Auxiliary decision-making method and auxiliary decision-making system for cloud virtual network node deployment

Abstract

The embodiment of the invention provides an auxiliary decision-making method and an auxiliary decision-making system for cloud virtual network node deployment, and the method comprises the steps: obtaining time delay data forwarded by each node in a complete set of network candidate nodes in a cloud, and carrying out the cleaning of all time delay data of each node, carrying out weighted fusion on the cleaned data under different time periods and different network flows to obtain final time delay data of each node matched with an actual scene; according to the final time delay data of each node, a minimum connected graph under preset network traffic is obtained, according to the final time delay data of each node, a shortest path between any two nodes in the minimum connected graph is traversed, an improved variable neighborhood search algorithm is adopted to expand and search a neighborhood of the shortest path, and the neighborhood of the shortest path is searched for. And obtaining the network topological graph meeting the optimal time delay and the degree of each node in the network topological graph. According to the method, the problem that traditional rough deployment may affect the time delay of the virtual network is optimized.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to an auxiliary decision-making method for cloud virtual network node deployment, an auxiliary decision-making system for cloud virtual network node deployment, an electronic device and a computer-readable medium. Background Art

[0002] With the rapid development of cloud computing, customers have higher and higher requirements for network latency. The basic idea of ​​network virtualization in cloud computing is to explore the direction of hardware universalization. The optimization of transmission rate in network virtualization is roughly divided into scheduling optimization and transmission optimization, and transmission optimization can be divided into hardware device optimization and transmission line optimization. The optimization of transmission lines generally only equates the bottleneck with the upper limit of the physical network, while the difference in transmission rate between hundreds or thousands of physical devices is often ignored.

[0003] The current deployment process for cloud virtual networks usually mechanically plans the types and quantities of various virtual network devices in given physical devices, and assumes that the transmission rates between physical devices are completely consistent, but there are actual differences. If deployed in the traditional way, the following two problems may occur:

[0004] (1) The transmission rate between two high-traffic network elements is not high, resulting in a latency bottleneck that limits the network due to deployment issues.

[0005] (2) Some high-traffic network elements are not located in most of the shorter latency paths or are deployed on edge nodes. When they collect and send VM traffic, they cannot provide the shortest average latency. Summary of the invention

[0006] An embodiment of the present invention provides a cloud virtual network node deployment auxiliary decision-making method, a cloud virtual network node deployment auxiliary decision-making system, an electronic device and a computer-readable medium to solve the problem in the prior art that the network delay bottleneck is limited by deployment problems and the shortest delay cannot be achieved.

[0007] An embodiment of the present invention discloses a cloud virtual network node deployment auxiliary decision method, comprising:

[0008] Obtain the delay data forwarded by each node in the entire set of candidate nodes in the cloud network, where the delay data is the data forwarded by each node under each network flow when the cloud virtual network is online by means of a streaming system;

[0009] All delay data of each node are cleaned, and the cleaned data of different time periods and different network traffic are weighted and integrated to obtain the final delay data of each node that matches the actual scenario;

[0010] According to the final delay data of each node, a minimum connectivity graph under a preset network traffic is obtained, wherein the minimum connectivity graph is a star topology structure including delay data within a specified range between any two nodes;

[0011] According to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology that meets the optimal delay and the degree of each node in the network topology.

[0012] Optionally, the delay data of each node forwarding in the entire set of candidate nodes in the cloud network is obtained, including:

[0013] By using the streaming system to simulate the cloud virtual network without load, the first delay data between any two nodes at different time intervals is obtained;

[0014] When the virtual network traffic of the cloud reaches n specified network traffic by using the streaming system simulation, the second delay data between any two nodes at different time intervals is obtained; n is greater than or equal to 1;

[0015] The first delay data and the second delay data constitute the delay data forwarded by each node.

[0016] Optionally, all delay data of each node are cleaned, and weighted fusion is performed on the cleaned data in different time periods and under different network traffic, including:

[0017] For all the delay data of each node, the density-based outlier detection method is used to remove the outliers in the delay data, and the k-nearest neighbor interpolation method is used to process the noise points in the delay data to obtain the preprocessed delay data;

[0018] According to the network traffic after the actual implementation of the cloud virtual network, the pre-processed delay data at different times and network traffic are weighted and fused to obtain the final delay data.

[0019] Optionally, according to the final delay data of each node, a minimum connectivity graph under a preset network traffic is obtained, including:

[0020] According to the preset network traffic information and the final delay data of each node, the Prim algorithm is used to generate the minimum connectivity graph of the current cloud virtual network under the specified termination conditions;

[0021] The preset network traffic information is the network traffic information after the cloud virtual network is actually implemented.

