.
Further, the first delay and the third delay between the switch and the controller include an average delay and a maximum delay, respectively:
the second delay between the controllers includes an average delay and a maximum delay:
wherein ,LS-C-avg Representing the average delay between the switch and the controller, L S-C-max Representing the maximum delay between the switch and the controller, L C-C-avg Representing the average delay between controllers, L C-C-max Represents the maximum delay between controllers, d (s, c m ) Represents the distance between the controller and the mth exchanger, d (c) i ,c j ) Represents the distance between the ith controller and the jth controller, v= { V 1 ,v 2 ,...,v n The number k represents the number of controllers.
From the above description, the time delay is divided into an average time delay and a maximum time delay, and the corresponding time delay is selected for calculation according to the requirement of the actual scene, so that the accuracy of calculating data is improved.
Further, the preset load balancing constraint is:
wherein a (n) represents the direction of the switch to the controller c i The maximum request stream sent, a (u), represents the switch to controller c j The maximum request flow sent, C, represents the set of controllers made up of all controllers in the SD-WAN network, epsilon represents the load balancing constraint factor.
From the above description, the load balancing constraint averages the loads in each sub-domain in a single SD-WAN network, so as to avoid the situation that the controller in the single sub-domain is too high while the rest sub-domains are idle, and improve the robustness of the SD-WAN network.
Further, before S2, the method further includes: acquiring load constraints on the subdomains:
where a (n) represents the maximum request flow sent by the switch to the controller, A (c) m ) Representing the maximum load capacity of the controller m.
The step S2 is specifically as follows: and dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure and the load constraint.
From the above description, the load constraint of a single subdomain ensures the stable and normal operation of the controller, and avoids the overload condition of the controller to a certain extent.
Further, the S2 specifically is:
and dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay by adopting an InfoMap algorithm according to the topological structure.
From the above description, it can be seen that the benefit of the SD-WAN network formed by the generated subdomains is ensured while the automatic generation of the subdomains is realized by dividing the SD-WAN network by adopting the InfoMap algorithm.
Further, calculating the average bit value L (K) in the SD-WAN network is specifically:
wherein ,
to enter the entropy of information for the occurrence of a subdomain event,to jump the event or leave the probability of the event of the sub-domain within the sub-domain,/for example>Information entropy representing a jump event of a worker within a sub-domain or an exit event from the sub-domain;
probability of entering subdomain wherein ,vx V y Representing nodes in two different sub-domains.
From the above description, it can be seen that the information entropy is used as one of the reference factors for dividing the subdomains, so as to ensure the overall performance of the finally obtained SD-WAN network.
Further, the step S2 further includes:
setting time delay constraint min (alpha L) s-c-avg +βL c-c-avg )；
Wherein, alpha and beta are weight coefficients, alpha+beta=1, alpha and beta are [0,1].
From the above description, it can be seen that the time delay constraint is set, the time delay is used as an optimization target, the operation efficiency of the SD-WAN network is ensured, the average time delay is used as a reference value, the optimization of the time delay and the system overhead are considered, and a scheme with high cost performance can be obtained.
Further, the step S2 further includes:
setting time delay constraint min (alpha' L) s-c-max +β'L c-c-max )；
Wherein α ', β' are weight coefficients, α '+β' =1, α ', β' ∈ [0,1].
From the above description, the time delay constraint is set, the time delay is used as an optimization target, the operation efficiency of the SD-WAN network is ensured, and the maximum time delay is used as a reference value, so that high efficiency and high speed are realized. Low latency SD-WAN network deployment.
Referring to fig. 2, a multi-controller deployment terminal in SD-WAN environment includes a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor implements the following steps when executing the computer program:
s1, acquiring a topological structure of an SD-WAN (secure digital-wide area network), a first time delay between a switch and a controller in the SD-WAN and a second time delay between controllers in the SD-WAN;
s2, dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure;
s3, acquiring a third time delay between the switch and the controller in each sub-domain and a preset load balancing constraint;
s4, determining the position of the controller in each sub-domain which minimizes the first time delay, the second time delay and the third time delay according to the load balancing constraint.
The invention has the beneficial effects that: aiming at the problem of SD-WAN multi-controller deployment, taking the minimum propagation delay between controllers as an optimization target, and carrying out subdomain division on a software defined network; and meanwhile, the deployment position of the controller in each sub-domain is determined by taking the time delay between the minimum controller and the switch as an optimization target on the basis, so that the time delay between the controller and the switch and the time delay between the switch can be minimized under the condition of ensuring the load balance of multiple controllers as much as possible, and the overall performance of the SD-WAN is ensured.
