A Cross-Cloud Resource Migration Method and System

Abstract

The present invention relates to the technical field of cloud computing resource migration, and specifically provides a cross-cloud resource migration method and system, which includes the following steps: Based on the processing capacity of virtual machine instances, the read / write rate of block storage, and the network latency time, proportionally allocate the processing capacity of virtual machine instances, perform weighted averaging on the read / write rate of block storage and the network latency time, integrate them into a two-dimensional matrix, sort and eliminate low-threshold nodes, and generate a preliminary resource optimization dataset. In the present invention, by proportionally allocating the processing capacity of virtual machines and weighted averaging the read / write rate of block storage and network latency, the performance index is processed in a matrix form to improve the accuracy of resource optimization. Combining the performance weight difference and path parameters to optimize the performance dependence expression, dynamically adjust the node transmission efficiency and priority, enhance the elasticity of the transmission path, and monitor and adjust the resource status at the target end in real time to ensure migration stability. This process improves the accuracy of resource integration and business controllability in a multi-cloud environment, meeting complex application requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing resource migration, and particularly to a cross-cloud resource migration method and system. Background Art

[0002] The technical field of cloud computing resource migration mainly studies how to achieve efficient migration and management of resources (including data, application programs, computing tasks, etc.) between different cloud computing environments (such as public clouds, private clouds, and hybrid clouds). This technical field aims to solve problems faced by enterprises in multi-cloud or hybrid cloud environments, such as resource scheduling, data consistency, and system compatibility. By using automated tools, containerization technology, and cross-platform compatibility technology, the resource migration process can be made faster, safer, and more controllable. The application scenarios of this field are extensive, including enterprise business expansion, disaster recovery switching, load balancing, and cost optimization, etc., helping enterprises to achieve efficient integration and utilization of resources in the multi-cloud strategy.

[0003] Among them, the cross-cloud resource migration method is a method and system for efficiently migrating resources between different cloud environments. By designing tools and strategies adapted to multi-cloud environments, it can achieve flexible resource scheduling, seamless migration, and secure data transmission, avoiding business interruptions caused by differences or compatibility issues of cloud service providers. Its purpose is to help enterprises achieve dynamic allocation and optimization of resources in multi-cloud or hybrid cloud environments, improve the flexibility and scalability of business, support digital transformation and continuous innovation and development of business.

[0004] The existing technology has insufficient handling of compatibility issues in different cloud environments in cloud resource migration, lacks in-depth integration of multiple performance indicators such as virtual machine instances, storage read / write rates, and network latency, resulting in insufficient resource scheduling accuracy and a preference process that relies more on empirical judgment. The description of the resource performance dependence path is too single, making it difficult to comprehensively reflect the complex dependence relationships between multiple nodes, and problems such as resource conflicts or path failures occur during the migration process. When the existing technology adjusts the resource migration path, it lacks a dynamic priority adjustment mechanism for high-load nodes, easily causing an increase in data transmission delay and a decline in performance. The monitoring and expansion of resource status also lack real-time performance, often resulting in the inability to detect and correct performance deviation problems in a timely manner, affecting the load balancing and stable operation of nodes after migration. For example, during the cross-cloud migration process, the lagged monitoring of resource status causes some nodes in the migration path to be overloaded, ultimately leading to low migration efficiency and business interruption, restricting the application adaptation ability and migration efficiency of the existing technology in multi-cloud or hybrid cloud environments, and having relatively weak dynamic response ability to business requirements. Summary of the Invention

[0005] The purpose of the present invention is to solve the disadvantages existing in the prior art, and to propose a cross-cloud resource migration method and system.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A cross-cloud resource migration method, comprising the following steps:

[0007] S1: Based on the processing capacity of the virtual machine instance, the block storage read / write rate, and the network latency time, proportionally allocate the processing capacity of the virtual machine instance, perform a weighted average on the block storage read / write rate and the network latency time, integrate them into a two-dimensional matrix, sort and eliminate low-threshold nodes, and generate a preliminary resource optimization dataset;