[0022] Optionally, according to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology graph that satisfies the optimal delay, including:

[0023] The termination condition of the improved variable neighborhood search algorithm includes: the termination condition after the neighborhood vibration operator, the local search and the search optimization optimal value are integrated;

[0024] The three neighborhood vibration operators in the improved variable neighborhood search algorithm are used to expand the neighborhood, and the three neighborhood vibration operators include: node replacement, node deletion and node addition;

[0025] For node replacement, if there is more than one node between two nodes, a node in E that does not appear in the minimum connected graph is randomly selected to replace one of the three nodes and form a new path;

[0026] For node deletion, if there is more than one node between two nodes, randomly delete one of the three nodes and form a new path directly connecting the remaining nodes.

[0027] For node addition, for two directly connected nodes in the minimum connectivity, randomly select a node that does not appear in the minimum connectivity graph in E, add it between the two nodes, and form a new path;

[0028] In the iterative process of the shortest path using the improved variable neighborhood search algorithm, a new path is formed with the help of three neighborhood vibration operators to iteratively obtain the optimal delay between each node and obtain the network topology diagram with the optimal delay. E is the set of edges of the cloud virtual network.

[0029] Optionally, according to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology graph that satisfies the optimal delay, further comprising:

[0030] Input T = (V, D) and G = (V, E) to the improved variable neighborhood search algorithm, and the improved variable neighborhood search algorithm outputs G r =(V,E r ), T is the minimum connected graph, G is the cloud virtual network, V is the set of all nodes in the minimum connected graph, D is the set of edges in the minimum connected graph; E is the set of edges in the cloud virtual network, E r is the minimum time delay edge between all nodes in the output network topology, k is the neighborhood vibration operator, iter is the number of iterations, and t is the operation time;

[0031] Initial conditions: iter = 0, t = 0, E r=D;

[0032] Iterative process, k = 1, v1, v2 select random nodes in V, and select any line E' between them; when k ≤ k max When , the neighborhood solution E' is generated according to the kth neighborhood vibration operator; if len(E")>1, then repeat the traversal in E to search for two lines E'' with the same physical nodes and the same len; if the delay of E'' is less than E", then set E'=E'' until there is no E'' that meets the conditions; if the delay of the edge e'' between two nodes in E is higher than E'', then set k=1, E r =E r +E”,E r =E r -e"; otherwise, k = k + 1;

[0033] k=k+1, repeat the above iterative process until t reaches the termination time or the number of iterations iter reaches the maximum value, and G is obtained. r .

[0034] Optionally, it also includes:

[0035] According to the output G r , obtain the degree of each node in the network topology graph, and G r The degree of each node serves as auxiliary decision-making information when the cloud virtual network is actually implemented.

[0036] The embodiment of the present invention also discloses a cloud virtual network node deployment decision-making auxiliary system, including:

[0037] The data simulation acquisition unit is used to obtain the delay data forwarded by each node in the entire set of candidate nodes in the cloud network. The delay data is the data forwarded by each node under each network flow when the cloud virtual network is online by simulating the streaming system; all the delay data of each node are cleaned, and the cleaned data under different time periods and different network flows are weighted and integrated to obtain the final delay data of each node that matches the actual scenario;

[0038] The network topology acquisition unit is used to acquire a minimum connectivity graph under a preset network flow according to the final delay data of each node, wherein the minimum connectivity graph is a star-shaped topology structure containing delay data within a specified range between any two nodes; according to the final delay data of each node, the shortest path between any two nodes in the minimum connectivity graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology graph that meets the optimal delay and the degree of each node in the network topology graph.

[0039] The embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0040] The memory is used to store computer programs;

[0041] The processor is used to implement the method described in the embodiment of the present invention when executing the program stored in the memory.

[0042] The embodiment of the present invention further discloses one or more computer-readable media on which instructions are stored. When executed by one or more processors, the processors are enabled to execute the method described in the embodiment of the present invention.

[0043] The embodiments of the present invention include the following advantages:

[0044] The auxiliary decision-making method provided by the embodiment of the present invention detects the delay data of each node in the cloud virtual network, and after preprocessing, obtains the minimum connectivity graph under the preset network traffic based on the preprocessed delay data, and then combines the delay between the candidate physical nodes, expands and searches the neighborhood of any node in the minimum connectivity graph through the variable neighborhood search algorithm, so as to obtain the network topology diagram with the shortest delay for each point solution; that is, by optimizing the cloud virtual network design scheme, it can give suggestions for the deployment carriers of different network element nodes, eliminating the subtle impact on network performance caused by deployment reasons in the prior art. At the same time, it eliminates the impact of high latency of network element node communication on the overall network performance after rough deployment in the past.

[0045] The method of the embodiment of the present invention solves the limitation of the transmission delay between physical devices on the traffic transmission delay on the cloud virtual network in the prior art. The method of the embodiment of the present invention further optimizes the impact of traditional rough deployment on virtual network latency through a complete solution from simulation, monitoring, optimization calculation to deployment recommendations for network element nodes.