Referring to fig. 1, a first embodiment of the present invention is as follows:
specifically comprises:
s1, acquiring a topological structure of an SD-WAN (secure digital-wide area network), a first time delay between a switch and a controller in the SD-WAN and a second time delay between controllers in the SD-WAN;
the topology structure comprises node positions of nodes in the SD-WAN network, wherein the nodes comprise switches and controllers;
s2, dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure;
in this embodiment, S2 further includes, before: acquiring load constraints on the subdomains:
where a (n) represents the maximum request flow sent by the switch to the controller, A (c) m ) Representing the maximum load capacity of the controller m;
s2 is specifically:
dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the second time delay by adopting an InfoMap algorithm according to the topological structure and the load constraint;
the InfoMap algorithm specifically comprises the following steps:
in the undirected weight network graph,representing node v x Pointing to node v y The weight of (which may be distance, link occupancy, etc.); v (V) x→ Representation and v x Set of all nodes connected, then node v x To v y The transition probability of (2) isNode v x Arrival probability of->The essence of the InfoMap algorithm is that the arrival probability of each node is updated by continuous iterative calculation until convergence;
in this embodiment, S2 is specifically:
s21, taking each node in the SD-WAN network as an independent alternative subdomain, and calculating a first average bit value in the SD-WAN network at the moment;
s22, sampling all nodes in the topological structure to obtain a node sequence;
s23, acquiring a first node and a second node adjacent to the first node according to the node sequence, and acquiring a first alternative subdomain corresponding to the first node and a second alternative subdomain corresponding to the second node;
s24, merging the first node into the second alternative subdomain to obtain a third alternative subdomain, calculating a second average bit in the SD-WAN network at the moment, comparing the first average bit value with the second average bit value, if the first average bit value is smaller than the second average bit value, restoring the first node into the first alternative subdomain and returning to S22, otherwise, reserving the third alternative subdomain and returning to S22;
wherein, calculating the average bit value L (K) in the SD-WAN network is specifically:
wherein ,
to enter the entropy of information for the occurrence of a subdomain event,to jump the event or leave the probability of the event of the sub-domain within the sub-domain,/for example>Information entropy representing a jump event of a worker within a sub-domain or an exit event from the sub-domain; gamma (c) m ) Controller c located in subfield gamma indicating the occurrence of the above-mentioned event m ；
Probability of entering a subdomain event
Probability of leaving subdomain event
wherein ,vx V y Representing nodes in two different sub-domains;
s3, acquiring a third time delay between the switch and the controller in each sub-domain and a preset load balancing constraint;
wherein the load balancing constraint is:
wherein a (n) represents the direction of the switch to the controller c i The maximum request stream sent, a (u), represents the switch to controller c j The maximum request flow is sent, C represents a controller set formed by all controllers in the SD-WAN network, epsilon represents a load balancing constraint factor;
s4, determining the position of a controller in each sub-domain which minimizes the first time delay, the second time delay and the third time delay according to the load balancing constraint;
in this embodiment, the delay between the switch and the controller (including the first delay and the third delay) includes an average delay and a maximum delay:
the second delay between the controllers includes an average delay and a maximum delay:
wherein ,LS-C-avg Representing the average delay between the switch and the controller, L S-C-max Representing the maximum delay between the switch and the controller, L C-C-avg Representing the average delay between controllers, L C-C-max Represents the maximum delay between controllers, d (s, c m ) Represents the distance between the controller and the mth exchanger, d (c) i ,c j ) Represents the distance between the ith controller and the jth controller, v= { V 1 ,v 2 ,...,v n ' represents a set of SDN switchesTogether, k represents the number of controllers;
s2, further comprises the following steps:
setting time delay constraint min (alpha L) s-c-avg +βL c-c-avg )；
Wherein, alpha and beta are weight coefficients, alpha+beta=1, alpha, beta epsilon [0,1];
time delay constraint min (alpha' L) s-c-max +β'L c-c-max )；
Wherein α ', β' are weight coefficients, α '+β' =1, α ', β' ∈ [0,1].
