SDN dual-end reliability evaluation method and system based on OBDD

Abstract

This invention discloses a two-end reliability assessment method and system for SDN based on OBDD. Leveraging the centralized control and easy programming characteristics of SDN, this method establishes an SDN two-end reliability assessment model. By designing the topology management module and path calculation module within the network model, a two-end minimum path set that meets bandwidth and latency constraints is obtained. A minimum path set-breadth-first search algorithm is used to determine the variable ordering of the BDD. The corresponding OBDD is constructed using the Shannon formula in the OBDD construction module. Finally, the reliability assessment module recursively calculates network reliability from the bottom up, completing the SDN two-end reliability assessment. This method combines methods such as the constrained adjacency endpoint matrix algorithm, the minimum path set-breadth-first search algorithm, and the ordered binary decision diagram (OBDD) to assess connectivity reliability in SDN, demonstrating good scalability.

Description

Technical Field

[0001] The present invention belongs to the field of network reliability assessment, in particular to an SDN two-end reliability assessment method based on a binary ordered decision diagram (OBDD). Background Art

[0002] Network reliability assessment, an effective method for evaluating network reliability within the field of network security assessment, has been widely used in various network systems. Reliability refers to the effectiveness and likelihood of a component, product, or system to complete a specific task without failure within a certain timeframe and under certain conditions. Network reliability, a probabilistic measure, is a network reliability parameter that uses probability to measure network connectivity.

[0003] Currently, methods for calculating network reliability can be divided into two categories: exact algorithms and approximate algorithms. There are five main exact algorithms: state enumeration, inclusion-exclusion principle, disjoint product-sum method, factorization, and state-space decomposition. The state enumeration method enumerates all network states that meet network transmission conditions to calculate network reliability. The inclusion-exclusion principle method expresses network reliability as the union of all minimal path sets, then uses the inclusion-exclusion principle to remove the intersecting components of compatible events and calculate the network reliability of the remaining components. The disjoint product-sum method expresses network reliability as the union of all minimal path sets, then converts this union into the sum of disjoint terms and calculates the network reliability. The factorization method decomposes the network into subnetworks and recursively decomposes the resulting subnetworks until they cannot be decomposed further, obtaining network reliability through continuous iteration. The state-space decomposition method decomposes the network state space into three sets of states and then decomposes the uncertain state set into two sets of states by continuously decomposing the state space. The sum of the probabilities of all acceptable state sets then represents the reliability of the entire network. However, because calculating network reliability is an NP-hard problem, exact algorithms are primarily used to determine the reliability of small networks or networks with specialized topologies. Approximate algorithms are often used to determine the reliability of medium-sized or large networks. Approximate algorithms reduce computational complexity at the expense of reliability accuracy. Common approaches include upper and lower bounds, Monte Carlo methods, graph transformations, ant colony algorithms, genetic algorithms, and neural network algorithms.

[0004] As network size increases, the computational complexity of calculating network reliability using minimum path sets or minimum cut sets can dramatically reduce the efficiency of the algorithm. This has led to the emergence of a new method: binary decision diagrams (BDDs). Binary decision diagrams (BDDs) are considered the most advanced data structure for efficiently storing large Boolean terms. A BDD is a directed acyclic graph consisting of a root node, several intermediate nodes, and two terminal nodes. If different Boolean variables appear in the same order on all paths from the root node, the BDD is called an ordered binary decision diagram (OBDD). A significant advantage of the OBDD approach is that the computational requirements for analysis and processing are linearly proportional to the size of the OBDD graph. OBDD-based methods reduce computational effort and shorten computation time, making them accurate and efficient.

[0005] However, the computational complexity of reliability algorithms based on OBDDs (observable data sets) is highly dependent on the size of the OBDD graph being calculated. Furthermore, the size of the OBDD is determined by the selected variable ordering. Existing methods for finding the optimal variable ordering can be categorized into two categories: static and dynamic. In the static approach, most methods propose an optimal variable ordering before actually constructing the network OBDD. By analyzing various information related to the resulting Boolean functions, the variable order is determined and then the OBDD is constructed. In the dynamic approach, methods primarily reduce the size of the OBDD by adjusting the order of existing variables. Among these, screening algorithms are among the most successful. They perform a local search by repeatedly swapping the positions of adjacent variables to iteratively find the optimal position for all variables. In recent years, many modern intelligent algorithms have been incorporated into OBDD minimization algorithms, such as simulated annealing and genetic algorithms. Genetic algorithm-based methods have produced promising results.