[0046] The auxiliary decision-making system of the embodiment of the present invention can calculate the network topology optimization model of massive physical nodes with the improved variable neighborhood search algorithm as the core support, and generate a topology map containing the route with shorter delay between any two nodes for complex network topology, that is, provide feasible, reliable and efficient network element node deployment suggestions for network element node deployment, and provide solution suggestions under different circumstances due to factors such as actual network conditions, geographical environment and computer room environment. Furthermore, it fully balances the accuracy of the results and the time consumption of calculation, so that the entire system can be used from being available to being easy to use. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1It is a flowchart of the steps of a cloud virtual network node deployment auxiliary decision method provided in an embodiment of the present invention;

[0048] Figure 2 is a flow chart of another auxiliary decision-making method for cloud virtual network node deployment provided in an embodiment of the present invention;

[0049] Figure 3 is an example diagram of the minimum connected graph provided in an embodiment of the present invention;

[0050] Figure 4 is an example diagram of a physical node replacement as a neighborhood vibration operator in an embodiment of the present invention;

[0051] Figure 5 is an example diagram of deleting a physical node as a neighborhood vibration operator in an embodiment of the present invention;

[0052] Figure 6 is an example diagram of a physical node added as a neighborhood vibration operator in an embodiment of the present invention;

[0053] Figure 7 is an example diagram of a network topology diagram of an optimal delay in an embodiment of the present invention;

[0054] Figure 8 It is a structural block diagram of a cloud virtual network node deployment and decision-making assistance system provided in an embodiment of the present invention;

[0055] Fig. 9 is a block diagram of an electronic device provided in an embodiment of the present invention;

[0056] Fig.10 is a schematic diagram of a computer-readable medium provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] In the embodiment of the present invention, the node for acquiring delay data is a node in the entire set of candidate nodes, and each node in the entire set of candidate nodes is a physical device that may serve as a network element node.

[0059] Reference Figure 1 , shows a flowchart of a method for assisting decision-making in cloud virtual network node deployment provided in an embodiment of the present invention, which may specifically include the following steps:

[0060] 101. Obtain delay data forwarded by each node in the entire set of candidate nodes in the cloud network, wherein the delay data is data forwarded by each node under each network traffic when the cloud virtual network is online by means of a streaming system.

[0061] In this embodiment, the first delay data between any two nodes at different time intervals can be obtained with the help of the streaming system to simulate that the cloud virtual network is unloaded; and the second delay data between any two nodes at different time intervals can be obtained with the help of the streaming system to simulate that the cloud virtual network traffic reaches n specified network traffic; n is greater than or equal to 1; at this time, the first delay data and the second delay data constitute the delay data forwarded by each node.

[0062] The virtual network elements in this embodiment are deployed on physical nodes, and the subject of delay detection in step 101 is the network node in the cloud (ie, the physical node), at which point the virtual network elements have not yet been deployed.

[0063] In this embodiment, when the cloud virtual network has no actual load, different network traffic is simulated through the streaming system to collect basic latency data, which helps to establish a performance benchmark for comparing network performance under different load conditions.

[0064] Gradually increase the network load to n specified traffic levels, each of which should include different data transmission rates and types (such as TCP / UDP). At each load level, record the latency data between any two nodes, which can help identify network bottlenecks and performance degradation points.

[0065] In addition to recording the latency data at a specific time point, the average latency over a period of time should also be calculated to better understand the overall performance trend of the network. In addition, the distribution of latency data can also be analyzed, such as statistical indicators such as minimum, maximum, and median, which can reveal the volatility and stability of network latency.

[0066] 102. All delay data of each node are cleaned, and the cleaned data in different time periods and under different network traffic are weightedly fused to obtain the final delay data of each node that matches the actual scenario.

[0067] For example, for all the delay data of each node, the density-based outlier detection method is used to remove outliers in the delay data, and the k-nearest neighbor interpolation method is used to process the noise points in the delay data to obtain the pre-processed delay data; according to the network traffic after the actual implementation of the cloud virtual network, the pre-processed delay data at different times and network traffic are weighted and fused to obtain the final delay data.

[0068] In this embodiment, cleaning and weighted fusion of the delay data of each node in the virtual network is a key step to ensure the accuracy and effectiveness of data analysis.

[0069] When performing data cleaning, the density-based outlier detection method is used to remove outliers in the time delay data. Specifically, the following steps are used:

[0070] Using density-based clustering algorithms such as DBSCAN (density clustering algorithm), outliers with significantly different densities from other data points are automatically identified and marked. This method is particularly suitable for processing mutation points in delay data, and it can effectively remove outliers caused by network failures, hardware problems or other abnormal factors.

[0071] The k-nearest neighbor interpolation method is used to process noise in time-delay data, including:

[0072] For data points marked as outliers, the average value of the k nearest neighboring data points around them is used to replace them. This method can smooth out local noise while retaining the overall trend of the data. At the same time, it is very important to choose a suitable k value. Usually, the best k value needs to be determined through experiments to balance the denoising effect and data authenticity.

[0073] Furthermore, it is necessary to process the missing data of the delay data:

[0074] For missing data in a short period of time, linear interpolation can be used to fill the gaps, that is, linear interpolation is performed based on the time delay values ​​at the time points before and after the missing data;

[0075] For long periods of missing data, the polynomial interpolation method can be used to fit multiple surrounding data points to generate a smoother delay curve.