Referring to fig. 3 to 5, a second embodiment of the present invention is as follows:
the multi-controller deployment method under the SD-WAN environment is applied to the actual scene:
in this embodiment, the community is a subdomain;
referring to fig. 4, a sub-domain consisting of 3 nodes is given as an example, where l represents the time delay between two nodes, and as can be seen from the figure, v is selected 1 When a node is used as a controller position, the minimum total time delay in the community can be expressed as l 12 +l 13 The method comprises the steps of carrying out a first treatment on the surface of the Next, the controller position is traversed, when v 3 When a node is used as a controller position, the minimum total time delay in the community can be expressed as l 13 +l 23 The method comprises the steps of carrying out a first treatment on the surface of the At this time, there is l 12 +l 13 >l 13 +l 23 Description and node v 1 In contrast, node v 3 More suitable for being used as a controller position; consistent with the method, the optimal controller deployment position is determined to be the node v by traversing other nodes 2 ；
Referring to fig. 5, for the case of two subfields, solid nodes represent deployment positions of the controller, hollow nodes represent switch nodes in the topology, and the deployment positions of the controller are respectively node v 2 and v4 . At this time, the shortest time delay between the deployment positions of the controller is l 23 +l 34 According to the formula min (alpha L s-c-avg +βL c-c-avg ) Can calculate the target time delay asThen, change node v 4 Controller position on to node v 5 Calculating the time delay between the transformed controllers as l 25 Compared with the existing time delay, is smaller than the existing minimum time delay 23 +l 34 Calculating a target time delay according to the transformed deployment position of the controller, namelyFinally comparing the target delay sizes of the two deployment positions, and taking the optimal deployment position with a smaller value;
repeating the steps, continuously iterating the controller positions, and exhausting all combinations to possibly calculate the optimal multi-controller deployment position conforming to the objective function.
Referring to fig. 2, a third embodiment of the present invention is as follows:
a multi-controller deployment terminal 1 in an SD-WAN environment, comprising a processor 2, a memory 3 and a computer program stored on the memory 3 and executable on the processor 2, the processor 2 implementing the steps of either embodiment one or embodiment two when executing the computer program.
In summary, the present invention provides a multi-controller deployment method and a terminal in SD-WAN environment, and provides an SDN multi-controller deployment method that considers both delay and balanced controller load between different objects, including division of software defined network sub-domains and determination of controller positions in each sub-domain, where the method of software defined network sub-domain division based on community discovery optimizes the software defined network sub-domain division by taking minimizing propagation delay and balanced load between switches as an optimization objective; and meanwhile, the deployment position of the controller in each subdomain is determined by taking the time delay between the minimum controller and the switch as an optimization target on the basis, so that the time delay between the controller and the switch and the time delay between the switch are minimized under the condition of ensuring the load balance of multiple controllers as much as possible, the overall performance of the SD-WAN is ensured, and the InfoMap algorithm is used in the optimization process, so that the automatic determination and the optimal selection of the subdomain number are realized.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent changes made by the specification and drawings of the present invention, or direct or indirect application in the relevant art, are included in the scope of the present invention.
Claims (9)
1. The multi-controller deployment method in the SD-WAN environment is characterized by comprising the following steps:
s1, acquiring a topological structure of an SD-WAN (secure digital-wide area network), a first time delay between a switch and a controller in the SD-WAN and a second time delay between controllers in the SD-WAN;
s2, dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure;
s3, acquiring a third time delay between the switch and the controller in each sub-domain and a preset load balancing constraint;
s4, determining the position of a controller in each sub-domain which minimizes the first time delay, the second time delay and the third time delay according to the load balancing constraint;
the topological structure in the S1 is characterized by comprising node positions of nodes in an SD-WAN network, wherein the nodes comprise a switch and a controller;
the step S2 is specifically as follows:
s21, taking each node in the SD-WAN network as an independent alternative subdomain, and calculating a first average bit value in the SD-WAN network at the moment;
s22, sampling all nodes in the topological structure to obtain a node sequence;
s23, acquiring a first node and a second node adjacent to the first node according to the node sequence, and acquiring a first alternative subdomain corresponding to the first node and a second alternative subdomain corresponding to the second node;
s24, merging the first node into the second alternative subdomain to obtain a third alternative subdomain, calculating a second average bit in the SD-WAN network at the moment, comparing the first average bit value with the second average bit value, if the first average bit value is smaller than the second average bit value, restoring the first node into the first alternative subdomain and returning to S22, otherwise, reserving the third alternative subdomain and returning to S22.