[0006] The size of the OBDD obtained using dynamic techniques is typically smaller than that obtained using heuristic sorting methods, but they are often very time-consuming. Heuristic sorting methods with shorter construction times are often chosen for network reliability assessment. Breadth-first search (BFS) and depth-first search (DFS) are widely used to find variable orders. BFS starts at a source node, visits all associated edges, and then continues from the nodes connected to the edges until all edges and nodes have been visited. Variables are then sorted according to the order in which the edges appear during the traversal. DFS starts at a source node, visits one associated edge, and then continues from the connected nodes until all edges and nodes have been visited. In this case, the variable order is the order in which the edges appear during the traversal. Du proposed a progressive sorting method that performs a different variable ordering on each branch of the OBDD vertex. While this does not guarantee a unique OBDD structure, it has a higher probability of producing a smaller OBDD that shares the same properties as the OBDD generated by the classical algorithm. Using the concept of boundary sets, Pan et al. proposed a new concept of "boundary length" and designed an edge sorting strategy based on boundary length. This strategy aims to maintain the minimum boundary length during variable sorting. This method can produce a good OBDD scale and provide an important reference for edge sorting in specific networks. Based on Hardy's decomposition method, Minh et al. derived a dynamic sorting heuristic method that is applicable to various network structures and can produce smaller OBDDs in irregular network structures. Summary of the Invention

[0007] The present invention aims to address the problems of existing technologies and accurately evaluate the reliability of SDN two-ends. It provides an SDN two-end reliability assessment method based on a binary ordered decision diagram (OBDD). Leveraging the centralized control and easy programming characteristics of SDN, an SDN two-end reliability assessment model is established. By designing the topology management module and path calculation module within the network model, a two-end minimum path set that meets bandwidth and latency constraints is obtained. The minimum path set-breadth-first search algorithm is used to determine the variable ordering of the BDD. The corresponding OBDD is constructed using the Shannon formula in the OBDD construction module. Finally, the reliability assessment module recursively calculates network reliability from the bottom up, completing the SDN two-end reliability assessment. The present method combines methods such as the constrained adjacency endpoint matrix algorithm, the minimum path set-breadth-first search algorithm, and the ordered binary decision diagram (OBDD) to assess connectivity reliability in SDN, demonstrating good scalability.

[0008] The technical solution to achieve the purpose of the present invention is: an SDN two-end reliability evaluation method based on OBDD, the method comprising the following steps:

[0009] Step 1: Based on the network structure of the target software-defined network (SDN), the node and link information in the network is obtained using the protocol of the southbound interface of the SDN controller, and a target SDN two-end reliability evaluation model is constructed, hereinafter referred to as the target network, which includes a topology management module, a path calculation module, an ordered binary decision diagram (OBDD) construction module, and a reliability evaluation module; the topology management module is used to dynamically collect network information, including link information and node information of the current network; the path calculation module is used to calculate the minimum path set of all communication nodes that meet the constraints; the OBDD construction module is used to construct the OBDD model based on the results of the path calculation module and the current topology information; and the reliability evaluation module is used to recursively calculate the two-end reliability in the OBDD model.

[0010] Step 2: The dynamic target network topology information is obtained from the topology management module. The improved constrained adjacency endpoint matrix algorithm is used in the path calculation module to find the minimum path set from the source node to the destination node in the target SDN two-end reliability evaluation model, which satisfies the given bandwidth and delay constraints.

[0011] Step 3: Based on the minimum path set, convert the minimum path into an equivalent Boolean function form;

[0012] Step 4: Based on the minimum path set, the minimum path set-breadth-first search algorithm is used to calculate the variable ranking from the source node to the destination node;

[0013] Step 5: In the OBDD construction module, the Boolean function obtained by converting the minimum path set that meets the constraint requirements is used to construct the OBDD from the source node to the destination node using the variable sorting obtained by the minimum path set-breadth-first search algorithm and the Shannon decomposition formula;

[0014] In step 6, based on the constructed OBDD and the availability probability of the links in the target network, the reliability from the source node to the destination node is calculated in a bottom-up recursive manner in the reliability evaluation module to achieve a two-end reliability evaluation of the target network.

[0015] Furthermore, in step 1, the OpenFlow protocol of the southbound interface of the SDN controller is used to obtain node and link information in the network, and a target SDN two-end reliability evaluation model is constructed, which is described as a nine-tuple:

[0016] G=(V, E, C, T, P, c0, t0, s, d)

[0017] in,

[0018] (1) V = (v1, v2, ..., v n ) represents the node set of the target network, n is the number of nodes in the network, v n is the nth node;

[0019] (2) E=(e1,e2,…,e m ) represents the link set of the target network, m is the number of links in the network, e m is the mth link;

[0020] (3) C={c i |1≤i≤m}, c i Indicates link e i The maximum remaining bandwidth;

[0021] (4)T={t i |1≤i≤m},t i Indicates link e i Current latency;

[0022] (5)P={p i |1≤i≤m},p i Indicates link e i The probability of availability;

[0023] (6) c0, t0 are the given bandwidth constraint and delay constraint respectively;

[0024] (7) s, d represent the source node and destination node in the target network, respectively.

[0025] Furthermore, in step 2, the dynamic target network topology information is obtained from the topology management module to generate the adjacency matrix The terminal matrix R = [r jk ], i, j, k∈[1, n], r∈[1, n-1], where,

[0026]

[0027]

[0028] Matrix B r Contains all minimum paths of length r from node i to node j in the target network. The matrix R represents the end point of each link in the target network.