[0076] In addition, time series models such as ARIMA (Autoregressive Integrated Moving Average) can be used to predict the delay value of the missing part based on historical data.

[0077] When performing weighted fusion, the weight can be set based on time or network traffic. Specifically:

[0078] The latency data in different time periods may have different representativeness. For example, the network traffic during working hours is usually higher than that during non-working hours. Therefore, a higher weight can be given to the data during working hours. That is, a reasonable weight distribution strategy can be set according to actual business needs and network usage patterns.

[0079] The latency data under different network traffic reflects the performance of the network under different load conditions. Different weights can be assigned to the data under different traffic based on the common traffic levels in actual application scenarios. For example, if the network traffic is at a medium level most of the time in actual application scenarios, a higher weight can be given to the latency data under medium traffic.

[0080] 103. According to the final delay data of each node, a minimum connectivity graph under a preset network traffic is obtained, where the minimum connectivity graph is a star topology structure including delay data within a specified range between any two nodes.

[0081] In the specific implementation process, this embodiment can use the Prim algorithm to generate the minimum connected graph of the current cloud virtual network under the specified termination condition based on the preset network traffic information and the final delay data of each node. The preset network traffic information here is the network traffic information after the cloud virtual network is actually implemented.

[0082] Prim's algorithm is a classic algorithm for generating a minimum spanning tree. It starts from an initial node and gradually adds edges until all nodes are included to form a connected graph.

[0083] The specific steps of the algorithm are:

[0084] Select an initial node as the starting point;

[0085] Create a set s, which initially contains only the starting point node;

[0086] Create a priority queue Q to store all nodes that have not joined s and their minimum edge weights (delays) with nodes in s;

[0087] Remove the node u with the smallest edge weight from the priority queue Q and add it to the set s;

[0088] Update the edge weights of the nodes adjacent to u in the priority queue Q. If the delay of an adjacent node v reaching u is less than the currently recorded delay, update the edge weight and parent node of v;

[0089] Repeat the above steps until all nodes are added to the set;

[0090] When all nodes are added to the set s, the algorithm terminates and the generated graph is the minimum connected graph.

[0091] When selecting edges, ensure that the latency of the newly added edge is within the preset range. If the latency of an edge is out of the range, do not select the edge and continue to look for other edges that meet the conditions.

[0092] When generating the minimum connected graph, the preset network traffic information is taken into account to ensure that the generated graph has the best performance under actual traffic.

[0093] 104. According to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology that meets the optimal delay and the degree of each node in the network topology.

[0094] In this embodiment, the termination condition of the improved variable neighborhood search algorithm may include: a termination condition after the neighborhood vibration operator, the local search and the search optimization optimal value are integrated.

[0095] The three neighborhood vibration operators in the improved variable neighborhood search algorithm are used to expand the neighborhood, and the three neighborhood vibration operators include: node replacement, node deletion and node addition;

[0096] For node replacement, if there is more than one node between two nodes, a node that does not appear in the minimum connected graph in E is randomly selected to replace one of the three nodes and form a new path;

[0097] For node deletion, if there is more than one node between two nodes, one of the three nodes is randomly deleted and a new path is formed with the remaining nodes directly connected;

[0098] For node addition, for two directly connected nodes in the minimum connectivity, randomly select a node that does not appear in the minimum connectivity graph in E, add it between the two nodes, and form a new path;

[0099] For example, for a selected pair of nodes A and B, if there is at least one other site between them, a node C is randomly selected from the node set E that is not in the current connected graph, and C is used to replace a node between A and B to form a new path.

[0100] Similarly, for a pair of nodes A and B, if there is at least one other node between them, one of the nodes D is randomly selected for deletion, so that A and B are directly connected, forming a shorter path.

[0101] For two directly connected nodes A and B, a node F is randomly selected from the node set E that is not in the current connected graph and inserted between A and B, which increases the complexity of the network and improves the latency performance.

[0102] In the iterative process of the shortest path using the improved variable neighborhood search algorithm, a new path is formed with the help of three neighborhood vibration operators to iteratively obtain the optimal delay between each node and obtain the network topology diagram with the optimal delay. E is the set of edges of the cloud virtual network.

[0103] Specifically, at the beginning of each iteration, one of the three defined neighborhood vibration operators is selected to be applied, either in a fixed order or randomly;

[0104] Perform corresponding operations (replace, delete or add nodes) according to the selected neighborhood vibration operator to generate candidate solutions; calculate the total delay of the newly generated network topology graph and compare it with the current best solution;

[0105] If the candidate solution is better than the current best solution, the candidate solution is accepted as the new current solution and the search continues. Otherwise, a decision is made based on certain acceptance criteria to decide whether to accept the worse solution to avoid falling into the local optimum.

[0106] The algorithm terminates when the preset maximum number of iterations is reached, the current solution is not improved after several consecutive iterations, or all possible neighborhoods have been explored and no better solution has been found.