2. The method for deploying multiple controllers in an SD-WAN environment according to claim 1, wherein the first delay and the third delay between the switch and the controller comprise an average delay and a maximum delay, respectively:
the second delay between the controllers includes an average delay and a maximum delay:
wherein ,LS-C-avg Representing the average delay between the switch and the controller, L S-C-max Representing the maximum delay between the switch and the controller, L C-C-avg Representing the average delay between controllers, L C-C-max Represents the maximum delay between controllers, d (s, c m ) Represents the distance between the switch and the mth controller, d (c i ,c j ) Represents the distance between the ith controller and the jth controller, v= { V 1 ,v 2 ,...,v n ' represents a set of SDN switchesTogether, k represents the number of controllers; gamma (c) m ) Representing a controller c located in a sub-field gamma m 。
3. The method for deploying multiple controllers in an SD-WAN environment according to claim 1, wherein the preset load balancing constraint is:
wherein a (n) represents the direction of the switch to the controller c i The maximum request stream sent, a (u), represents the switch to controller c j The maximum request flow sent, C, represents the set of controllers made up of all controllers in the SD-WAN network, epsilon represents the load balancing constraint factor.
4. The method for multi-controller deployment in SD-WAN environment according to claim 1, wherein said S2 further comprises before: acquiring load constraints on the subdomains:
where a (n) represents the maximum request flow sent by the switch to the controller, A (c) m ) Representing the maximum load capacity of the controller m;
the step S2 is specifically as follows: and dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay according to the topological structure and the load constraint.
5. The method for deploying multiple controllers in an SD-WAN environment according to claim 1, wherein S2 is specifically:
and dividing the SD-WAN network into a first preset number of subfields with the minimum first time delay and the minimum second time delay by adopting an InfoMap algorithm according to the topological structure.
6. The method for deploying multiple controllers in an SD-WAN environment according to claim 1, wherein calculating the average bit value L (K) in the SD-WAN network is specifically:
wherein ,
to enter the entropy of information for the occurrence of a subdomain event,to jump the event or leave the probability of the event of the sub-domain within the sub-domain,/for example>Information entropy representing jump events or leave sub-domain events within a sub-domain;
probability of entering subdomain
Probability of leaving subdomain event
wherein ,vx V y Representing nodes in two different sub-domains.
7. The method for multi-controller deployment in an SD-WAN environment according to claim 2, wherein said S2 further comprises:
setting time delay constraint min (alpha L) s-c-avg +βL c-c-avg )；
Wherein, alpha and beta are weight coefficients, alpha+beta=1, alpha and beta are [0,1].
8. The method for multi-controller deployment in an SD-WAN environment according to claim 2, wherein said S2 further comprises:
setting time delay constraint min (alpha' L) s-c-max +β'L c-c-max )；
Wherein α ', β' are weight coefficients, α '+β' =1, α ', β' ∈ [0,1].
9. A multi-controller deployment terminal in an SD-WAN environment, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements a multi-controller deployment method in an SD-WAN environment according to any of claims 1-8 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011322406.3A CN112702186B (en) | 2020-11-23 | 2020-11-23 | Multi-controller deployment method and terminal in SD-WAN environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011322406.3A CN112702186B (en) | 2020-11-23 | 2020-11-23 | Multi-controller deployment method and terminal in SD-WAN environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112702186A CN112702186A (en) | 2021-04-23 |
| CN112702186B true CN112702186B (en) | 2023-10-27 |
Family
ID=75506015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011322406.3A Active CN112702186B (en) | 2020-11-23 | 2020-11-23 | Multi-controller deployment method and terminal in SD-WAN environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112702186B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113452552B (en) * | 2021-06-16 | 2022-10-21 | 西安电子科技大学 | Information entropy perception-based super-multi-target controller placement method |
| CN114679730A (en) * | 2022-03-15 | 2022-06-28 | 国网智能电网研究院有限公司 | Software defined network switching node layout method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101772041A (en) * | 2010-01-26 | 2010-07-07 | 福建师范大学 | Wireless multi-hop network performance optimization method based on cross-layer Bargaining game model |