[0029] The improved constrained adjacency endpoint matrix algorithm is used to calculate the minimum path set that satisfies the given bandwidth and delay constraints. The specific process is as follows:

[0030] Step 2.1: Obtain all link and node information of the target network from the topology management module, and construct a link set E′ that meets the constraints using the given bandwidth constraint c0 and delay constraint t0;

[0031] Step 2.2: Generate the one-step adjacency matrix B1 and the endpoint matrix R based on the current link set E′ and the node set V of the target network model G;

[0032] Step 2.3, according to algorithm B r =B r-1 *R, calculated matrix And record the matrix B r All the elements in row s and column d in the , that is, the minimum path node set L = {l1, l2, ..., l w};

[0033] Step 2.4: Based on the calculated minimum path node set L, the corresponding link relationship between two nodes is used to convert the minimum path node set into the minimum path link set, and the minimum path link set L′={l i |l i ∈L,C i ≥c0,T i ≤t0}, l i represents the i-th minimum path link, C i Indicates link l i Available bandwidth, T i Indicates link l i The maximum delay.

[0034] Furthermore, step 3 converts the minimum path into an equivalent Boolean function form, specifically:

[0035] Let the i-th minimum path l in the minimum path link set L′ be i l i =(e i1 , e i2 ,...,e iu ), where u is the minimum path l i The number of links in e iu is the minimum path l i The u-th link in the i The corresponding equivalent Boolean function is expressed as f i =(e i1 ·e i2 …)·e iu , then the equivalent Boolean function of the minimum path link set L′ is expressed as:

[0036] f=f1+f2+…+f N

[0037] Where N represents the minimum number of paths in the minimum path link set L′.

[0038] Furthermore, the minimum path set-breadth-first search algorithm is used in step 4 to calculate the variable ranking from the source node to the destination node. The specific process includes:

[0039] Step 4.1, initialize an empty variable sorting table List to record variable sorting;

[0040] Step 4.2: For each minimum path l in the minimum path link set L′ i The same layer link is recorded. If at this time l i If the same-layer link has been recorded, its count is increased by one, otherwise its count is recorded as one; the same-layer link refers to the link belonging to the same layer during breadth traversal;

[0041] Step 4.3, use the quick sort algorithm to sort the links in the record in descending order according to their count size;

[0042] Step 4.4: If the variable corresponding to the link with the highest count is not in the List, add its corresponding variable to the List and check the remaining links at the same level in turn until all the variables corresponding to the current links at the same level have been added to the List.

[0043] Step 4.5, check the List. If all the variables corresponding to the links are added to the List, the algorithm ends and the output variable sorting π(x1, x2, ..., x z ), otherwise jump to step 4.2 to select the next level of links for statistical information recording.

[0044] Furthermore, step 5 is specifically as follows:

[0045] The OBDD building block constructs the Boolean function of step 3 as f(x1, x2, ..., x z ),for Both

[0046] Sort the variables π(x1, x2, ..., x) calculated by the minimum path set-breadth-first search algorithm in step 4 z ) According to Shannon decomposition theorem, the constructed Boolean function f(x1, x2, ..., x z ) is processed, and the calculation formula is:

[0047]

[0048] The root node represents the Boolean function f(x1, x2, ..., x z ) itself, two branches are drawn from the root node, representing the variables x in the variable sequence π i The input pattern (x1, ..., x i-1,0,x i+1 ,…,x z ) and (x1,…,x i-1 , 1, x i+1 ,…,x z ) under the Boolean function f(x1,…,x i-1 ,0,x i+1 ,…,x z ) and f(x1,…,x i-1 , 1, x i+1 ,…,x z ).

[0049] Furthermore, in step 6, the reliability is calculated using the following formula based on the link availability probability:

[0050]

[0051] Among them, Pr{f} represents the two-end reliability of the target network, Pr{x i} represents the variable x i The corresponding link availability probability, represents the reliability of the 1-component branch, Represents x i The corresponding link failure probability, Represents the reliability of the 0-component branch.

[0052] Compared with the existing technology, the present invention has the following significant advantages: 1) It utilizes the centralized and easy-to-program features of SDN to collect topology information in the network and model both ends at the application layer, with a clear structure and easy programming; 2) It proposes a new heuristic variable sorting algorithm, the minimum path set link statistical sorting algorithm, which has better performance than the traditional static sorting method; 3) It uses the improved OBDD to accurately calculate the reliability of the SDN two ends for the Boolean function of the minimum path set under given bandwidth and delay constraints, and evaluates the connectivity reliability of the SDN two ends, which has good scalability.

[0053] The present invention is further described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 Flowchart of the SDN reliability evaluation method based on OBDD.

[0055] Figure 2 The following is the architecture diagram of SDN reliability evaluation based on OBDD.

[0056] Figure 3 This is the structural diagram of the SDN reliability evaluation system based on OBDD. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0058] It should be noted that if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in this field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0059] Aiming at the problem that link failure in SDN environment leads to the failure of the entire network communication, the present invention uses the improved adjacency endpoint matrix algorithm, the minimum path set-breadth-first search algorithm, the ordered binary decision diagram (OBDD) and other methods, and considers the factors of link failure and link being constrained by bandwidth and delay. An SDN two-end reliability evaluation model is established and an SDN two-end reliability evaluation method based on OBDD is proposed.