[0107] In particular, the search space is considered to be fully explored when the neighborhood shaking operator can no longer produce valid changes (i.e., no new paths can be formed by replacing, deleting, or adding nodes).

[0108] The final network topology should meet the communication requirements between all nodes and have the best latency performance. The degree of each node in the network topology, that is, the number of edges connected to each node, is calculated and output.

[0109] To facilitate understanding of the above step 104, the step may be described in detail as follows:

[0110] Input T = (V, D) and G = (V, E) to the improved variable neighborhood search algorithm, and the improved variable neighborhood search algorithm outputs G r =(V,E r ), T is the minimum connected graph, G is the cloud virtual network, V is the set of all nodes in the minimum connected graph, D is the set of edges in the minimum connected graph; E is the set of edges in the cloud virtual network, E r is the minimum time delay edge between all nodes in the output network topology, k is the neighborhood vibration operator, iter is the number of iterations, and t is the operation time;

[0111] Initial conditions: iter = 0, t = 0, E r =D;

[0112] Iterative process, k = 1, v1, v2 select random nodes in V, and select any line E' between them; when k ≤ k max When , the neighborhood solution E' is generated according to the kth neighborhood vibration operator; if len(E")>1, then repeat the traversal in E to search for two lines E'' with the same physical nodes and the same len; if the delay of E'' is less than E", then set E'=E'' until there is no E'' that meets the conditions; if the delay of the edge e'' between two nodes in E is higher than E'', then set k=1, E r =E r +E”,E r =E r -e"; otherwise, k = k + 1;

[0113] k=k+1, repeat the above iterative process until t reaches the termination time or the number of iterations iter reaches the maximum value, and G is obtained. r .

[0114] Furthermore, in this embodiment, according to the output G r , obtain the degree of each node in the network topology graph, and G r The degree of each node serves as auxiliary decision-making information when the cloud virtual network is actually implemented.

[0115] The method of this embodiment detects the delay data of each node in the cloud virtual network, and after preprocessing, obtains the minimum connectivity graph under the preset network traffic based on the preprocessed delay data, and then combines the delay between the candidate physical nodes, expands and searches the neighborhood of any node in the minimum connectivity graph through the variable neighborhood search algorithm, so as to obtain the network topology diagram with the shortest delay for each point solution; that is, by optimizing the cloud virtual network design scheme, it is possible to give suggestions for the deployment carriers of different network element nodes, eliminating the subtle impact on network performance caused by deployment reasons in the prior art. At the same time, it eliminates the impact of high latency of network element node communication on the overall network performance after rough deployment in the past.

[0116] That is, the above method solves the limitation of the transmission delay between physical devices in the prior art on the traffic transmission delay on the cloud virtual network. The method of the embodiment of the present invention further optimizes the impact of traditional rough deployment on virtual network latency through a complete solution of simulation, monitoring, optimization calculation to deployment recommendations for network element nodes.

[0117] In particular, the method of this embodiment can be used to simulate traffic in multiple scenarios, and by using multiple rounds of data collection and data noise optimization methods based on physical network characteristics, a traffic simulation model that is prepared and fits the usage scenario is provided for the entire optimization step. It also provides a feasible solution for other studies on network traffic simulation and detection.

[0118] like Figure 2 As shown, Figure 2 A flowchart of another cloud virtual network node deployment auxiliary decision method is shown, and the steps are described as follows:

[0119] 201. Simulate the actual traffic data of each node in the entire set of candidate nodes of the cloud virtual network to obtain the latency data of each node collected in multiple rounds under multiple scenarios.

[0120] In this embodiment, in order to realize the forwarding capability between physical nodes under the network traffic scale of the actual online simulated cloud virtual network, the delay between each node of the physical network is collected through network traffic simulation and multi-scenario and multi-round collection, thereby providing accurate delay data for the subsequent network topology optimization processing stage, and at the same time, it can eliminate the pollution of data caused by occasional vibrations and planned turbulence of the physical network during collection.

[0121] Network traffic simulation and multi-scenario multi-round collection are detected by integrating the following constraints to ensure that the collected data can fully describe the actual network conditions. First, simulate the idle time of the cloud virtual network (referred to as the network), that is, test and record the delay between any two network element nodes when the network is unloaded; second, simulate the busy time of the network for data collection, set n pre-made flow scales through the preset traffic scale, and conduct flow between the physical network. By monitoring the physical switching equipment of the flow system, dynamically adjust the flow scale at the physical node, thereby adjusting the scale of the traffic in the entire network, and collect network delay when the expected traffic scale is reached; third, in order to avoid regular interference caused by periodic detection, use uncertain time intervals for detection, and continue long-term monitoring to collect network delay data over a long span of time, so as to observe the trend of network performance over time.

[0122] When conducting detection at uncertain time intervals, the collected data also needs to be cleaned to identify and eliminate data anomalies caused by factors such as occasional network vibrations and planned turbulence.

