WO2020252895A1 - 混合软件自定义网络的部署方法、装置、设备及存储介质 - Google Patents

混合软件自定义网络的部署方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2020252895A1
WO2020252895A1 PCT/CN2019/102469 CN2019102469W WO2020252895A1 WO 2020252895 A1 WO2020252895 A1 WO 2020252895A1 CN 2019102469 W CN2019102469 W CN 2019102469W WO 2020252895 A1 WO2020252895 A1 WO 2020252895A1
Authority
WO
WIPO (PCT)
Prior art keywords
sdn
vlan
network
hybrid
node
Prior art date
Application number
PCT/CN2019/102469
Other languages
English (en)
French (fr)
Inventor
陈闽川
马骏
王少军
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2020252895A1 publication Critical patent/WO2020252895A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H04L12/4641—Virtual LANs, VLANs, e.g. virtual private networks [VPN]
    • 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
    • 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
    • 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
    • H04L41/0893—Assignment of logical groups to network elements
    • 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

Definitions

  • This application relates to the field of distributed deployment, and in particular to a deployment method, device, equipment, and storage medium of a hybrid software custom network.
  • SDN Software-defined network
  • OpenFlow OpenFlow
  • hybrid SDN SDN Hybrid networking with traditional networks, called hybrid SDN, can implement the transition of the network architecture in an economical and iterative manner without business interruption due to complete replacement.
  • the inventor realizes that in the current hybrid SDN, the deployment strategy stays at static deployment, and only deploys for current resources, without considering that the transition from a traditional network to a software-defined network is a gradual process. Moreover, the communication method in the hybrid software-defined network is too single, and the advantages of the software-defined network structure are not fully utilized.
  • This application provides a deployment method, device, equipment, and storage medium for a hybrid software custom network, which are used to make the deployment process smoother and more effective, and the coordination between network control rate and link load balance is more reasonable, and network work is improved effectiveness.
  • the first aspect of the embodiments of the present application provides a method for deploying a hybrid software-defined network, including: obtaining node information of a target network; judging whether the target network is a hybrid software-defined network SDN according to the node information; If the target network is not the hybrid SDN, the initial deployment strategy of the hybrid SDN network is executed.
  • the initial deployment strategy of the hybrid SDN network is used to calculate the initial SDN node deployment location; according to the initial SDN node deployment location and preset reinforcement learning algorithm Create a virtual local network VLAN group.
  • the second aspect of the embodiments of the present application provides a device for deploying a hybrid software customized network, including: an acquiring unit, configured to acquire node information of a target network; and a determining unit, configured to determine the target network based on the node information Whether it is a hybrid software-defined network SDN; the first execution unit, if the target network is not the hybrid SDN, it is used to execute the hybrid SDN network initial deployment strategy, and the hybrid SDN network initial deployment strategy is used to calculate the initial SDN node Deployment location; the first establishment unit, according to the initial SDN node deployment location and preset reinforcement learning algorithm to establish a virtual local network VLAN group.
  • the third aspect of the embodiments of the present application provides a deployment device for a hybrid software custom network, which includes a memory, a processor, and a computer program stored in the memory and running on the processor.
  • the processor When the computer program is executed, the deployment method of the hybrid software self-defined network is realized.
  • the fourth aspect of the embodiments of the present application provides a computer-readable storage medium that stores instructions in the computer-readable storage medium.
  • the computer executes the above hybrid software custom network. Steps of the deployment method.
  • FIG. 1 is a schematic diagram of an embodiment of a method for deploying a hybrid software customized network in an embodiment of this application;
  • FIG. 2 is a schematic diagram of an application scenario for executing an initial deployment strategy of a hybrid SDN network in an embodiment of the application;
  • FIG. 3 is a schematic diagram of an application scenario of the VLAN combination merge process in an embodiment of the application
  • FIG. 4 is a schematic diagram of another application scenario for implementing a transition deployment strategy of a hybrid SDN network in an embodiment of the application;
  • FIG. 5 is a schematic diagram of a dynamic deployment cost change value of a hybrid SDN network in an embodiment of this application
  • FIG. 6 is a schematic diagram of an embodiment of a deployment device for a hybrid software customized network in an embodiment of the application
  • FIG. 7 is a schematic diagram of another embodiment of a deployment device for a hybrid software customized network in an embodiment of this application.
  • FIG. 8 is a schematic diagram of an embodiment of a deployment device for a hybrid software customized network in an embodiment of this application.
  • This application provides a deployment method, device, equipment, and storage medium for a hybrid software custom network, which are used to make the deployment process smoother and more effective, and the coordination between network control rate and link load balance is more reasonable, and network work is improved effectiveness.
  • FIG. 1 a flowchart of a method for deploying a hybrid software customized network provided by an embodiment of the present application, which specifically includes:
  • the server obtains the node information of the target network.
  • the terminal can obtain the name identification, address identification, and configuration identifier of the target network node, the path to the target network node, etc., and other node information that needs to be used ,
  • the specifics are not limited here.
  • the execution subject may be a server or a terminal, as long as the deployment of a hybrid software custom network can be implemented, and this application uses a server as an example for description.
  • the server determines whether the target network is a hybrid software-defined network SDN according to the node information. Specifically, the server determines whether the strategy currently executed by the target network is a hybrid SDN strategy according to the node information.
  • the hybrid SDN strategy is a rule that the nodes of the target network can execute, that is, the nodes of the target network manage the network according to certain rules. Among them, the strategy is divided into two situations, one is to upgrade the traditional network to a hybrid SDN network, that is, the initial deployment strategy of the hybrid SDN network; the other is to upgrade the hybrid SDN network by adding SDN nodes (switches), that is, hybrid SDN Network transition deployment strategy. In the process of implementing the initial deployment strategy of the hybrid SDN network, only hardware resources need to be considered.
  • the target network is not a hybrid SDN
  • execute the initial deployment strategy of the hybrid SDN network and the initial deployment strategy of the hybrid SDN network is used to calculate the initial deployment location of the SDN node.
  • the server implements the initial deployment strategy of the hybrid SDN network, and the server upgrades the target network, that is, upgrades from a traditional network to a hybrid SDN network.
  • the initial deployment strategy of the hybrid SDN network is used to calculate the initial deployment of SDN nodes position. Specifically, multiple target deployment nodes are selected according to the initial deployment strategy of the hybrid SDN network, the switches in the multiple target deployment nodes are replaced with SDN switches, and VLAN grouping is performed.
  • FIG 2 is a schematic diagram of the process of upgrading a traditional network to a hybrid SDN network (that is, implementing the initial deployment strategy of the hybrid SDN network).
  • the target network includes traditional switch A, traditional switch B, traditional switch C, traditional switch D, Traditional switch E, traditional switch F, traditional switch G, traditional switch H, traditional switch I, and traditional switch J; calculated according to the preset reinforcement learning algorithm, traditional switches D and E need to be replaced with SDN switches D and E. Then configure some of the links through the hybrid SDN network communication module to configure VLAN groups.
  • the server establishes a virtual local network VLAN group according to the initial deployment location of the SDN node and the preset reinforcement learning algorithm. For example, first use a preset reinforcement learning algorithm to select the optimal deployment node. After multiple deployment nodes are selected, VLAN group division is performed to obtain a collection of VLAN groups. The specific process of VLAN group division is as follows: first perform SDN node VLAN grouping on SDN switch nodes to obtain the current stage VLAN grouping set and isolated switch set, and then perform isolated node VLAN grouping, VLAN combination and VLAN group decomposition in turn.
  • the VLAN grouping process of SDN nodes is as follows: 1. First, the single switch vi in the SDN switch set V S in the hybrid SDN network is sorted from large to small according to degree, because the deployment node has already considered the link problem, and only switches will be considered here. (That is, the degree of the node), as the priority of VLAN grouping. 2. Set two variables Valready and Valone, Valready represents the switch that has been VLAN grouped, Valone represents the remaining switches without VLAN grouping, Valready and Valone are initialized to the empty set and all switches V-Vs. 3. Traverse the set V of all switches in the hybrid SDN network for the SDN switch nodes that have been sorted.
  • Valready is an empty set, add the node vi directly to the set, and continue to traverse the next node. If Valready has only one node, calculate the shortest path of the only node in Valready through vi, establish a VLAN group, add the information to the set VL of VLANs in the hybrid SDN network, and remove the nodes on the new VLAN group from the Valone Remove. If there are at least two nodes in Valready, calculate the two nodes closest to vi in Valready and establish two new VLAN groups, and add the two new VLAN groups established to the set VL of VLANs in the hybrid SDN network.