[0060] Combine Figure 1 In one embodiment, the OBDD-based SDN two-end reliability evaluation method specifically includes the following steps:

[0061] Step 1: Analyze the target network structure and use protocols such as OpenFlow on the southbound interface of the SDN controller to obtain information about nodes and links in the network. Build an SDN two-end reliability evaluation model that includes four modules: topology management, path calculation, OBDD construction, and reliability evaluation.

[0062] Step 2: The topology management module of the network model constructed in Step 1 obtains network topology information. Using an improved constrained adjacency endpoint matrix algorithm, the path calculation module calculates the minimum path set from the source node to the destination node of the model. Based on the bandwidth constraint c0 and the delay constraint t0, the minimum paths that do not meet the constraints are deleted to obtain the minimum path set that meets the bandwidth constraint requirements.

[0063] Step 3: Based on the minimum path set that meets the constraint requirements in step 2, convert the minimum path into an equivalent Boolean function form;

[0064] Step 4: Based on the minimum path set that meets the constraints in step 2, use the minimum path set-breadth-first search algorithm to calculate the variable order from the source node to the destination node;

[0065] Step 5: Based on the Boolean function obtained in step 3 and the variable order obtained by the minimum path set-breadth-first search algorithm in step 4, the OBDD from the source node to the destination node is constructed using the Shannon formula in the OBDD construction module;

[0066] In step 6, based on the OBDD constructed in step 5 and the availability probability of each link in the network, a bottom-up recursive method is used in the reliability evaluation module to calculate the reliability from the source node to the destination node, thus realizing the dual-end reliability evaluation of SDN.

[0067] A more detailed description is given below.

[0068] 1. SDN two-end reliability evaluation model

[0069] Definition 1 defines the SDN two-end reliability evaluation model and describes in detail the elements that constitute the SDN two-end reliability evaluation model. The specific definition 1 is as follows:

[0070] Definition 1. The SDN two-end reliability evaluation model G is represented as a nine-tuple:

[0071] G=(V, E, C, T, P, c0, t0, s, d)

[0072] Where V=(v1,v2,…,v n ) represents the node set of the target network, n is the number of nodes in the network, E=(e1,e2,…,e m ) represents the link set of the target network, m is the number of links in the network, C = {c i |1≤i≤m}, c i Indicates link e i The maximum remaining bandwidth, P = {p i |1≤i≤m},p i Indicates link e i The availability probability of a link is c0, t0 are the given bandwidth constraints and delay constraints, respectively, and s and d represent the source and destination nodes of the target network, respectively. The SDN controller obtains all link information in the network by continuously sending LLDP probe packets to connected switches, thereby obtaining the entire network topology connection information. Network data bandwidth can be obtained through the OpenFlow protocol. The bandwidth of a link is determined by the capabilities of the two ports of the switch, that is, the link traffic can be obtained by obtaining the traffic of the port. The structure of the port statistics information obtained by OpenFlow1.3 is shown in Table 1:

[0073] Table 1 OpenFlow port statistics structure

[0074] property describe dpid Data path ID port_no Port number rx_packets Number of received packets tx_packets Number of packets sent rx_bytes Number of bytes received tx_bytes Number of bytes sent duration_sec Port activity time

[0075] The number of packets sent and received, the number of bytes, and the duration of the statistics can be obtained from the message format described in Table 1. Subtracting the number of bytes of statistical messages from two different times and then dividing it by the time difference between the two statistical times can obtain the statistical traffic speed. The maximum remaining bandwidth can be obtained by subtracting the current traffic bandwidth from the maximum bandwidth of the port. The link round-trip delay RTT can also be measured through LLDP packets. After the LLDP packet is sent from the controller to the switch, it is forwarded once and sent back to the controller. By recording the total delay t between the two forwardings, lldp1 and t lldp2 The round-trip delay from the controller to the two switches is t echo1 and t echo2 The round trip delay of the link between the two switches can be calculated as RRT 1，2 ,RTT 1，2 =t lldp1 +t lldp2 -(t echo1 +t echo2 The given bandwidth constraint c0 and delay constraint t0 are set by both nodes according to the situation.

[0076] 2. Obtain the minimum path set subject to bandwidth and delay constraints

[0077] Definition 2. A minimum path is a set of edges that connect communication nodes. If one of the edges fails, the communication nodes are no longer connected. The minimum path set refers to the set of all minimum paths.

[0078] Obtain network topology information from the topology management module of the network reliability assessment model G and generate an adjacency matrix and the terminal matrix R = [r jk ],in,

[0079]

[0080]

[0081] The matrix B1 contains all paths of length 1 in the network, and the matrix R represents the end point of each link in the network. Elements in Obtained by the following calculation formula:

[0082]

[0083] in,

[0084]

[0085] Matrix B1 contains all paths of length 1 in the network. Matrix R represents the endpoints of each link in the network. The improved constrained adjacency endpoint matrix algorithm is used to calculate the minimum set of paths that satisfy certain bandwidth and delay constraints. The specific algorithm is as follows:

[0086] Step 2.1: Obtain all the link and node information of the current network from the topology management module of the network model G. Using the given bandwidth constraint c0 and delay constraint t0, construct a link set E′ that meets the constraint conditions.