[0123] Among them, simulating network idle time specifically includes: in the absence of additional load, selecting any two nodes in the network for delay testing, recording the delay data between different node pairs, mainly to maintain the consistency of test conditions, such as test time period, test frequency, etc.; analyzing these delay data to understand the basic delay characteristics of the network under no-load state.

[0124] 202. Data cleaning.

[0125] The missing values ​​and anomaly optimization of the delay data processing of each node obtained in 201 are implemented to achieve optimization tailored to the actual application scenario.

[0126] In this embodiment, a density-based outlier detection method is used for outlier detection. At the same time, after eliminating the pollution of data by human factors such as network maintenance through the identification of deployment personnel, the k-nearest neighbor interpolation method is used to process noise points considering characteristics such as network stability.

[0127] 203. Data fusion.

[0128] In this embodiment, according to the network traffic scale after the actual implementation of this network, the data collected at different times and different traffic scales can be weighted and fused to provide data that fits the actual scenario for subsequent optimization.

[0129] For example, the bandwidth of the physical node construction under the actual network traffic scale after landing is 10M, and the delays of 1M, 2M, 3M, 4M, 5M, 6M, 7M, 8M, 9M, and 10M are collected / detected in multiple rounds and multiple scenarios; then, before planning the network element, it is determined that the scheduled load may be the most in the 80% scenario, so the delay data detected under 8M is set with a higher weight such as 40%, and the delay data weight data of the bandwidth around 8M can be less such as 10%, and the others are not weighted to find an average value. The above-mentioned 1M delay data weight is 20%, 2M delay data weight is 15%, 3M delay data weight is 15%, 4M delay data weight is 10%, 5M delay data weight is 10%, 6M delay data weight is 10%, 7M delay data weight is 6%, 8M delay data weight is 5%, 9M delay data weight is 5%, and 10M delay data weight is 4%. The above is only an example of data fusion, and this embodiment does not limit it and can be adjusted according to actual needs.

[0130] 204. Network topology optimization.

[0131] In order to achieve a star topology constructed by a line with a small delay between any two nodes under a given network traffic scale, any two nodes can be connected. In order to ensure that a better solution can be obtained within the specified time during the network topology optimization process, the minimum connectivity graph of the current network is generated by the Prim algorithm, such as Figure 3 As shown, the shortest path between any two physical nodes is traversed, and the neighborhood of the shortest path is expanded and searched within the neighborhood based on the improved variable neighborhood search algorithm, so as to find the optimal solution between any two physical nodes, and the termination condition after the neighborhood vibration factor, local search and search optimization optimal value are used to balance the calculation time of the optimization algorithm and the credibility of the results, so as to obtain a star topology diagram, as shown in Figure 7 shown.

[0132] Among them, the improved variable neighborhood search algorithm in this embodiment introduces three neighborhood vibration operators (such as physical node deletion, physical node replacement, and physical node addition) to expand the neighborhood, thereby preventing the calculation of the minimum delay line between any two points from being trapped in the local area and failing to obtain a more optimized line, while also taking into account the consideration of calculation time.

[0133] Physical node replacement, if there is more than one node between two nodes (such as edge network element nodes), such as Figure 4(a) shows PM1, PM2, and PM3. A node PM4 is randomly selected from the physical nodes of the non-minimum connected graph as node 3. That is, a physical node other than the existing route point in E is randomly selected to replace node 3 (original PM2). The direct edge between the replaced node and the original node is searched to form a new line, i.e., a new path. Figure 4 (b); otherwise, no action is taken.

[0134] Physical node deletion, if there is more than one node between two edge nodes, such as Figure 5 PM1, PM2, and PM3 shown in (a) randomly select a node (PM2) to delete, and replace the edges between the nodes at both ends of the deleted node with direct connecting edges, as shown in Figure 5 (b); otherwise, no action is taken.

[0135] Physical node addition, currently Figure 6 As shown in (a), in the route between two nodes (PM1, PM2), a physical node (PM4) is added in E except for the existing route points, and a direct edge is added between the newly added node and the original node to form a new route, as shown in Figure 6 (b) as shown.

[0136] 205. Network element node recommendation.

[0137] In order to translate the optimized network topology into a recommendation for the deployment of network element nodes. At this time, the network constructed by the route with the minimum delay between any two nodes is obtained according to the network topology. At this time, the topology is no longer a minimum connected graph, but a network containing a degree of each physical node that is not all 1, such as Figure 7 As shown in FIG. 1 , a node with a larger degree at this time means that the node has more edges with smaller delays connecting other nodes, and it is more suitable as a network element node.

[0138] Specifically, based on the optimized network topology, we can determine which nodes are more suitable for deployment as network element nodes. These nodes usually have higher degrees and more connections with other nodes, and the latency of these connections is also smaller. The following are the specific steps and methods:

[0139] ①Analyze the optimized network topology:

[0140] Calculate the degree of each node: The degree of a node refers to the number of edges directly connected to the node. In the optimized network topology, nodes with larger degrees usually have more low-latency connections.