| CN107204874A (en) * | 2017-05-09 | 2017-09-26 | 天津大学 | Ensure the minimum SDN multi-controller dispositions method of time delay |
| CN107276662A (en) * | 2017-07-27 | 2017-10-20 | 大连大学 | A kind of software definition Information Network multi-controller dynamic deployment method |
| CN108449265A (en) * | 2018-02-05 | 2018-08-24 | 西安交通大学 | Path calculation method and device in a kind of SD-WAN environment |
| CN109286528A (en) * | 2018-10-16 | 2019-01-29 | 四川长虹电器股份有限公司 | A kind of SDN network multi-controller dispositions method based on time delay |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8832264B2 (en) * | 2012-03-01 | 2014-09-09 | Justin Pauley | Network appliance for monitoring network requests for multimedia content |
| US10225187B2 (en) * | 2017-03-22 | 2019-03-05 | Cisco Technology, Inc. | System and method for providing a bit indexed service chain |
| US20200077355A1 (en) * | 2018-08-29 | 2020-03-05 | Commscope Technologies Llc | Clock synchronization in a centralized radio access network having multiple controllers |
-
2020
- 2020-11-23 CN CN202011322406.3A patent/CN112702186B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101772041A (en) * | 2010-01-26 | 2010-07-07 | 福建师范大学 | Wireless multi-hop network performance optimization method based on cross-layer Bargaining game model |
| CN107204874A (en) * | 2017-05-09 | 2017-09-26 | 天津大学 | Ensure the minimum SDN multi-controller dispositions method of time delay |
| CN107276662A (en) * | 2017-07-27 | 2017-10-20 | 大连大学 | A kind of software definition Information Network multi-controller dynamic deployment method |
| CN108449265A (en) * | 2018-02-05 | 2018-08-24 | 西安交通大学 | Path calculation method and device in a kind of SD-WAN environment |
| CN109286528A (en) * | 2018-10-16 | 2019-01-29 | 四川长虹电器股份有限公司 | A kind of SDN network multi-controller dispositions method based on time delay |
Non-Patent Citations (5)
| Title |
|---|
| Ahtisham Ali Ansari等.An SD-WAN Controller for Delay Jitter Minimization in Coded Multi-access Systems.《2019 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS)》.2020,全文. * |
| David Erickson等.Using Network Knowledge to Improve Workload Performance in Virtualized Data Centers.《2014 IEEE International Conference on Cloud Engineering》.2014,全文. * |
| Xuancheng Guo等.Deep-Reinforcement-Learning-Based QoS-Aware Secure Routing for SDN-IoT.《IEEE INTERNET OF THINGS JOURNAL》.2020,第7卷(第7期),全文. * |
| 周文辉 ; 刘永伟 ; .EBoD:打造开放SD-WAN网络服务平台.中兴通讯技术.2019,(第02期),全文. * |
| 郭烜成等.软件定义广域网中控制器部署与交换机动态迁移策略.《计算机应用》.2019,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112702186A (en) | 2021-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112702186B (en) | Multi-controller deployment method and terminal in SD-WAN environment | |
| Pang et al. | SDN-based data center networking with collaboration of multipath TCP and segment routing | |
| US20010029543A1 (en) | Link state routing techniques | |
| US20030236887A1 (en) | Cluster bandwidth management algorithms | |
| CN112350949B (en) | Rerouting congestion control method and system based on flow scheduling in software defined network | |
| CN108684046B (en) | Random learning-based access network service function chain deployment method | |
| JPH0766835A (en) | Communication network and method for selection of route in said network | |
| CN112543151B (en) | SDN controller deployment method and device, electronic equipment and storage medium | |
| JP4611319B2 (en) | Network architecture | |
| CN116708280B (en) | Data center network multipath transmission method based on disorder tolerance | |
| WO2009055777A1 (en) | Methods and systems for providing efficient provisioning of data flows | |
| CN1787475B (en) | Access network system and method of moving internal network relay devices | |
| WO2015069761A1 (en) | System and method for delay management for traffic engineering | |
| CN113162863B (en) | Fair multipath routing method based on wheel disc algorithm | |
| CN112165371B (en) | Method, medium, equipment and device for BIER automatic configuration management (BSL) | |
| Yeh et al. | VIP: Joint traffic engineering and caching in Named Data Networks | |
| Aly | A novel controller placement using Petri-nets for SDNs | |
| CN109921991B (en) | SDN controller deployment method based on Dinkelbach | |
| CN109450809B (en) | Data center scheduling system and method | |
| CN112511453B (en) | SDN controller deployment method, device and storage medium | |
| Ma | Adaptive distributed load balancing routing mechanism for LEO satellite IP networks | |
| Tsukuda et al. | Reducing Inconsistency between Software-Defined Networking Controllers | |
| CN112311584B (en) | Deployment method of software defined network controller | |
| CN114793215A (en) | Distributed system based on load balancing mode asynchronous agent | |
| CN115174489B (en) | Data center network flow scheduling method for minimizing network congestion and Qos guarantee |
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 |