[0087] Step 2.2: Generate the one-step adjacency matrix B1 and the endpoint matrix R based on the current link set E′ and the node set V of the network model G;

[0088] Step 2.3, according to algorithm B r =B r-1 *R, you can get the matrix Matrix B r Contains all the minimum paths from node i to node j with a length of r, and the record matrix B r All the elements in row s and column d in (r=1, 2, ..., n-1) are obtained, that is, the minimum path node set L={l1, l2, ..., l w};

[0089] Step 2.4: Based on the calculated minimum path node set L, the corresponding link relationship between two nodes is used to convert the minimum path node set into the minimum path link set, and the minimum path link set L′={l i |l i ∈L,C i ≥c0,T i ≤t0}.

[0090] 3. Convert to equivalent Boolean function form

[0091] The minimum path l in the minimum path set L′ that satisfies the delay and bandwidth constraints i Convert to an equivalent Boolean function form, let l i =(e i1 , e i2 ,...,e iu ), where u is the minimum path l i The number of links in l i The corresponding equivalent Boolean function is expressed as f i (l i )=(ei1 ·e i2 …)·e iu , the equivalent Boolean function corresponding to the minimum path set L′ is expressed as f=f1+f2+…+f N , where N represents the minimum number of paths in the minimum path set L′;

[0092] 4. Use the minimum path set-breadth-first search algorithm to calculate the variable ranking

[0093] The algorithm is based on the traditional heuristic BFS variable sorting method. In the BFS sorting process, variables are sorted only according to the order of exiting the queue, without considering the frequency of occurrence of variables on the same level in the minimum path set. Therefore, the present invention statistically sorts the frequency of occurrence of variables on the same level and sorts the variables in descending order. The steps for variable sorting are as follows:

[0094] Step 4.1, initialize an empty variable sorting table List to record variable sorting;

[0095] Step 4.2: For each minimum path l in the minimum path link set L′ i The same layer link is recorded. If at this time l i If the same-layer link has been recorded, its count is increased by one, otherwise its count is recorded as one; the same-layer link refers to the link belonging to the same layer during breadth traversal;

[0096] Step 4.3: Use the quick sort algorithm to sort the links in the record from largest to smallest according to their counts.

[0097] Step 4.4: If the variable corresponding to the current highest-counted peer link is not in the List, add its corresponding variable to the List, and then check the remaining peer links in turn until all variables corresponding to the current peer link have been added to the List.

[0098] Step 4.5: Check the List. When all the variables corresponding to the links are added to the List, the algorithm ends and the output variable sorting π(x1, x2, ..., x z ), otherwise jump to step 4.2 to select the next level of links for statistical information recording.

[0099] 5. Construction of OBDD

[0100] The OBDD building module in the network reliability evaluation model G uses the generated minimum path set that meets the bandwidth constraint to construct the Boolean function f(x1, x2, ..., x z ),for Both And the variable order π(x1, x2, ..., x z ) According to the following Shannon formula, the constructed Boolean function f(x1, x2, ..., x z ) for processing:

[0101]

[0102] The root node represents the Boolean function f(x1, x2, ..., x z ) itself, two branches are drawn from the root node, representing the variables x in the variable sequence π i The input pattern (x1, ..., x i-1 ,0,x i+1 ,…,x z ) and (x i ,…,x i-1 , 1, x i+1 ,…,x z ) under the Boolean function f(x1,…,x i-1 ,0,x i+1 ,…,x z ) and f(x1,…,x i-1 , 1, x i+1 ,…,x z ). f(x1,…,x i-1 ,0,x i+1 ,…,x z ) can be further Shannon expanded, and they are connected with the 0-component and 1-component obtained by their respective expansions, and the remaining components are recursively Shannon expanded until a 0 node or a 1 node is reached. The two corresponding simplification rules are as follows:

[0103] (1) Deletion rule: For a node u in the OBDD, if u.low = u.high, that is, the 0-component branch and the 1-component branch of the node u point to the same node, then delete the node u and connect the parent node of the node u directly to the node corresponding to u.low;

[0104] (2) Merge rule: For nodes u and v in OBDD, if u.var = v.var, u.low = v.low, and u.high = v.high, that is, the corresponding variables of nodes u and v are the same and the 0-component branch and the 1-component branch point to the same thing, then nodes u and v are merged, node u is deleted, and the parent node of node u is directly connected to node v.

[0105] 6. According to the availability probability of a given link P = {p i |1≤i≤m}, calculate the reliability:

[0106]

[0107] Where Pr{f} represents the desired two-terminal reliability, Pr{x i} represents the variable x i The corresponding link availability probability, represents the reliability of the 1-component branch, Represents the variable x i The corresponding link failure probability, Represents the reliability of a 0-component branch. For leaf nodes, the reliability corresponding to a 0-leaf node is 0, and the reliability corresponding to a 1-leaf node is 1. Using a bottom-up recursive approach, the two-end reliability is calculated to implement two-end reliability assessment for SDN.