  • the VLAN grouping process of isolated nodes is as follows: rank the isolated nodes in the isolated node set with degree as the priority to obtain the set of nodes that have established VLAN group communication among the isolated nodes, and then group the isolated nodes that have established VLAN group communication. Nodes outside the node set are traversed, a new VLAN group is established according to the number of public paths in the link path set between the last two nodes, and the new VLAN group is added to the current VLAN grouping set to obtain the first Transition VLAN grouping collection. Specifically: 1. Sort the input isolated nodes with degree as the priority, and then initialize a Valready set, which represents the nodes that have established VLAN group communication among the isolated nodes. 2. First traverse the isolated node set Valone.
  • the node that has established VLAN group communication is skipped, if not, calculate the two SDN nodes closest to vi in the Vs set, and record the two link path sets E1 and E2 .
  • these two link path sets E1 and E2 if there is a common path, then only the link path set of the closest SDN node to vi is left, a new VLAN group is created from the path set, and its information is added to the VL Group, and finally add the switches on the new VLAN group to the Valready collection. If the two link paths do not have a common path, then take the union of the two link paths, create a new VLAN group, add its information to the VL group, and finally add the switches on the new VLAN group to Valready set.
  • the VLAN combination and merging process is: determining the link with the most VLAN groups in the first transitional VLAN grouping set, and selecting the VLAN groups on any two links to merge to obtain the second transitional VLAN grouping set. Specifically: 1. Find the link with the most VLAN groups among the links in the hybrid SDN network. In the left picture of Figure 3, there are three links e B, D , e D, G, and e H, J. The VLAN groups on the first two links are combined.
  • links e B, D are taken as an example
  • VLmine is the VLAN2 group at this time
  • VLmaxe is the VLAN1 group.
  • the VLAN group decomposition process is: according to the second transitional VLAN grouping set, determine the VLAN group set that needs to be reorganized, the traditional switch node set that needs to be reorganized, and the traditional switch set that has completed the reorganization of the VLAN, and then the VLAN group set that needs to be reorganized and all The traditional switch node set that needs to be reorganized is assigned, and the VLAN set that needs to be reorganized is deleted from the VLAN group set. Finally, the newly added SDN switch and the remaining traditional switches are divided into VLAN groups to obtain the reorganized VLAN group set. Specifically: 1.
  • VLregroup Vnewc
  • Valready which respectively represent the set of VLAN groups that need to be reorganized, the set of traditional switch nodes that need to be reorganized, and the set of traditional switches that have completed VLAN reorganization.
  • VLregroup Assign values to the VLregroup and Vnewc variables, and remove the VLAN set to be reorganized from the VLAN group set VL.
  • the new SDN switch is divided into VLAN groups, and the division method is the same as the VLAN grouping process of SDN nodes.
  • the remaining traditional switches are divided into VLAN groups in the same way as the VLAN grouping process of isolated nodes. 5.
  • the entire network can be regarded as a mesh structure
  • routers and servers are regarded as nodes of the mesh structure
  • the degree refers to the number of links connected by the router nodes.
  • the embodiment of this application implements the initial deployment strategy of the hybrid SDN network for the traditional network according to the actual situation of the network, so that the deployment process is smoother and more effective, and the coordination between network control rate and link load balance is more reasonable; at the same time, virtual local network groups are adopted.
  • the method of division divides the entire hybrid software custom network into different local virtual network groups, and each network node needs to communicate through the local virtual network group, which improves network efficiency.
  • the method further includes: if the target network is all For the hybrid SDN, the hybrid SDN network transition deployment strategy is executed, and the hybrid SDN network transition deployment strategy is used to calculate the deployment position of the newly added SDN node.
  • the server implements a hybrid SDN network transition deployment strategy, and the server upgrades the target network, that is, adding SDN nodes (switches) to the hybrid SDN network for network upgrades to enhance network flow control and network fault tolerance Ability etc.
  • the hybrid SDN network transition deployment strategy is used to calculate the deployment location of newly added SDN nodes (switches).
  • Figure 4 is a schematic diagram of the process of adding SDN nodes to a hybrid SDN network (ie, implementing a hybrid SDN network transition deployment strategy).
  • the target network includes traditional switch A, traditional switch B, traditional switch C, SDN switch D, SDN Switch E, traditional switch F, traditional switch G, traditional switch H, traditional switch I, traditional switch J; calculated according to the preset reinforcement learning algorithm, the traditional switch H needs to be replaced with the SDN switch H.
  • the specific process is similar to that in Figure 2, and will not be repeated here.
  • the embodiment of the application implements a hybrid SDN network transition deployment strategy for the hybrid SDN according to the actual situation of the network, so that the deployment process is smoother and more effective, and the coordination between network control rate and link load balance is more reasonable.
  • the method further includes: establishing a newly-added virtual local network VLAN group according to the deployment position of the newly-added SDN node, and the newly-added VLAN group includes VLAN group decomposition and VLAN combination.
  • the server creates a new virtual local network VLAN group according to the deployment location of the new SDN node, and the new VLAN group includes VLAN group decomposition and VLAN combination. For example, as shown in Figure 4, after determining the need to add SDN switch H, reconfigure the VLAN group through the hybrid SDN network communication module.
  • the entire hybrid SDN network can make full use of the functions of the SDN network's custom forwarding mode and traffic statistics, so that more traffic is controlled by the SDN controller.
  • the deployment position and preset of the initial SDN node Establishing a virtual local network VLAN group by the reinforcement learning algorithm includes: performing VLAN grouping of SDN nodes according to the initial SDN node deployment position and preset reinforcement learning algorithm to obtain the current stage VLAN grouping set and isolated node set; The isolated nodes in the set are sorted by priority to obtain a set of nodes that have established VLAN group communication among the isolated nodes, and then traverse the nodes outside the set of nodes that have established VLAN group communication among the isolated nodes.
  • the number of public paths in the link path set between two nodes establishes a new VLAN group, and adds the new VLAN group to the current VLAN grouping set to obtain a first transitional VLAN grouping set; determining the first transitional VLAN
  • the link with the most VLAN groups in the grouping set is selected to merge the VLAN groups on any two links to obtain the second transitional VLAN grouping set; according to the second transitional VLAN grouping set, the VLAN group set that needs to be reorganized is determined, and the grouping needs to be reorganized.
  • the set of traditional switch nodes and the set of traditional switches that have completed the reorganization of VLANs assign values to the set of VLAN groups that need to be reorganized and the set of traditional switch nodes that need to be reorganized, and remove the VLAN sets that need to be reorganized from the VLAN group set Delete, and finally divide the newly added SDN switch and the remaining traditional switches into VLAN groups to obtain the reorganized VLAN group set.
  • the SDN switch is dynamically deployed by using the reinforcement learning algorithm to solve the problem of smooth transition from the entire network to the SDN network, and to meet the requirement of gradually replacing the old switch in a traditional network or a hybrid SDN network.
  • the deployment position and preset of the initial SDN node performs VLAN grouping of SDN nodes to obtain the current VLAN grouping set and isolated node set, including: according to the preset reinforcement learning algorithm, the single switch vi in the SDN switch set Vs in the hybrid SDN network is processed from large to large Small order; set two variables Valready and Valone, the Valready represents the set of nodes that have been VLAN grouped, the Valone represents the remaining node set without VLAN grouping, the Valready is initialized to an empty set, and the Valone is initialized to V- Vs, the V represents all nodes; traverse the SDN nodes that have been prioritized; if the Valready is an empty set, add the node vi directly to the Valready set, and continue to traverse the next node; If there is only one node in
  • the method further includes: when the hybrid SDN fails At the time, the hybrid SDN is optimized.
  • the server optimizes the hybrid SDN, and the server calculates the dynamic deployment cost change value of the hybrid SDN network through a preset formula.
  • the cost value is calculated by inverting the optimization goal.
  • the minimum cost value is the optimization goal.
  • the maximum network control rate and link load balance are both 50%.
  • the X-axis is the number of iterations, each iteration in the loop, and the Y-axis is the cost value.
  • This experiment loops 120 times, and selects the first, sixtieth, and one hundred and twenty times
  • each loop iteration is 600 times (only 400 times are shown in the figure)
  • the learning rate is 1e-3
  • the reward decay is 0.99
  • the random factor is 0.4
  • the VLAN group is fixed at 10 groups
  • the selected replacement traditional switch is Three
  • the selected network topology is TW Telecom.
  • the optimized value target is defined as:
  • ⁇ and ⁇ are network traffic coefficients and link load balancing coefficients, the sum of ⁇ and ⁇ is 1, ConMax (G) is the maximum control network function, and LinkBal (G) is the load balancing function.
  • the cost value will fluctuate greatly. Later, it can be seen that it basically tends to ease, but because in the loop, the state matrix is constantly being trained. After many trainings, it can be found that after 450 iterations, the data after the first, 60th, and 120th iterations, the random factor factor can be ignored, and the cost value of the 120th iteration Obviously smaller than the other two times, the 60th cycle will occasionally jitter afterwards, because the state matrix is not comprehensive enough, which leads to a random selection of the best points when facing a new state, which may cause the cost value to jitter. Although jitter also occurs at the 120th time, it is relatively stable compared to the 60th time.
  • the hybrid SDN when a failure occurs in the hybrid SDN, the hybrid SDN is optimized to eliminate the impact of the failure on the hybrid SDN and improve work efficiency.