[0141] Calculate the average delay of each node: For each node, calculate the average delay of the minimum delay path between it and all other nodes, thereby evaluating the importance and performance of the node in the entire network.

[0142] ②Determine the key nodes:

[0143] Select nodes with higher degrees: set a degree threshold, and select nodes with degrees greater than or equal to the threshold as candidate network element nodes; sort the nodes from high to low according to their degrees, and select the top N nodes as preliminary candidate nodes.

[0144] Comprehensively consider the average delay: set a threshold for the average delay and select nodes with an average delay below the threshold; score each node based on the node's degree and average delay. For example, you can use a weighted sum method, where the degree has a higher weight and the average delay has a lower weight.

[0145] The method of this embodiment can provide suggestions for the deployment decisions of various forwarding devices in the physical network and provide solutions for optimizing the lines between physical network regions. It provides a feasible solution for other studies on network traffic simulation and detection. At the same time, it optimizes the delay of physical network transmission between different regions and provides new ideas for network traffic simulation, monitoring and collection of cloud service systems.

[0146] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0147] Reference Figure 8 , shows a structural block diagram of a cloud virtual network node deployment and decision-making assistance system provided in an embodiment of the present invention, which may specifically include the following units:

[0148] The data simulation acquisition unit is used to obtain the delay data forwarded by each node in the entire set of candidate nodes in the cloud network. The delay data is the data forwarded by each node under each network flow when the cloud virtual network is online by simulating the streaming system; all the delay data of each node are cleaned, and the cleaned data under different time periods and different network flows are weighted and integrated to obtain the final delay data of each node that matches the actual scenario;

[0149] The network topology acquisition unit is used to acquire a minimum connectivity graph under a preset network flow according to the final delay data of each node, wherein the minimum connectivity graph is a star-shaped topology structure containing delay data within a specified range between any two nodes; according to the final delay data of each node, the shortest path between any two nodes in the minimum connectivity graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology graph that meets the optimal delay and the degree of each node in the network topology graph.

[0150] The auxiliary decision-making system of this embodiment can calculate the network topology optimization model of massive physical nodes with the improved variable neighborhood search algorithm as the core support, and generate a topology map containing routes with shorter delays between any two nodes for complex network topologies, that is, provide feasible, reliable, and efficient network element node deployment suggestions for network element node deployment, and provide solution suggestions under different circumstances due to factors such as actual network conditions, geographical environment, and computer room environment. Furthermore, it fully balances the accuracy of the results and the time consumption of calculation, so that the entire system can be used from being available to being easy to use.

[0151] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0152] In addition, an embodiment of the present invention further provides an electronic device, such as Fig. 9 As shown, it includes a processor 1301, a communication interface 1302, a memory 1303 and a communication bus 1304, wherein the processor 1301, the communication interface 1302, and the memory 1303 communicate with each other through the communication bus 1304.

[0153] Memory 1303, used for storing computer programs;

[0154] The processor 1301 is used to implement the following steps when executing the program stored in the memory 1303:

[0155] Obtain the delay data forwarded by each node in the cloud network, where the delay data is the data forwarded by each node under each network flow when the cloud virtual network is online by means of a streaming system;

[0156] All delay data of each node are cleaned, and the cleaned data of different time periods and different network traffic are weighted and integrated to obtain the final delay data of each node that matches the actual scenario;

[0157] According to the final delay data of each node, a minimum connectivity graph under a preset network traffic is obtained, wherein the minimum connectivity graph is a star topology structure including delay data within a specified range between any two nodes;

[0158] According to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology that meets the optimal delay and the degree of each node in the network topology.

[0159] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0160] The communication interface is used for communication between the above terminal and other devices.

[0161] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0162] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0163] like Fig.10 As shown, in another embodiment provided by the present invention, a computer-readable storage medium 1401 is also provided, in which instructions are stored. When the computer-readable storage medium 1401 is run on a computer, the computer executes an auxiliary decision-making method for cloud virtual network node deployment described in the above embodiment.

[0164] In another embodiment provided by the present invention, a computer program product containing instructions is also provided. When the computer is run on the computer, the computer executes the auxiliary decision-making method for cloud virtual network node deployment described in the above embodiment.

[0165] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0166] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0167] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0168] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A cloud virtual network node deployment auxiliary decision method, characterized in that: include: Obtain the delay data forwarded by each node in the entire set of candidate nodes in the cloud network. The delay data is the data forwarded by each node under each network flow when the cloud virtual network is online by means of a streaming system; All delay data of each node are cleaned, and the cleaned data of different time periods and different network traffic are weighted and integrated to obtain the final delay data of each node that matches the actual scenario; According to the final delay data of each node, a minimum connectivity graph under a preset network traffic is obtained, wherein the minimum connectivity graph is a star topology structure including delay data within a specified range between any two nodes; According to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology that meets the optimal delay and the degree of each node in the network topology.