[0108] In one embodiment, a system for evaluating SDN two-end reliability based on OBDD is provided, the system comprising:

[0109] The first module is used to obtain node and link information in the network based on the network structure of the target software-defined network (SDN) using the protocol of the southbound interface of the SDN controller, and to build a target SDN two-end reliability evaluation model, hereinafter referred to as the target network, which includes a topology management module, a path calculation module, an ordered binary decision diagram (OBDD) construction module, and a reliability evaluation module. The topology management module is used to dynamically collect network information, including link information and node information of the current network; the path calculation module is used to calculate the minimum set of all paths between communication nodes that meet the constraints; the OBDD construction module is used to build the OBDD model based on the results of the path calculation module and the current topology information; and the reliability evaluation module is used to recursively calculate the two-end reliability in the OBDD model.

[0110] The second module is used to obtain dynamic target network topology information from the topology management module and use the improved constrained adjacency endpoint matrix algorithm in the path calculation module to calculate the minimum path set from the source node to the destination node in the target SDN two-end reliability assessment model, which satisfies the given bandwidth and delay constraints.

[0111] The third module is used to convert the minimum path into an equivalent Boolean function form based on the minimum path set;

[0112] The fourth module is used to calculate the variable ranking from the source node to the destination node based on the minimum path set using the minimum path set-breadth-first search algorithm;

[0113] The fifth module is used to convert the minimum path set that meets the constraint requirements into a Boolean function in the OBDD construction module, and use the variable sorting obtained by the minimum path set-breadth-first search algorithm and the Shannon decomposition formula to construct the OBDD from the source node to the destination node;

[0114] The sixth module is used to calculate the reliability of the source node to the destination node in a bottom-up recursive manner in the reliability evaluation module based on the constructed OBDD and the availability probability of the link in the target network, thereby realizing the two-end reliability evaluation of the target network.

[0115] The specific definitions of the OBDD-based SDN two-terminal reliability assessment system can be found in the definitions of the OBDD-based SDN two-terminal reliability assessment method above and will not be repeated here. Each module in the aforementioned OBDD-based SDN two-terminal reliability assessment system can be implemented in whole or in part through software, hardware, or a combination thereof. Each of these modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0116] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0117] Step 1: Based on the network structure of the target software-defined network (SDN), the node and link information in the network is obtained using the protocol of the southbound interface of the SDN controller, and a target SDN two-end reliability evaluation model is constructed, hereinafter referred to as the target network, which includes a topology management module, a path calculation module, an ordered binary decision diagram (OBDD) construction module, and a reliability evaluation module; the topology management module is used to dynamically collect network information, including link information and node information of the current network; the path calculation module is used to calculate the minimum path set of all communication nodes that meet the constraints; the OBDD construction module is used to construct the OBDD model based on the results of the path calculation module and the current topology information; and the reliability evaluation module is used to recursively calculate the two-end reliability in the OBDD model.

[0118] Step 2: The dynamic target network topology information is obtained from the topology management module. The improved constrained adjacency endpoint matrix algorithm is used in the path calculation module to find the minimum path set from the source node to the destination node in the target SDN two-end reliability evaluation model, which satisfies the given bandwidth and delay constraints.

[0119] Step 3: Based on the minimum path set, convert the minimum path into an equivalent Boolean function form;

[0120] Step 4: Based on the minimum path set, the minimum path set-breadth-first search algorithm is used to calculate the variable ranking from the source node to the destination node;

[0121] Step 5: In the OBDD construction module, the Boolean function obtained by converting the minimum path set that meets the constraint requirements is used to construct the OBDD from the source node to the destination node using the variable sorting obtained by the minimum path set-breadth-first search algorithm and the Shannon decomposition formula;

[0122] In step 6, based on the constructed OBDD and the availability probability of the links in the target network, the reliability from the source node to the destination node is calculated in a bottom-up recursive manner in the reliability evaluation module to achieve a two-end reliability evaluation of the target network.

[0123] For the specific limitations of each step, please refer to the limitations of the OBDD-based SDN two-terminal reliability evaluation method above, which will not be repeated here.

[0124] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0125] Step 1: Based on the network structure of the target software-defined network (SDN), the node and link information in the network is obtained using the protocol of the southbound interface of the SDN controller, and a target SDN two-end reliability evaluation model is constructed, hereinafter referred to as the target network, which includes a topology management module, a path calculation module, an ordered binary decision diagram (OBDD) construction module, and a reliability evaluation module; the topology management module is used to dynamically collect network information, including link information and node information of the current network; the path calculation module is used to calculate the minimum path set of all communication nodes that meet the constraints; the OBDD construction module is used to construct the OBDD model based on the results of the path calculation module and the current topology information; and the reliability evaluation module is used to recursively calculate the two-end reliability in the OBDD model.

[0126] Step 2: The dynamic target network topology information is obtained from the topology management module. The improved constrained adjacency endpoint matrix algorithm is used in the path calculation module to find the minimum path set from the source node to the destination node in the target SDN two-end reliability evaluation model, which satisfies the given bandwidth and delay constraints.