  • the The optimization of the hybrid SDN includes: when the hybrid SDN fails, determining the type of failure of the hybrid SDN, the type of failure includes communication failure within a VLAN group, communication failure between VLAN groups, and communication failure between SDN groups; When the fault type is a communication failure within a VLAN group, implement the fault tolerance strategy within the VLAN group and recalculate the fault-tolerant link; when the fault type is a communication failure between VLAN groups, execute the inter-VLAN fault tolerance strategy and recalculate the fault-tolerant link; When the fault type is a communication failure between SDN groups, the fault tolerance strategy between SDN groups is implemented, and the fault-tolerant link is recalculated.
  • the embodiments of the present application adopt different fault tolerance strategies for different fault scenarios, and improve the recovery performance of hybrid software self-defined network fault tolerance.
  • the deployment method of the hybrid software custom network in the embodiment of this application is described above, and the deployment device of the hybrid software custom network in the embodiment of this application is described below. Please refer to FIG. 6, the hybrid software custom network in the embodiment of this application An embodiment of the deployment device includes:
  • the obtaining unit 601 is configured to obtain node information of the target network
  • the determining unit 602 is configured to determine whether the target network is a hybrid software-defined network SDN according to the node information;
  • the first establishing unit 604 establishes a virtual local network VLAN group according to the initial SDN node deployment position and a preset reinforcement learning algorithm.
  • the embodiment of this application implements the initial deployment strategy of the hybrid SDN network for the traditional network according to the actual situation of the network, so that the deployment process is smoother and more effective, and the coordination between network control rate and link load balance is more reasonable; at the same time, virtual local network groups are adopted.
  • the method of division divides the entire hybrid software custom network into different local virtual network groups, and each network node needs to communicate through the local virtual network group, which improves network efficiency.
  • another embodiment of the deployment apparatus for a hybrid software customized network in the embodiment of the present application includes:
  • the obtaining unit 601 is configured to obtain node information of the target network
  • the determining unit 602 is configured to determine whether the target network is a hybrid software-defined network SDN according to the node information;
  • the first establishing unit 604 establishes a virtual local network VLAN group according to the initial SDN node deployment position and a preset reinforcement learning algorithm.
  • the deployment apparatus for a hybrid software-defined network further includes: a second execution unit 605, configured to execute a hybrid SDN network transition deployment strategy if the target network is the hybrid SDN, and the hybrid SDN network transition deployment The strategy is used to calculate the deployment location of a new SDN node.
  • the deployment device of the hybrid software custom network further includes: a second establishing unit 606, configured to establish a newly-added virtual local network VLAN group according to the deployment position of the newly-added SDN node, and the newly-added VLAN group includes a VLAN group Decomposition and VLAN combination.
  • the first establishing unit 603 is specifically configured to:
  • the first establishing unit 603 is specifically further configured to:
  • the single switch vi in the SDN switch set Vs in the hybrid SDN network is sorted from largest to smallest by degree; two variables Valready and Valone are set, and Valready represents the set of nodes that have been grouped by VLAN.
  • the Valone represents the remaining node set without VLAN grouping, the Valready is initialized to an empty set, the Valone is initialized to V-Vs, and the V represents all nodes; the priority SDN nodes are traversed; If Valready is an empty set, add the node vi directly to the Valready set, and continue to traverse the next node; if there is only one node in the Valready, the shortest path of the only node in the Valready is calculated through the vi , And create a new VLAN group, add the information of the new VLAN group to the VL, and remove the nodes on the new VLAN group from the Valone; if there are at least two nodes in the Valready, then The two nodes closest to the vi in the Valready calculate and establish two new VLAN groups, and add the two new VLAN groups established to the set VL of VLANs in the hybrid SDN network, and remove them from the Valone Nodes on the two new VLAN groups; obtain the two values of VL and Valone, where VL represents a
  • the deployment device of the hybrid software self-defined network further includes: an optimization unit 607, which is used to optimize the hybrid SDN when the hybrid SDN fails.
  • the optimization unit 607 is specifically configured to: when the hybrid SDN fails, determine the type of failure of the hybrid SDN, the type of failure includes communication failures within VLAN groups, communication failures between VLAN groups, and communication between SDN groups Fault; when the fault type is a communication failure within a VLAN group, the fault tolerance strategy in the VLAN group is executed and the fault tolerant link is recalculated; when the fault type is a communication failure between VLAN groups, the fault tolerance strategy between VLAN groups is executed and the fault tolerance chain is recalculated When the fault type is a communication failure between SDN groups, implement the fault tolerance strategy between SDN groups and recalculate the fault-tolerant link.
  • the embodiment of this application implements the initial deployment strategy of the hybrid SDN network for the traditional network according to the actual situation of the network, so that the deployment process is smoother and more effective, and the coordination between network control rate and link load balance is more reasonable; at the same time, virtual local network groups are adopted.
  • the method of division divides the entire hybrid software custom network into different local virtual network groups, and each network node needs to communicate through the local virtual network group, which improves network efficiency.
  • different fault tolerance strategies are adopted for different fault scenarios to improve the recovery performance of hybrid software custom network fault tolerance.
  • FIGS 6 to 7 above describe in detail the deployment device of the hybrid software custom network in the embodiment of the present application from the perspective of modular functional entities.
  • the following describes the deployment device of the hybrid software custom network in the embodiment of the present application from the perspective of hardware processing. Give a detailed description.
  • FIG. 8 is a schematic structural diagram of a deployment device of a hybrid software-defined network provided by an embodiment of the present application.
  • the deployment device 800 of the hybrid software-defined network may have relatively large differences due to differences in configuration or performance, and may include one or One or more processors (central processing units, CPU) 801 (for example, one or more processors) and memory 809, and one or more storage media 808 for storing application programs 807 or data 806 (for example, one or one storage device with a large amount of ).
  • the memory 809 and the storage medium 808 may be short-term storage or persistent storage.
  • the program stored in the storage medium 808 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the deployment device of the hybrid software custom network. Further, the processor 801 may be configured to communicate with the storage medium 808, and execute a series of instruction operations in the storage medium 808 on the deployment device 800 of the hybrid software custom network.
  • the deployment device 800 of the hybrid software customized network may also include one or more power supplies 802, one or more wired or wireless network interfaces 803, one or more input and output interfaces 804, and/or, one or more operating systems 805 , Such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD and so on.
  • operating systems 805 Such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD and so on.
  • the processor 801 can execute the functions of the acquisition unit 601, the judgment unit 602, the first execution unit 603, the first establishment unit 604, the second execution unit 605, the second establishment unit 606, and the optimization unit 607 in the foregoing embodiment.
  • the processor 801 is the control center of the deployment device of the hybrid software custom network, and can perform processing according to the set deployment method of the hybrid software custom network.
  • the processor 801 uses various interfaces and lines to connect various parts of the deployment equipment of the entire hybrid software customized network, and by running or executing software programs and/or modules stored in the memory 809, and calling data stored in the memory 809, Perform various functions and processing data of the deployment device of the hybrid software custom network, thereby realizing the deployment of the hybrid software custom network.
  • the storage medium 808 and the memory 809 are both carriers for storing data.
  • the storage medium 808 may refer to an internal memory with a small storage capacity but a fast speed, and the storage medium 809 may have a large storage capacity but a slow storage speed. External memory.
  • the memory 809 may be used to store software programs and modules.
  • the processor 801 executes various functional applications and data processing of the deployment device 800 of the hybrid software customized network by running the software programs and modules stored in the memory 809.
  • the memory 809 may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, an application program required by at least one function (for example, determine whether the target network is a hybrid software-defined network SDN, etc.) according to node information;
  • the storage data area can store data (such as a virtual local network VLAN group, etc.) created according to the use of deployment equipment of the hybrid software custom network.
  • the memory 809 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • a non-volatile memory such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • the computer-readable storage medium may be a non-volatile computer-readable storage medium.
  • the computer-readable storage medium stores instructions. When the instructions run on a computer At the time, the computer is made to execute the steps of the following hybrid software custom network deployment method:
  • Groove filling is performed on the inside of the target closed curve to obtain a lesion image of the target fundus lesion.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本申请公开了基于分布式部署的混合软件自定义网络的部署方法、装置、设备及存储介质，用于让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理，提高了网络工作效率。本申请方法包括：获取目标网络的节点信息；根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；若所述目标网络不为所述混合SDN，则执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。