[0008] S2: Based on the preliminary resource optimization dataset, extract the performance weight difference between the virtual machine and the private subnet, accumulate the distributed file dependency path parameters to the node path, generate a resource performance dependency weight value, and fill it into the mapping relationship graph path set, perform optimal connection screening on the critical path, and generate a resource mapping optimization path set;

[0009] S3: Based on the resource mapping optimization path set, analyze the transmission latency of the virtual machine and the bandwidth occupancy of the load balancer, match the memory free rate, recursively adjust the transmission efficiency distribution value, reallocate the transmission priority for nodes exceeding the critical point, and perform iterative adjustment to generate a resource priority migration sequence list;

[0010] S4: Based on the resource priority migration sequence list, real-time update the running state of the target virtual machine and the network latency, compare the memory usage rate and the read / write rate, perform resource expansion on nodes with difference values exceeding the range, correct the performance offset value of the target node to within the standard range, generate a migration completed node performance status table, and monitor and adjust the node running load to generate a migration completed monitoring result.

[0011] As a further solution of the present invention, the obtaining step of the preliminary resource optimization dataset is specifically as follows:

[0012] S111: Based on the processing capacity of the virtual machine instance, the block storage read / write rate, and the network latency time, proportionally allocate the processing capacity of the virtual machine instance, and generate a processing capacity weight value of the virtual machine instance by comparing the load conditions and required resources of different virtual machine instances;

[0013] S112: Based on the processing capacity weight value of the virtual machine instance, perform a weighted average calculation on the block storage read / write rate and the network latency time, using the formula:

[0014] ;

[0015] Generate a weighted block storage and network latency weight value;

[0016] wherein, is the weighted average value, is the weight coefficient of the block storage read / write rate, is the weight coefficient of network delay time, is the block storage read and write rate, is the network delay time;

[0017] S113: Integrate the weighted block storage and network delay weight values and the processing capacity weight into a two-dimensional matrix, sort according to the weighted results, screen qualified virtual machine instances, remove instances below the threshold, and generate a preliminary resource optimization data set.

[0018] As a further solution of the present invention, the step of obtaining the resource performance dependency weight value is specifically as follows:

[0019] S211: Based on the preliminary resource optimization data set, extract the performance indicator data between the virtual machine and the private subnet, perform a ratio analysis on the network traffic of the virtual machine and the subnet traffic, analyze the difference between the traffic ratio and the throughput and delay parameters, and generate a performance weight difference value between the virtual machine and the private subnet;

[0020] S212: Based on the performance weight difference between the virtual machine and the private subnet, call the path parameter set of the distributed file, identify the node path weight and bandwidth ratio in the path, superimpose the influence of the number of path dependencies, and use the formula:

[0021] ;

[0022] Calculate the node path dependency weight;

[0023] in, represents the node path dependency weight, Indicates the path The path weight of each node, Indicates the path The bandwidth share of each node, represents the cumulative number of path dependencies, Represents the total number of nodes in the path, is the weight adjustment coefficient;

[0024] S213: Based on the node path dependency weight, combined with the performance weight difference value between the virtual machine and the private subnet, the weight data of the two are normalized, the weight dependency of the overall resource performance is analyzed, and the resource performance dependency weight value is generated.

[0025] As a further solution of the present invention, the step of obtaining the resource mapping optimization path set is specifically:

[0026] S221: Based on the resource performance dependence weight value, extract each performance parameter in the resource set. By comparing the performance values between resources, determine the weight difference for each pair of resources, classify and statistically analyze the weight differences, and generate a resource weight mapping node set.

[0027] S222: Based on the resource weight mapping node set, by screening the node and path weight combinations, analyze each path weight cumulative value item by item, measure the dependence strength of the associated resource node combinations, match the measured results with the weight values, and generate a key resource path set.

[0028] S223: Based on the key resource path set, by comparing the node dependence weight relationships within the paths, screen the path segments with the optimal weight values and connect them for processing. Verify the integrity of the connected paths and adjust the structure, and add the optimized paths to the mapping set to obtain a resource mapping optimized path set.