[0127] Step 3: Based on the minimum path set, convert the minimum path into an equivalent Boolean function form;

[0128] Step 4: Based on the minimum path set, the minimum path set-breadth-first search algorithm is used to calculate the variable ranking from the source node to the destination node;

[0129] Step 5: In the OBDD construction module, the Boolean function obtained by converting the minimum path set that meets the constraint requirements is used to construct the OBDD from the source node to the destination node using the variable sorting obtained by the minimum path set-breadth-first search algorithm and the Shannon decomposition formula;

[0130] In step 6, based on the constructed OBDD and the availability probability of the links in the target network, the reliability from the source node to the destination node is calculated in a bottom-up recursive manner in the reliability evaluation module to achieve a two-end reliability evaluation of the target network.

[0131] For the specific limitations of each step, please refer to the limitations of the OBDD-based SDN two-terminal reliability evaluation method above, which will not be repeated here.

[0132] The method provided by the present invention can perform accurate two-end reliability analysis on SDN, effectively reduce memory overhead and time overhead, and can accurately and quickly complete the two-end reliability evaluation of SDN.

[0133] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only illustrative of the principles of the present invention. Without departing from the spirit and scope of the present invention, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for evaluating the reliability of SDN dual-ends based on OBDD, characterized in that: The method comprises the following steps: Step 1: Based on the network structure of the target software-defined network (SDN), which is denoted as the target network, the node and link information in the target network is obtained using the protocol of the southbound interface of the SDN controller, and a target SDN two-end reliability evaluation model is constructed, which includes a topology management module, a path calculation module, an ordered binary decision diagram (OBDD) construction module, and a reliability evaluation module. The topology management module is used to dynamically collect network information, including link information and node information of the current network. The path calculation module is used to calculate the minimum set of paths between communication nodes that meet all constraints. The OBDD construction module is used to build the OBDD model based on the results of the path calculation module and the current topology information. The reliability evaluation module is used to recursively calculate the two-end reliability in the OBDD model. Step 2: The dynamic target network topology information is obtained from the topology management module. The improved constrained adjacency endpoint matrix algorithm is used in the path calculation module to find the minimum path set from the source node to the destination node in the target SDN two-end reliability evaluation model, which satisfies the given bandwidth and delay constraints. Step 3: Based on the minimum path set, convert the minimum path into an equivalent Boolean function form; Step 4: Based on the minimum path set, the minimum path set-breadth-first search algorithm is used to calculate the variable ranking from the source node to the destination node; Step 5: In the OBDD construction module, the Boolean function obtained by converting the minimum path set that meets the constraint requirements is used to construct the OBDD from the source node to the destination node using the variable sorting obtained by the minimum path set-breadth-first search algorithm and the Shannon decomposition formula; In step 6, based on the constructed OBDD and the availability probability of the links in the target network, the reliability from the source node to the destination node is calculated in a bottom-up recursive manner in the reliability evaluation module to achieve a two-end reliability evaluation of the target network.

2. The OBDD-based SDN dual-end reliability evaluation method according to claim 1, characterized in that: In step 1, the OpenFlow protocol of the southbound interface of the SDN controller is used to obtain node and link information in the network and build the target SDN two-end reliability evaluation model, which is described as a nine-tuple: G=(V, E, C, T, P, c0, t0, s, d) in, (1) V = (v1, v2, ..., v n ) represents the node set of the target network, n is the number of nodes in the network, v n is the nth node; (2) E=(e1,e2,…,e m ) represents the link set of the target network, m is the number of links in the network, e m is the mth link; (3) C={c i |1≤i≤m}, c i Indicates link e i The maximum remaining bandwidth; (4)T={t i |1≤i≤m},t i Indicates link e i Current latency; (5)P={p i |1≤i≤m},p i Indicates link e i The probability of availability; (6) c0, t0 are the given bandwidth constraint and delay constraint respectively; (7) s, d represent the source node and destination node in the target network, respectively.

3. The OBDD-based SDN dual-end reliability evaluation method according to claim 1 or 2, characterized in that: In step 2, the dynamic target network topology information is obtained from the topology management module to generate the adjacency matrix The terminal matrix R = [r jk ], i, j, k∈[1, n], r∈[1, n-1], where, Matrix B r Contains all minimum paths of length r from node i to node j in the target network. The matrix R represents the end point of each link in the target network. The improved constrained adjacency endpoint matrix algorithm is used to calculate the minimum path set that satisfies the given bandwidth and delay constraints. The specific process is as follows: Step 2.1: Obtain all link and node information of the target network from the topology management module, and construct a link set E′ that meets the constraints using the given bandwidth constraint c0 and delay constraint t0; Step 2.2: Generate the one-step adjacency matrix B1 and the endpoint matrix R based on the current link set E′ and the node set V of the target network model G; Step 2.3, according to algorithm B r =B r-1 *R, calculated matrix And record the matrix B r All the elements in row s and column d in the , that is, the minimum path node set L = {l1, l2, ..., l w }; Step 2.4: Based on the calculated minimum path node set L, the corresponding link relationship between two nodes is used to convert the minimum path node set into the minimum path link set, and the minimum path link set L′={l i |l i ∈L,C i ≥c0,T i ≤t0}, l i represents the i-th minimum path link, C i Indicates link l i Available bandwidth, T i Indicates link l i The maximum delay.