Description

混合软件自定义网络的部署方法、装置、设备及存储介质
本申请要求于2019年6月17日提交中国专利局、申请号为201910521265.9、发明名称为“混合软件自定义网络的部署方法、装置、设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在申请中。
技术领域
本申请涉及分布式部署领域，尤其涉及一种混合软件自定义网络的部署方法、装置、设备及存储介质。
背景技术
软件定义网络(software defined network，SDN)，是Emulex网络一种新型网络创新架构，是网络虚拟化的一种实现方式，其核心技术OpenFlow通过将网络设备控制面与数据面分离开来，从而实现了网络流量的灵活控制，使网络作为管道变得更加智能。
由于计算、存储资源的虚拟化，上层设备(服务器、存储)的资源打通了，如果网络架构依旧保持原样，要想保证系统的稳定运行，就需要对计算、存储资源池做相应的配置，SDN和传统网络混合组网，称为混合SDN，能够以经济实惠的迭代方式实现网络架构的过渡，而不会因全部更换导致业务中断。
发明人意识到目前在混合SDN中，其部署策略停留在静态部署上，只针对当前资源进行部署，而没有考虑到从传统网络过渡到软件自定义网络是一个渐进的过程。并且在混合软件自定义网络中的通信方式也过于单一，并没有充分利用好软件自定义网络结构的优点。
发明内容
本申请提供了一种混合软件自定义网络的部署方法、装置、设备及存储介质，用于让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理，提高了网络工作效率。
本申请实施例的第一方面提供一种混合软件自定义网络的部署方法，包括：获取目标网络的节点信息；根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；若所述目标网络不为所述混合SDN，则执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
本申请实施例的第二方面提供了一种混合软件自定义网络的部署装置，包括：获取单元，用于获取目标网络的节点信息；判断单元，用于根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；第一执行单元，若所述目标网络不为所述混合SDN，则用于执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；第一建立单元，根据所述初始SDN节点部署位置和预置的强化学习算法建立虚 拟本地网络VLAN组。
本申请实施例的第三方面提供了一种混合软件自定义网络的部署设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述混合软件自定义网络的部署方法。
本申请实施例的第四方面提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在计算机上运行时，使得计算机执行上述混合软件自定义网络的部署方法的步骤。
本申请实施例，根据网络的实际情况，采用不同的部署策略，让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理。
附图说明
图1为本申请实施例中混合软件自定义网络的部署方法的一个实施例示意图；
图2为本申请实施例中执行混合SDN网络初期部署策略的一个应用场景示意图；
图3为本申请实施例中VLAN组合并过程的一个应用场景示意图；
图4为本申请实施例中执行混合SDN网络过渡部署策略的另一个应用场景示意图；
图5为本申请实施例中混合SDN网络动态部署cost变化值的一个示意图；
图6为本申请实施例中混合软件自定义网络的部署装置的一个实施例示意图；
图7为本申请实施例中混合软件自定义网络的部署装置的另一个实施例示意图；
图8为本申请实施例中混合软件自定义网络的部署设备的一个实施例示意图。
具体实施方式
本申请提供了一种混合软件自定义网络的部署方法、装置、设备及存储介质，用于让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理，提高了网络工作效率。
请参阅图1，本申请实施例提供的一种混合软件自定义网络的部署方法的流程图，具体包括：
101、获取目标网络的节点信息。
服务器获取目标网络的节点信息，例如，终端可以获取目标网络节点的名称标识，地址标识，以及配置信息的配置标识符，访问该目标网络节点的路径等，还可以是其他需要用到的节点信息，具体此处不做限定。需要说明的是，执行主体可以是服务器或者是终端，只要能实现混合软件自定义网络的部署即可，本申请以服务器为例进行说明。
102、根据节点信息判断目标网络是否为混合软件定义网络SDN。
服务器根据节点信息判断目标网络是否为混合软件定义网络SDN。具体的，服务器根据节点信息判断目标网络当前执行的策略是否为混合SDN策略。混合SDN策略为目标网络的节点可以执行的规则，即目标网络的节点按照一定的规则对网络进行管理。其中，策略分为两种情况，一种为将传统网络升级为混合SDN网络，即混合SDN网络初期部署策略；另一种为混合SDN网络通过增加SDN节点(交换机)进行网络升级，即混合SDN网络过渡部署策略。在执行混合SDN网络初期部署策略过程中，只需要考虑硬件资源的问题，而在执行混合SDN网络过渡部署策略过程中，除了要考虑硬件资源的问题，还需要考虑对现有虚 拟本地网络(virtual local area network，VLAN)组的破坏，和对现有硬件资源的移动(即SDN节点(交换机)的资源迁移)。
103、若目标网络不为混合SDN，则执行混合SDN网络初期部署策略，该混合SDN网络初期部署策略用于计算初始SDN节点部署位置。
若目标网络不为混合SDN，则服务器执行混合SDN网络初期部署策略，服务器对目标网络进行升级，即从传统网络升级为混合SDN网络，其中，混合SDN网络初期部署策略用于计算初始SDN节点部署位置。具体的，根据混合SDN网络初期部署策略选出多个目标部署节点，将多个目标部署节点中的交换机替换为SDN交换机，并进行VLAN分组。
如图2所示，图2为传统网络升级为混合SDN网络过程的一个示意图(即执行混合SDN网络初期部署策略)，目标网络包括传统交换机A、传统交换机B、传统交换机C、传统交换机D、传统交换机E、传统交换机F、传统交换机G、传统交换机H、传统交换机I、传统交换机J；根据预置的强化学习算法计算得到需要将传统交换机D、E替换成SDN交换机D、E。然后将其中的一些链路通过混合SDN网络通信模块配置VLAN组，具体的，传统交换机A与传统交换机B、传统交换机C之间，传统交换机C与传统交换机G之间，传统交换机G与传统交换机J之间，传统交换机B与SDN交换机E之间，传统交换机H与SDN交换机D之间，传统交换机H与传统交换机J之间的链路都为VLAN1链路；传统交换机J与传统交换机H、传统交换机I之间，传统交换机I与传统交换机F之间，传统交换机F与SDN交换机E之间，传统交换机F与SDN交换机E之间的链路都为VLAN2链路；SDN交换机D与SDN交换机E之间的链路为VLAN3链路，其中，传统交换机H与传统交换机J之间的链路既属于VLAN1又属于VLAN2，而SDN交换机D与传统交换机B、传统交换机G之间为普通链路。
104、根据初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
服务器根据初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。例如，首先采用预置的强化学习算法选择出最优的部署节点。选出多个部署节点后，进行VLAN组划分，得到VLAN组的集合。VLAN组划分得过程具体为：先对SDN交换机节点进行SDN节点VLAN分组,得到现阶段VLAN分组集合与孤立交换机集合，然后依次进行孤立节点VLAN分组、VLAN组合并和VLAN组分解。
SDN节点VLAN分组过程为：1、先对混合SDN网络中SDN交换机集合V S中的单个交换机vi按照度进行由大到小排序，因为部署节点已经考虑了链路问题，在这里将只考虑交换机(即节点)的度，将其作为VLAN分组的优先级。2、设置两个变量Valready和Valone，Valready表示已经进行VLAN分组的交换机，Valone表示剩余没有VLAN分组的交换机，Valready和Valone初始化分别为空集和所有交换机V-Vs。3、对已经排序了优先级的SDN交换机节点进行遍历混合SDN网络中所有交换机的集合V。若Valready为空集，那么将该节点vi直接加入到集合中，继续遍历下一个节点。若Valready只有一个节点，那么通过vi 计算出Valready中唯一节点的最短路径，并建立VLAN组，将信息加入到混合SDN网络中VLAN的集合VL中，并将新VLAN组上的节点从所述Valone去除。若Valready中存在至少两个节点，则对Valready中与vi最近的两个节点进行计算并建立两个新VLAN组，将建立的两个新VLAN组加入到混合SDN网络中VLAN的集合VL中，并从Valone去除两个新VLAN组上的节点；若所述Valready中存在至少两个节点，则对集合Valready中与vi最近的两个节点进行计算并建立两个新VLAN组，在这里可能会存在两个VLAN组有公用链路，其实是在浪费VLAN资源，所以对其中两个VLAN组缩减为一个，只将与vi最近节点建立的VLAN组加入到VL中，并从Valone去除该VLAN组上的交换机。最终返回VL和Valone两个值，分别代表现阶段VLAN组集合和孤立的交换机(即未进行VLAN分组的交换机)。
孤立节点VLAN分组过程为：对所述孤立节点集合中孤立节点以度为优先级进行排序，得到孤立节点中已经建立VLAN组通信的节点集合，然后对所述孤立节点中已经建立VLAN组通信的节点集合之外的节点进行遍历，根据最近两个节点之间链路路径集合中的公共路径数量建立新VLAN组，并将所述新VLAN组添加到所述现阶段VLAN分组集合，得到第一过渡VLAN分组集合。具体的：1、对输入的孤立节点以度为优先级进行排序，然后初始化一个Valready集合，该集合表示孤立节点中已经建立VLAN组通信的节点。2、先对孤立节点集合Valone进行遍历，若已经建立VLAN组通信的节点略过，若没有则计算出Vs集合中与vi最近的两个SDN节点，记录下两个链路路径集合E1和E2。在这两个链路路径集合E1和E2中，若有公共路径，那么只留下与vi最近SDN节点的链路路径集合，将该路径集合建立一个新VLAN组，并加其信息加入到VL组，最后将新VLAN组上的交换机加入到Valready集合中。若这两个链路路径没有公共路径，那么将这两个链路路径的集合取并集，建立一个新VLAN组，将其信息加入到VL组，最后将新VLAN组上的交换机加入到Valready集合。