2. The method according to claim 1, characterized in that Obtain the forwarding delay data of each node in the entire set of candidate nodes in the cloud network, including: With the help of the streaming system, the first delay data between any two nodes at different time intervals is obtained when the cloud virtual network is unloaded; When the virtual network traffic of the cloud reaches n specified network traffic by means of the streaming system simulation, the second delay data between any two nodes at different time intervals is obtained; n is greater than or equal to 1; The first delay data and the second delay data constitute the delay data forwarded by each node.

3. The method according to claim 1, characterized in that All delay data of each node is cleaned, and the cleaned data of different time periods and different network traffic are weighted and integrated, including: For all the delay data of each node, the density-based outlier detection method is used to remove the outliers in the delay data, and the k-nearest neighbor interpolation method is used to process the noise points in the delay data to obtain the preprocessed delay data; According to the network traffic after the actual implementation of the cloud virtual network, the pre-processed delay data at different times and network traffic are weighted and fused to obtain the final delay data.

4. The method according to claim 1, characterized in that According to the final delay data of each node, the minimum connectivity graph under the preset network traffic is obtained, including: According to the preset network traffic information and the final delay data of each node, the Prim algorithm is used to generate the minimum connectivity graph of the current cloud virtual network under the specified termination conditions; The preset network traffic information is the network traffic information after the cloud virtual network is actually implemented.

5. The method according to claim 1, characterized in that According to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology that satisfies the optimal delay, including: The termination condition of the improved variable neighborhood search algorithm includes: the termination condition after the neighborhood vibration operator, the local search and the search optimization optimal value are integrated; The three neighborhood vibration operators in the improved variable neighborhood search algorithm are used to expand the neighborhood, and the three neighborhood vibration operators include: node replacement, node deletion and node addition; For node replacement, if there is more than one node between two nodes, a node in E that does not appear in the minimum connected graph is randomly selected to replace one of the three nodes and form a new path; For node deletion, if there is more than one node between two nodes, randomly delete one of the three nodes and form a new path directly connecting the remaining nodes. For node addition, for two directly connected nodes in the minimum connectivity, randomly select a node that does not appear in the minimum connectivity graph in E, add it between the two nodes, and form a new path; In the iterative process of the shortest path using the improved variable neighborhood search algorithm, a new path is formed with the help of three neighborhood vibration operators to iteratively obtain the optimal delay between each node and obtain the network topology diagram with the optimal delay. E is the set of edges of the cloud virtual network.

6. The method according to claim 5, characterized in that According to the final delay data of each node, the shortest path between any two nodes in the minimum connected graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology diagram that satisfies the optimal delay, which also includes: Input T = (V, D) and G = (V, E) to the improved variable neighborhood search algorithm, and the improved variable neighborhood search algorithm outputs G r =(V,E r ), T is the minimum connected graph, G is the cloud virtual network, V is the set of all nodes in the minimum connected graph, D is the set of edges in the minimum connected graph; E is the set of edges in the cloud virtual network, E r is the minimum time delay edge between all nodes in the output network topology, k is the neighborhood vibration operator, iter is the number of iterations, and t is the operation time; Initial conditions: iter = 0, t = 0, E r =D; Iterative process, k = 1, v1, v2 select random nodes in V, and select any line E' between them; when k ≤ k max When , the neighborhood solution E' is generated according to the kth neighborhood vibration operator; if len(E")>1, then repeat the traversal in E to search for two lines E'' with the same physical nodes and the same len; if the delay of E'' is less than E", then set E'=E'' until there is no E'' that meets the conditions; if the delay of the edge e'' between two nodes in E is higher than E'', then set k=1, E r =E r +E”,E r =E r -e"; otherwise, k = k + 1; k=k+1, repeat the above iterative process until t reaches the termination time or the number of iterations iter reaches the maximum value, and G is obtained. r .

7. The method according to claim 6, characterized in that Also includes: According to the output G r , obtain the degree of each node in the network topology graph, and G r The degree of each node serves as auxiliary decision-making information when the cloud virtual network is actually implemented.

8. A cloud virtual network node deployment decision support system, characterized in that: include: The data simulation acquisition unit is used to obtain the delay data forwarded by each node in the entire set of candidate nodes in the cloud network. The delay data is the data forwarded by each node under each network flow when the cloud virtual network is online by simulating the streaming system; all the delay data of each node are cleaned, and the cleaned data under different time periods and different network flows are weighted and integrated to obtain the final delay data of each node that matches the actual scenario; The network topology acquisition unit is used to acquire a minimum connectivity graph under a preset network flow according to the final delay data of each node, wherein the minimum connectivity graph is a star-shaped topology structure containing delay data within a specified range between any two nodes; according to the final delay data of each node, the shortest path between any two nodes in the minimum connectivity graph is traversed, and the neighborhood of the shortest path is expanded and searched using an improved variable neighborhood search algorithm to obtain a network topology graph that meets the optimal delay and the degree of each node in the network topology graph.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; The memory is used to store computer programs; The processor is used to implement the method according to any one of claims 1 to 6 when executing the program stored in the memory.

10. A computer-readable medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 6.

Images