4. The OBDD-based SDN dual-end reliability evaluation method according to claim 3, characterized in that: Step 3 converts the minimum path into an equivalent Boolean function form, specifically: Let the i-th minimum path l in the minimum path link set L′ be i l i =(e i1 , e i2 ,...,e iu ), where u is the minimum path l i The number of links in e iu is the minimum path l i The u-th link in the i The corresponding equivalent Boolean function is expressed as f i =(e i1 ·e i2 …)·e iu , then the equivalent Boolean function of the minimum path link set L′ is expressed as: f=f1+f2+…+f N Where N represents the minimum number of paths in the minimum path link set L′.

5. The OBDD-based SDN dual-end reliability evaluation method according to claim 4, characterized in that: In step 4, the minimum path set-breadth-first search algorithm is used to calculate the variable ranking from the source node to the destination node. The specific process includes: Step 4.1, initialize an empty variable sorting table List to record variable sorting; Step 4.2: For each minimum path l in the minimum path link set L′ i The same layer link is recorded. If at this time l i If the same-layer link has been recorded, its count is increased by one, otherwise its count is recorded as one; the same-layer link refers to the link belonging to the same layer during breadth traversal; Step 4.3, use the quick sort algorithm to sort the links in the record in descending order according to their count size; Step 4.4: If the variable corresponding to the link with the highest count is not in the List, add its corresponding variable to the List and check the remaining links at the same level in turn until all the variables corresponding to the current links at the same level have been added to the List. Step 4.5, check the List. If all the variables corresponding to the links are added to the List, the algorithm ends and the output variable sorting π(x1, x2, ..., x z ), otherwise jump to step 4.2 to select the next level of links for statistical information recording.

6. The OBDD-based SDN dual-end reliability evaluation method according to claim 5, characterized in that: Step 5 is as follows: The OBDD building block constructs the Boolean function of step 3 as f(x1, x2, ..., x z ),for Both Sort the variables π(x1) calculated by the minimum path set-breadth first search algorithm in step 4 ， x2,…,x z ) According to Shannon decomposition theorem, the constructed Boolean function f(x1, x2, ..., x z ) is processed, and the calculation formula is: The root node represents the Boolean function f(x1, x2, ..., x z ) itself, two branches are drawn from the root node, representing the variables x in the variable sequence π i The input pattern (x1, ..., x i-1 ,0,x i+1 ,…,x z ) and (x1,…,x i-1 , 1, x i+1 ,…,x z ) under the Boolean function f(x1,…,x i-1 ,0,x i+1 ,…,x z ) and f(x 1， …, x i-1 , 1, x i+1 ,…,x z ).

7. The OBDD-based SDN dual-end reliability evaluation method according to claim 6, characterized in that: In step 6, the reliability is calculated based on the link availability probability using the following formula: Among them, Pr{f} represents the two-end reliability of the target network, Pr{x i } represents the variable x i The corresponding link availability probability, represents the reliability of the 1-component branch, Represents x i The corresponding link failure probability, Represents the reliability of the 0-component branch.

8. An OBDD-based SDN dual-end reliability evaluation system based on the method according to any one of claims 1 to 7, characterized in that: The system comprises: The first module is used to obtain node and link information in the network based on the network structure of the target software-defined network (SDN) using the protocol of the southbound interface of the SDN controller, and to build a target SDN two-end reliability evaluation model, hereinafter referred to as the target network, which includes a topology management module, a path calculation module, an ordered binary decision diagram (OBDD) construction module, and a reliability evaluation module. The topology management module is used to dynamically collect network information, including link information and node information of the current network; the path calculation module is used to calculate the minimum set of all paths between communication nodes that meet the constraints; the OBDD construction module is used to build the OBDD model based on the results of the path calculation module and the current topology information; and the reliability evaluation module is used to recursively calculate the two-end reliability in the OBDD model. The second module is used to obtain dynamic target network topology information from the topology management module and use the improved constrained adjacency endpoint matrix algorithm in the path calculation module to calculate the minimum path set from the source node to the destination node in the target SDN two-end reliability assessment model, which satisfies the given bandwidth and delay constraints. The third module is used to convert the minimum path into an equivalent Boolean function form based on the minimum path set; The fourth module is used to calculate the variable ranking from the source node to the destination node based on the minimum path set using the minimum path set-breadth-first search algorithm; The fifth module is used to convert the minimum path set that meets the constraint requirements into a Boolean function in the OBDD construction module, and use the variable sorting obtained by the minimum path set-breadth-first search algorithm and the Shannon decomposition formula to construct the OBDD from the source node to the destination node; The sixth module is used to calculate the reliability of the source node to the destination node in a bottom-up recursive manner in the reliability evaluation module based on the constructed OBDD and the availability probability of the link in the target network, thereby realizing the two-end reliability evaluation of the target network.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images