VLAN组合并过程为：确定第一过渡VLAN分组集合中VLAN组最多的链路，选择任意两条链路上的VLAN组进行合并，得到第二过渡VLAN分组集合。具体的：1、找出混合SDN网络中链路中VLAN组最多的链路。在图3的左图中为链路e B,D、e D,G和e H,J这三条，对前两条链路上的VLAN组进行合并，这里以链路e B,D为例，此时VLmine为VLAN2组，VLmaxe为VLAN1组。2、将VLmine和VLmaxe信息从VL中去除，然后将VLmine覆盖的传统交换机记为Vminc(传统交换机B)，SDN交换机记为Vmins(SDN交换机D、E)，将VLmaxe覆盖的传统交换机记为Vmaxc(传统交换机A、B、C和G)，SDN交换机记为Vmaxs(SDN交换机D、E)，设置SDN交换机并集Vs。3、对Vmaxc-Vminc中的传统交换机进行遍历，把交换机都加入到VLmaxe中。4、将VLmaxe加入到VL中，最终混合SDN网络VLAN信息如图3的右图所示。
VLAN组分解过程为：根据第二过渡VLAN分组集合确定需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN的传统交换机集合，然后对所述需要重组的VLAN组集合和所述需要重组的传统交换机节点集合进行赋值，并将需要重组的VLAN集合从VLAN组集合中删除，最后对新增SDN交换机和剩余传统交换机进行VLAN组划分，得到重组后的VLAN组集合。具体的：1、先初始化三个变量VLregroup、Vnewc和Valready，分别代表需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN的传统交换机集合。2、对VLregroup和Vnewc变量进行赋值，并将需要重组的VLAN集合从VLAN组集合VL去除。3、对新增SDN交换机进行VLAN组划分，其划分方式同SDN节点VLAN分组过程。4、是对剩余传统交换机进行VLAN组划分，其划分方式同孤立节点VLAN分组过程。5、得到重组后的VLAN组集合VL。
需要说明的是，整个网络可以看作是一个网状结构，路由器和服务器看作是网状结构的节点，度指的就是路由器节点连接的链路条数。
本申请实施例，根据网络的实际情况，对传统网络执行混合SDN网络初期部署策略，让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理；同时采用虚拟本地网络组划分的方式将整个混合软件自定义网络划分不同本地虚拟网络组，每一个网络节点都需要通过本地虚拟网络组进行通信，提高了网络工作效率。
可选的，在上述图1对应的实施例的基础上，本申请实施例提供的混合软件自定义网络的部署方法的可选实施例中，所述方法还包括：若所述目标网络为所述混合SDN，则执行混合SDN网络过渡部署策略，所述混合SDN网络过渡部署策略用于计算新增SDN节点部署位置。
例如，若目标网络为混合SDN，则服务器执行混合SDN网络过渡部署策略，服务器对目标网络进行升级，即为混合SDN网络增加SDN节点(交换机)进行网络升级，来增强网络的流量控制、网络容错能力等。其中，混合SDN网络过渡部署策略用于计算新增SDN节点(交换机)部署位置。如图4所示，图4为混合SDN网络增加SDN节点过程的一个示意图(即执行混合SDN网络过渡部署策略)，目标网络包括传统交换机A、传统交换机B、传统交换机C、SDN交换机D、SDN交换机E、传统交换机F、传统交换机G、传统交换机H、传统交换机I、传统交换机J；根据预置的强化学习算法计算得到需要将传统交换机H替换成SDN交换机H。具体过程与图2类似，此处不再赘述。
本申请实施例，根据网络的实际情况，对混合SDN执行混合SDN网络过渡部署策略，让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理。
可选的，在上述图1对应的实施例的基础上，本申请实施例提供的混合软件自定义网络的部署方法的可选实施例中，在所述执行混合SDN网络过渡部署策略之后，所述方法还包括：根据所述新增SDN节点部署位置建立新增虚拟本地网络VLAN组，所述新增VLAN组包括VLAN组分解和VLAN组合并。
例如，服务器根据新增SDN节点部署位置建立新增虚拟本地网络VLAN组，新增VLAN组包括VLAN组分解和VLAN组合并。例如图4所示，确定需要增加SDN交换机H后，重新 通过混合SDN网络通信模块配置VLAN组，VLAN组重新配置后，传统交换机A与传统交换机B、传统交换机C之间，传统交换机C与传统交换机G之间，传统交换机G与传统交换机J之间，传统交换机B与SDN交换机E之间，SDN交换机H与传统交换机J之间的链路都为VLAN1链路；传统交换机J与SDN交换机H、传统交换机I之间，传统交换机I与传统交换机F之间，传统交换机F与SDN交换机E之间的链路都为VLAN2链路；SDN交换机D与SDN交换机E之间的链路为VLAN3链路，SDN交换机H与SDN交换机D之间的链路为VLAN4链路，SDN交换机H与SDN交换机E之间的链路为VLAN5链路，其中，传统交换机H与传统交换机J之间的链路既属于VLAN1又属于VLAN2。
本申请实施例，通过VLAN组的划分，可以使得整个混合SDN网络具有充分利用SDN网络自定义转发方式、统计流量等功能，从而让更多流量受到SDN控制器的控制。
可选的，在上述图1对应的实施例的基础上，本申请实施例提供的混合软件自定义网络的部署方法的可选实施例中，所述根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组，包括：根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合；对所述孤立节点集合中孤立节点以度为优先级进行排序，得到孤立节点中已经建立VLAN组通信的节点集合，然后对所述孤立节点中已经建立VLAN组通信的节点集合之外的节点进行遍历，根据最近两个节点之间链路路径集合中的公共路径数量建立新VLAN组，并将所述新VLAN组添加到所述现阶段VLAN分组集合，得到第一过渡VLAN分组集合；确定所述第一过渡VLAN分组集合中VLAN组最多的链路，选择任意两条链路上的VLAN组进行合并，得到第二过渡VLAN分组集合；根据所述第二过渡VLAN分组集合确定需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN的传统交换机集合，然后对所述需要重组的VLAN组集合和所述需要重组的传统交换机节点集合进行赋值，并将需要重组的VLAN集合从VLAN组集合中删除，最后对新增SDN交换机和剩余传统交换机进行VLAN组划分，得到重组后的VLAN组集合。
本申请实施例，通过使用强化学习算法动态部署SDN交换机，解决整个网络向SDN网络平稳过渡的问题，并满足在传统网络中或混合SDN网络中，逐步替换旧交换机的需求。
可选的，在上述图1对应的实施例的基础上，本申请实施例提供的混合软件自定义网络的部署方法的可选实施例中，所述根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合，包括：根据预置的强化学习算法对混合SDN网络中SDN交换机集合Vs中的单个交换机vi按照度进行由大到小排序；设置两个变量Valready和Valone，所述Valready表示已经进行VLAN分组的节点集合，所述Valone表示剩余没有VLAN分组的节点集合，所述Valready初始化为空集，所述Valone初始化为V-Vs，所述V表示所有节点；对已经排序了优先级的SDN节点进行遍历；若所述Valready为空集，则将所述节点vi直接加入到Valready集合中，继续遍历下一个节点；若所述Valready只存在一个节点，则通过所述vi计算出所述Valready中唯一节点的最 短路径，并建立新VLAN组，将所述新VLAN组的信息加入到VL中，并将所述新VLAN组上的节点从所述Valone去除；若所述Valready中存在至少两个节点，则对所述Valready中与所述vi最近的两个节点进行计算并建立两个新VLAN组，将建立的所述两个新VLAN组加入到混合SDN网络中VLAN的集合VL中，并从所述Valone去除所述两个新VLAN组上的节点；得到所述VL和Valone两个值，所述VL表示现阶段VLAN组集合，所述Valone表示孤立的节点集合。本申请实施例，对SDN节点VLAN分组过程进行了细化，增加了本申请实施例的实现方式。
可选的，在上述图1对应的实施例的基础上，本申请实施例提供的混合软件自定义网络的部署方法的可选实施例中，所述方法还包括：当所述混合SDN发生故障时，对所述混合SDN进行优化。
例如，服务器对混合SDN进行优化，服务器通过预置公式计算出混合SDN网络动态部署cost变化值。该cost值由优化目标取反计算得出，求cost值最小值即为优化目标，网络最大控制率与链路负载均衡都取50％占比。例如，如图5所示，其中X轴为迭代次数，每一次循环中迭代，Y轴为cost值，本次实验循环120次，选取第一次、第六十次和第一百二十次作为参考，每一次循环迭代600次(图中仅显示了400次)，学习率为1e-3，奖励衰减度为0.99，随机因子为0.4，VLAN组固定为10组，选取的替换传统交换机为3个，选用的网络拓扑为TW Telecom。其中优化值目标定义为：
Figure PCTCN2019102469-appb-000001
其中，γ和β为网络流量系数和链路负载均衡系数，γ与β之和为1，ConMax(G)是最大化控制网络函数，LinkBal(G)是负载均衡函数。
可以发现在在前面的100次迭代，因为随机因子的原因，cost值会变得浮动较大，在之后可以看出基本趋于缓和，但是因为在循环中，状态矩阵是不断被训练。经过多次训练后，可以发现在450次迭代后，第一次、第六十次和第一百二十次后的数据，可以忽略随机因子的因素，第一百二十次循环的cost值明显小于其余两次，其中第六十次循环在之后偶尔会发生抖动，因为其状态矩阵不够全面，导致面对新的状态时，需要通过随机选取最优点，所以可能会导致cost值发生抖动。虽然第一百二十次也会发生抖动，但是相对于第六十次较为稳定。在第一次循环时其cost值可能因为被困于局部最优的原因，导致其cost值一直平稳并且没有什么变化。本申请实施例，在混合SDN发生故障，对混合SDN进行优化，消除故障对混合SDN的影响，提高了工作效率。
可选的，在上述图1对应的实施例的基础上，本申请实施例提供的混合软件自定义网络的部署方法的可选实施例中，所述当所述混合SDN发生故障时，对所述混合SDN进行优化，包括：当所述混合SDN发生故障时，确定所述混合SDN的故障类型，所述故障类型包括VLAN组内通信故障、VLAN组间通信故障和SDN组间通信故障；当故障类型为VLAN组内通信故障时，执行VLAN组内故障容错策略，重新计算容错链路；当故障类型为VLAN组间通信故障时，执行VLAN组间故障容错策略，重新计算容错链路；当故障类型为SDN组间通 信故障时，执行SDN组间故障容错策略，重新计算容错链路。本申请实施例，针对不同的故障场景，采用不同的故障容错策略，提高了混合软件自定义网络故障容错的恢复性能。
上面对本申请实施例中混合软件自定义网络的部署方法进行了描述，下面对本申请实施例中混合软件自定义网络的部署装置进行描述，请参阅图6，本申请实施例中混合软件自定义网络的部署装置的一个实施例包括：
获取单元601，用于获取目标网络的节点信息；
判断单元602，用于根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；
第一执行单元603，若所述目标网络不为所述混合SDN，则用于执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；
第一建立单元604，根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
本申请实施例，根据网络的实际情况，对传统网络执行混合SDN网络初期部署策略，让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理；同时采用虚拟本地网络组划分的方式将整个混合软件自定义网络划分不同本地虚拟网络组，每一个网络节点都需要通过本地虚拟网络组进行通信，提高了网络工作效率。
请参阅图7，本申请实施例中混合软件自定义网络的部署装置的另一个实施例包括：
获取单元601，用于获取目标网络的节点信息；
判断单元602，用于根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；
第一执行单元603，若所述目标网络不为所述混合SDN，则用于执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；
第一建立单元604，根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
可选的，混合软件自定义网络的部署装置还包括：第二执行单元605，若所述目标网络为所述混合SDN，则用于执行混合SDN网络过渡部署策略，所述混合SDN网络过渡部署策略用于计算新增SDN节点部署位置。
可选的，混合软件自定义网络的部署装置还包括：第二建立单元606，用于根据所述新增SDN节点部署位置建立新增虚拟本地网络VLAN组，所述新增VLAN组包括VLAN组分解和VLAN组合并。
可选的，第一建立单元603具体用于：
根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合；对所述孤立节点集合中孤立节点以度为优先级进行排序，得到孤立节点中已经建立VLAN组通信的节点集合，然后对所述孤立节点中已经建立VLAN组通信的节点集合之外的节点进行遍历，根据最近两个节点之间链路路径集合中的公共路径数量建立新VLAN组，并将所述新VLAN组添加到所述现阶段VLAN分组集合，得到第一过渡VLAN分组集合；确定所述第一过渡VLAN分组集合中VLAN组最多的链路，选择任意两条链路上的VLAN组进行合并，得到第二过渡VLAN分组集合；根据所述第二过渡VLAN分组集合确定需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN 的传统交换机集合，然后对所述需要重组的VLAN组集合和所述需要重组的传统交换机节点集合进行赋值，并将需要重组的VLAN集合从VLAN组集合中删除，最后对新增SDN交换机和剩余传统交换机进行VLAN组划分，得到重组后的VLAN组集合。
可选的，第一建立单元603具体还用于：
根据预置的强化学习算法对混合SDN网络中SDN交换机集合Vs中的单个交换机vi按照度进行由大到小排序；设置两个变量Valready和Valone，所述Valready表示已经进行VLAN分组的节点集合，所述Valone表示剩余没有VLAN分组的节点集合，所述Valready初始化为空集，所述Valone初始化为V-Vs，所述V表示所有节点；对已经排序了优先级的SDN节点进行遍历；若所述Valready为空集，则将所述节点vi直接加入到Valready集合中，继续遍历下一个节点；若所述Valready只存在一个节点，则通过所述vi计算出所述Valready中唯一节点的最短路径，并建立新VLAN组，将所述新VLAN组的信息加入到VL中，并将所述新VLAN组上的节点从所述Valone去除；若所述Valready中存在至少两个节点，则对所述Valready中与所述vi最近的两个节点进行计算并建立两个新VLAN组，将建立的所述两个新VLAN组加入到混合SDN网络中VLAN的集合VL中，并从所述Valone去除所述两个新VLAN组上的节点；得到所述VL和Valone两个值，所述VL表示现阶段VLAN组集合，所述Valone表示孤立的节点集合。
可选的，混合软件自定义网络的部署装置还包括：优化单元607，当所述混合SDN发生故障时，用于对所述混合SDN进行优化。
可选的，优化单元607具体用于：当所述混合SDN发生故障时，确定所述混合SDN的故障类型，所述故障类型包括VLAN组内通信故障、VLAN组间通信故障和SDN组间通信故障；当故障类型为VLAN组内通信故障时，执行VLAN组内故障容错策略，重新计算容错链路；当故障类型为VLAN组间通信故障时，执行VLAN组间故障容错策略，重新计算容错链路；当故障类型为SDN组间通信故障时，执行SDN组间故障容错策略，重新计算容错链路。
本申请实施例，根据网络的实际情况，对传统网络执行混合SDN网络初期部署策略，让部署过程更平滑和有效，网络控制率和链路负载均衡之间协调更合理；同时采用虚拟本地网络组划分的方式将整个混合软件自定义网络划分不同本地虚拟网络组，每一个网络节点都需要通过本地虚拟网络组进行通信，提高了网络工作效率。同时针对不同的故障场景，采用不同的故障容错策略，提高了混合软件自定义网络故障容错的恢复性能。
上面图6至图7从模块化功能实体的角度对本申请实施例中的混合软件自定义网络的部署装置进行详细描述，下面从硬件处理的角度对本申请实施例中混合软件自定义网络的部署设备进行详细描述。
图8是本申请实施例提供的一种混合软件自定义网络的部署设备的结构示意图，该混合软件自定义网络的部署设备800可因配置或性能不同而产生比较大的差异，可以包括一 个或一个以上处理器(central processing units，CPU)801(例如，一个或一个以上处理器)和存储器809，一个或一个以上存储应用程序807或数据806的存储介质808(例如一个或一个以上海量存储设备)。其中，存储器809和存储介质808可以是短暂存储或持久存储。存储在存储介质808的程序可以包括一个或一个以上模块(图示没标出)，每个模块可以包括对混合软件自定义网络的部署设备中的一系列指令操作。更进一步地，处理器801可以设置为与存储介质808通信，在混合软件自定义网络的部署设备800上执行存储介质808中的一系列指令操作。
混合软件自定义网络的部署设备800还可以包括一个或一个以上电源802，一个或一个以上有线或无线网络接口803，一个或一个以上输入输出接口804，和/或，一个或一个以上操作系统805，例如Windows Serve，Mac OS X，Unix，Linux，FreeBSD等等。本领域技术人员可以理解，图8中示出的混合软件自定义网络的部署设备结构并不构成对混合软件自定义网络的部署设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。处理器801可以执行上述实施例中获取单元601、判断单元602、第一执行单元603、第一建立单元604、第二执行单元605、第二建立单元606和优化单元607的功能。
下面结合图8对混合软件自定义网络的部署设备的各个构成部件进行具体的介绍：
处理器801是混合软件自定义网络的部署设备的控制中心，可以按照设置的混合软件自定义网络的部署方法进行处理。处理器801利用各种接口和线路连接整个混合软件自定义网络的部署设备的各个部分，通过运行或执行存储在存储器809内的软件程序和/或模块，以及调用存储在存储器809内的数据，执行混合软件自定义网络的部署设备的各种功能和处理数据，从而实现混合软件自定义网络的部署。存储介质808和存储器809都是存储数据的载体，本申请实施例中，存储介质808可以是指储存容量较小，但速度快的内存储器，而存储器809可以是储存容量大，但储存速度慢的外存储器。
存储器809可用于存储软件程序以及模块，处理器801通过运行存储在存储器809的软件程序以及模块，从而执行混合软件自定义网络的部署设备800的各种功能应用以及数据处理。存储器809可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如根据节点信息判断目标网络是否为混合软件定义网络SDN等)等；存储数据区可存储根据混合软件自定义网络的部署设备的使用所创建的数据(比如虚拟本地网络VLAN组等)等。此外，存储器809可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在本申请实施例中提供的混合软件自定义网络的部署方法程序和接收到的数据流存储在存储器中，当需要使用时，处理器801从存储器809中调用。
本申请还提供一种计算机可读存储介质，该计算机可读存储介质可以为非易失性计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在计算机上运行时，使得计算机执行如下混合软件自定义网络的部署方法的步骤：
获取原始眼底光学相干断层扫描技术OCT影像；
对所述原始眼底OCT影像进行分割预处理，得到预处理的眼底OCT图像；
在所述预处理的眼底OCT图像中生成一个初始闭合曲线，所述初始闭合曲线的内部包括目标眼底病灶；
通过预置的形态学蛇形曲线演化算法确定所述初始闭合曲线的初始曲线函数；
根据所述初始曲线函数对所述初始闭合曲线进行演化，得到目标闭合曲线，所述目标闭合曲线的目标曲线函数表示所述目标眼底病灶的轮廓；
对所述目标闭合曲线的内部进行凹槽填充，得到所述目标眼底病灶的病灶图像。
以上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (20)

  1. 一种混合软件自定义网络的部署方法，包括：
    获取目标网络的节点信息；
    根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；
    若所述目标网络不为所述混合SDN，则执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；
    根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
  2. 根据权利要求1所述的混合软件自定义网络的部署方法，所述方法还包括：
    若所述目标网络为所述混合SDN，则执行混合SDN网络过渡部署策略，所述混合SDN网络过渡部署策略用于计算新增SDN节点部署位置。
  3. 根据权利要求2所述的混合软件自定义网络的部署方法，在所述执行混合SDN网络过渡部署策略之后，所述方法还包括：
    根据所述新增SDN节点部署位置建立新增虚拟本地网络VLAN组，所述新增VLAN组包括VLAN组分解和VLAN组合并。
  4. 根据权利要求1所述的混合软件自定义网络的部署方法，所述根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组，包括：
    根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合；
    对所述孤立节点集合中孤立节点以度为优先级进行排序，得到孤立节点中已经建立VLAN组通信的节点集合，然后对所述孤立节点中已经建立VLAN组通信的节点集合之外的节点进行遍历，根据最近两个节点之间链路路径集合中的公共路径数量建立新VLAN组，并将所述新VLAN组添加到所述现阶段VLAN分组集合，得到第一过渡VLAN分组集合；
    确定所述第一过渡VLAN分组集合中VLAN组最多的链路，选择任意两条链路上的VLAN组进行合并，得到第二过渡VLAN分组集合；
    根据所述第二过渡VLAN分组集合确定需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN的传统交换机集合，然后对所述需要重组的VLAN组集合和所述需要重组的传统交换机节点集合进行赋值，并将需要重组的VLAN集合从VLAN组集合中删除，最后对新增SDN交换机和剩余传统交换机进行VLAN组划分，得到重组后的VLAN组集合。
  5. 根据权利要求4所述的混合软件自定义网络的部署方法，所述根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合，包括：
    根据预置的强化学习算法对混合SDN网络中SDN交换机集合Vs中的单个交换机vi按照度进行由大到小排序；
    设置两个变量Valready和Valone，所述Valready表示已经进行VLAN分组的节点集合，所述Valone表示剩余没有VLAN分组的节点集合，所述Valready初始化为空集，所述Valone初始化为V-Vs，所述V表示所有节点；
    对已经排序了优先级的SDN节点进行遍历；
    若所述Valready为空集，则将所述节点vi直接加入到Valready集合中，继续遍历下一个节点；
    若所述Valready只存在一个节点，则通过所述vi计算出所述Valready中唯一节点的最短路径，并建立新VLAN组，将所述新VLAN组的信息加入到VL中，并将所述新VLAN组上的节点从所述Valone去除；
    若所述Valready中存在至少两个节点，则对所述Valready中与所述vi最近的两个节点进行计算并建立两个新VLAN组，将建立的所述两个新VLAN组加入到混合SDN网络中VLAN的集合VL中，并从所述Valone去除所述两个新VLAN组上的节点；
    得到所述VL和Valone两个值，所述VL表示现阶段VLAN组集合，所述Valone表示孤立的节点集合。
  6. 根据权利要求1-5中任一所述的混合软件自定义网络的部署方法，所述方法还包括：
    当所述混合SDN发生故障时，对所述混合SDN进行优化。
  7. 根据权利要求6所述的混合软件自定义网络的部署方法，所述当所述混合SDN发生故障时，对所述混合SDN进行优化，包括：
    当所述混合SDN发生故障时，确定所述混合SDN的故障类型，所述故障类型包括VLAN组内通信故障、VLAN组间通信故障和SDN组间通信故障；
    当故障类型为VLAN组内通信故障时，执行VLAN组内故障容错策略，重新计算容错链路；
    当故障类型为VLAN组间通信故障时，执行VLAN组间故障容错策略，重新计算容错链路；
    当故障类型为SDN组间通信故障时，执行SDN组间故障容错策略，重新计算容错链路。
  8. 一种混合软件自定义网络的部署装置，包括：
    获取单元，用于获取目标网络的节点信息；
    判断单元，用于根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；
    第一执行单元，若所述目标网络不为所述混合SDN，则用于执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；
    第一建立单元，根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
  9. 根据权利要求8所述的混合软件自定义网络的部署装置，所述混合软件自定义网络的部署装置还包括：
    第二执行单元，若所述目标网络为所述混合SDN，则用于执行混合SDN网络过渡部署策略，所述混合SDN网络过渡部署策略用于计算新增SDN节点部署位置。
  10. 根据权利要求9所述的混合软件自定义网络的部署装置，所述混合软件自定义网络的部署装置还包括：
    第二建立单元，用于根据所述新增SDN节点部署位置建立新增虚拟本地网络VLAN组，所述新增VLAN组包括VLAN组分解和VLAN组合并。
  11. 根据权利要求8所述的混合软件自定义网络的部署装置，第一建立单元具体用于：
    根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合；
    对所述孤立节点集合中孤立节点以度为优先级进行排序，得到孤立节点中已经建立VLAN组通信的节点集合，然后对所述孤立节点中已经建立VLAN组通信的节点集合之外的节点进行遍历，根据最近两个节点之间链路路径集合中的公共路径数量建立新VLAN组，并将所述新VLAN组添加到所述现阶段VLAN分组集合，得到第一过渡VLAN分组集合；
    确定所述第一过渡VLAN分组集合中VLAN组最多的链路，选择任意两条链路上的VLAN组进行合并，得到第二过渡VLAN分组集合；
    根据所述第二过渡VLAN分组集合确定需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN的传统交换机集合，然后对所述需要重组的VLAN组集合和所述需要重组的传统交换机节点集合进行赋值，并将需要重组的VLAN集合从VLAN组集合中删除，最后对新增SDN交换机和剩余传统交换机进行VLAN组划分，得到重组后的VLAN组集合。
  12. 根据权利要求11所述的混合软件自定义网络的部署装置，第一建立单元具体还用于：
    根据预置的强化学习算法对混合SDN网络中SDN交换机集合Vs中的单个交换机vi按照度进行由大到小排序；
    设置两个变量Valready和Valone，所述Valready表示已经进行VLAN分组的节点集合，所述Valone表示剩余没有VLAN分组的节点集合，所述Valready初始化为空集，所述Valone初始化为V-Vs，所述V表示所有节点；
    对已经排序了优先级的SDN节点进行遍历；
    若所述Valready为空集，则将所述节点vi直接加入到Valready集合中，继续遍历下一个节点；
    若所述Valready只存在一个节点，则通过所述vi计算出所述Valready中唯一节点的最短路径，并建立新VLAN组，将所述新VLAN组的信息加入到VL中，并将所述新VLAN组上的节点从所述Valone去除；
    若所述Valready中存在至少两个节点，则对所述Valready中与所述vi最近的两个节点进行计算并建立两个新VLAN组，将建立的所述两个新VLAN组加入到混合SDN网络中VLAN的集合VL中，并从所述Valone去除所述两个新VLAN组上的节点；
    得到所述VL和Valone两个值，所述VL表示现阶段VLAN组集合，所述Valone表示孤立的节点集合。
  13. 根据权利要求8-12中任一所述的混合软件自定义网络的部署装置，混合软件自定义网络的部署装置还包括：
    优化单元，当所述混合SDN发生故障时，用于对所述混合SDN进行优化。
  14. 根据权利要求13所述的混合软件自定义网络的部署装置，所述优化单元具体用于：
    当所述混合SDN发生故障时，确定所述混合SDN的故障类型，所述故障类型包括VLAN组内通信故障、VLAN组间通信故障和SDN组间通信故障；
    当故障类型为VLAN组内通信故障时，执行VLAN组内故障容错策略，重新计算容错链路；
    当故障类型为VLAN组间通信故障时，执行VLAN组间故障容错策略，重新计算容错链路；
    当故障类型为SDN组间通信故障时，执行SDN组间故障容错策略，重新计算容错链路。
  15. 一种混合软件自定义网络的部署设备，其特征在于，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如下步骤：
    获取目标网络的节点信息；
    根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；
    若所述目标网络不为所述混合SDN，则执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；
    根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
  16. 根据权利要求15所述的混合软件自定义网络的部署设备，所述处理器执行所述计算机程序实现所述根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN之后，还包括以下步骤：
    若所述目标网络为所述混合SDN，则执行混合SDN网络过渡部署策略，所述混合SDN网络过渡部署策略用于计算新增SDN节点部署位置。
  17. 根据权利要求16所述的混合软件自定义网络的部署设备，所述处理器执行所述计算机程序实现所述执行混合SDN网络过渡部署策略之后，还包括以下步骤：
    根据所述新增SDN节点部署位置建立新增虚拟本地网络VLAN组，所述新增VLAN组包括VLAN组分解和VLAN组合并。
  18. 根据权利要求15所述的混合软件自定义网络的部署设备，所述处理器执行所述计算机程序实现所述根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组时，包括以下步骤：
    根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合；
    对所述孤立节点集合中孤立节点以度为优先级进行排序，得到孤立节点中已经建立VLAN组通信的节点集合，然后对所述孤立节点中已经建立VLAN组通信的节点集合之外的节点进行遍历，根据最近两个节点之间链路路径集合中的公共路径数量建立新VLAN组，并将所述新VLAN组添加到所述现阶段VLAN分组集合，得到第一过渡VLAN分组集合；
    确定所述第一过渡VLAN分组集合中VLAN组最多的链路，选择任意两条链路上的VLAN组进行合并，得到第二过渡VLAN分组集合；
    根据所述第二过渡VLAN分组集合确定需要重组的VLAN组集合、需要重组的传统交换机节点集合和已经完成重组VLAN的传统交换机集合，然后对所述需要重组的VLAN组集合和所述需要重组的传统交换机节点集合进行赋值，并将需要重组的VLAN集合从VLAN组集合中删除，最后对新增SDN交换机和剩余传统交换机进行VLAN组划分，得到重组后的VLAN组集合。
  19. 根据权利要求18所述的混合软件自定义网络的部署设备，所述处理器执行所述计算机程序实现所述根据所述初始SDN节点部署位置和预置的强化学习算法进行SDN节点VLAN分组,得到现阶段VLAN分组集合和孤立节点集合时，包括以下步骤：
    根据预置的强化学习算法对混合SDN网络中SDN交换机集合Vs中的单个交换机vi按照度进行由大到小排序；
    设置两个变量Valready和Valone，所述Valready表示已经进行VLAN分组的节点集合，所述Valone表示剩余没有VLAN分组的节点集合，所述Valready初始化为空集，所述Valone初始化为V-Vs，所述V表示所有节点；
    对已经排序了优先级的SDN节点进行遍历；
    若所述Valready为空集，则将所述节点vi直接加入到Valready集合中，继续遍历下一个节点；
    若所述Valready只存在一个节点，则通过所述vi计算出所述Valready中唯一节点的最短路径，并建立新VLAN组，将所述新VLAN组的信息加入到VL中，并将所述新VLAN组上的节点从所述Valone去除；
    若所述Valready中存在至少两个节点，则对所述Valready中与所述vi最近的两个节点进行计算并建立两个新VLAN组，将建立的所述两个新VLAN组加入到混合SDN网络中VLAN的集合VL中，并从所述Valone去除所述两个新VLAN组上的节点；
    得到所述VL和Valone两个值，所述VL表示现阶段VLAN组集合，所述Valone表示孤立的节点集合。
  20. 一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在计算机上运行时，使得计算机执行如下步骤：
    获取目标网络的节点信息；
    根据所述节点信息判断所述目标网络是否为混合软件定义网络SDN；
    若所述目标网络不为所述混合SDN，则执行混合SDN网络初期部署策略，所述混合SDN网络初期部署策略用于计算初始SDN节点部署位置；
    根据所述初始SDN节点部署位置和预置的强化学习算法建立虚拟本地网络VLAN组。
PCT/CN2019/102469 2019-06-17 2019-08-26 混合软件自定义网络的部署方法、装置、设备及存储介质 WO2020252895A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910521265.9 2019-06-17
CN201910521265.9A CN110417576B (zh) 2019-06-17 2019-06-17 混合软件自定义网络的部署方法、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2020252895A1 true WO2020252895A1 (zh) 2020-12-24

Family

ID=68359196

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/102469 WO2020252895A1 (zh) 2019-06-17 2019-08-26 混合软件自定义网络的部署方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN110417576B (zh)
WO (1) WO2020252895A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113965400B (zh) * 2021-11-01 2023-06-30 电子科技大学长三角研究院(衢州) 一种通信网络中流量关键点的确定方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015021248A1 (en) * 2013-08-09 2015-02-12 Nec Laboratories America, Inc. Hybrid network management
CN104954186A (zh) * 2015-06-19 2015-09-30 云南电网有限责任公司信息中心 一种面向应用的sdn网络策略控制方法
CN105471658A (zh) * 2015-12-11 2016-04-06 北京电子科技职业学院 一种sdn网络及其组网方法
CN105553746A (zh) * 2016-01-08 2016-05-04 广州西麦科技股份有限公司 一种基于sdn的配置自动迁移的系统及其方法
CN106060015A (zh) * 2016-05-18 2016-10-26 深圳信息职业技术学院 一种基于sdn的ip源地址验证方法
US20170006067A1 (en) * 2015-07-02 2017-01-05 Vencore Labs, Inc. Hybrid sdn/legacy policy enforcement

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9413646B2 (en) * 2014-08-25 2016-08-09 Nec Corporation Path selection in hybrid networks
CN106850454B (zh) * 2016-04-29 2020-07-14 大连理工大学 一种高流量调整能力的混合sdn网络部署方法
CN107835136A (zh) * 2017-12-14 2018-03-23 中国科学技术大学苏州研究院 现有网络向软件定义网络过渡的交换机部署与路由方法
CN108880909B (zh) * 2018-07-10 2021-04-30 北京邮电大学 一种基于强化学习的网络节能方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015021248A1 (en) * 2013-08-09 2015-02-12 Nec Laboratories America, Inc. Hybrid network management
CN104954186A (zh) * 2015-06-19 2015-09-30 云南电网有限责任公司信息中心 一种面向应用的sdn网络策略控制方法
US20170006067A1 (en) * 2015-07-02 2017-01-05 Vencore Labs, Inc. Hybrid sdn/legacy policy enforcement
CN105471658A (zh) * 2015-12-11 2016-04-06 北京电子科技职业学院 一种sdn网络及其组网方法
CN105553746A (zh) * 2016-01-08 2016-05-04 广州西麦科技股份有限公司 一种基于sdn的配置自动迁移的系统及其方法
CN106060015A (zh) * 2016-05-18 2016-10-26 深圳信息职业技术学院 一种基于sdn的ip源地址验证方法

Also Published As

Publication number Publication date
CN110417576A (zh) 2019-11-05
CN110417576B (zh) 2021-10-12

Similar Documents

Publication Publication Date Title
CN108260169B (zh) 一种基于QoS保障的服务功能链动态部署方法
US6704320B1 (en) Dynamic algorithm for determining a shortest path tree between network nodes
US10547563B2 (en) Efficient message forwarding for distributed resource orchestration
Carpio et al. Balancing the migration of virtual network functions with replications in data centers
US20150350023A1 (en) Data center network architecture
EP2774329B1 (en) Data center network architecture
CN106533966B (zh) 一种网络服务的资源编排方法及装置
CN109412963B (zh) 一种基于流拆分的服务功能链部署方法
US8745265B2 (en) Interconnection fabric connection
EP3328008B1 (en) Deadlock-free routing in lossless multidimensional cartesian topologies with minimal number of virtual buffers
EP3541028B1 (en) Adaptive load-balancing over a multi-point logical interface
WO2016086709A1 (zh) 实现容量规划的方法和装置
US9596137B1 (en) Methods and apparatus for configuration binding in a distributed switch
CN108337179A (zh) 链路流量控制方法及装置
CN108833271A (zh) 一种电网广域控制业务通信路径选择方法及服务器
Wu et al. Efficient and consistent flow update for software defined networks
CN105553855B (zh) 动态调整底层网络生成树拓扑结构的方法及系统
CN108400922B (zh) 虚拟局域网络配置系统与方法及其计算机可读存储介质
US10284457B2 (en) System and method for virtual link trunking
WO2020252895A1 (zh) 混合软件自定义网络的部署方法、装置、设备及存储介质
CN112015518B (zh) 增量式部署sdn环境下实现多虚拟机实时迁移方法及系统
US10855581B2 (en) System and method of computing ethernet routing paths
CN111770070A (zh) 一种基于sdn的安全服务链聚合部署方法
CN113810276A (zh) 一种段路由故障处理方法、装置、电子设备及存储介质
CN108712345B (zh) 基于逐跳计算的分布式负载均衡方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19933948

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19933948

Country of ref document: EP

Kind code of ref document: A1