[0029] As a further solution of the present invention, the steps for obtaining the resource priority migration sequence list are specifically as follows:

[0030] S311: Based on the resource mapping optimized path set, extract the virtual machine transmission delay data, calculate the average delay value for each path, and perform coupling analysis with the bandwidth occupancy value of the subnet load balancer to evaluate the transmission delay characteristics. Select the paths that exceed the target threshold, record the bandwidth occupancy ratio, and generate an analysis result of transmission delay and bandwidth occupancy.

[0031] S312: Based on the analysis result of transmission delay and bandwidth occupancy, evaluate the relationship between path transmission load and memory occupancy, recursively adjust the transmission efficiency distribution value, and use the formula:

[0032] ;

[0033] Generate a distribution adjustment result of path transmission efficiency;

[0034] Among them, represents the transmission efficiency distribution value, is the memory free rate of the th path, is the path bandwidth occupancy weight, is the path load value, is the adjustment coefficient, is the total weight of path transmission priorities, represents the total number of paths;

[0035] S313: Based on the distribution adjustment result of path transmission efficiency, reassign priorities to the path nodes with transmission efficiency exceeding the critical point, recursively adjust the node weights item by item according to the priorities, sort according to node criticality and load priority, determine the migration priority, and generate a resource priority migration sequence list.

[0036] As a further solution of the present invention, the step of obtaining the performance status table of the migrated completed nodes is specifically as follows:

[0037] S411: Based on the resource priority migration sequence table, monitor the running status of the virtual machine at the target end in real time, collect the network latency value and CPU occupancy rate of the target virtual machine, record the change data in combination with the node resource metrics, normalize the network latency and CPU occupancy rate, and obtain the real-time performance status of the target virtual machine;

[0038] S412: Based on the real-time performance status of the target virtual machine, compare the memory usage rate and the read / write rate, judge the change of the average memory occupancy through the difference value, calculate the performance deviation in combination with the fluctuation range of the read / write rate, correct the difference value, adjust the resource expansion parameter, and adopt the formula:

[0039] ;

[0040] Generate the performance deviation correction result of the target virtual machine;

[0041] Wherein, represents the performance deviation value, is the current memory usage rate, is the target memory usage rate, is the current read / write rate, is the target read / write rate;

[0042] S413: Based on the performance deviation correction result of the target virtual machine, execute resource expansion to adjust the nodes beyond the standard range, correct the real-time performance status of the nodes after expansion, normalize the adjusted performance value to the standard range, summarize the node performance correction status, and generate the performance status table of the migrated completed nodes.

[0043] As a further solution of the present invention, the step of obtaining the monitoring result of the migrated completed is specifically as follows:

[0044] S421: Based on the performance status table of the migrated completed nodes, extract the node performance parameter data, read and screen and sort out the operation metrics node by node, perform time series analysis on the CPU occupancy rate, memory usage rate, and network traffic value, and generate the node performance status record table;

[0045] S422: Based on the node performance status record table, analyze the node load occupancy situation, mark and classify the nodes exceeding the load threshold, transfer the tasks of the high-load nodes to the low-load nodes, evaluate the performance metrics of the low-load nodes, verify the adjusted node load parameters, and obtain the node load adjustment result;

[0046] S423: Based on the node load adjustment result, monitor the adjusted nodes in real time, capture the running load and performance parameter data, compare item by item the load stability and performance, record the dynamic information after node migration, summarize the monitoring data, and generate the monitoring result after migration completion.

[0047] A cross-cloud resource migration system, which is used to execute the above cross-cloud resource migration method, and the system includes:

[0048] The node screening module calculates the weighted value of the block storage read / write rate and the network latency time based on the processing capacity of the virtual machine instance, the block storage read / write rate, and the network latency time, combines the three parameters in proportion for evaluation, and sorts the nodes to screen the nodes with scores greater than the set threshold to obtain a high-score node set;

[0049] The path dependency module calculates the weights of the distributed file dependency path node attributes based on the high-score node set, accumulates the file path length to the node weight value, and screens the paths by combining the performance differences between the virtual machine and the subnet to generate the path weight optimization result;

[0050] The transmission priority module analyzes the transmission latency and bandwidth occupancy ratio of the nodes based on the path weight optimization result, re-analyzes the transmission efficiency distribution by combining the node memory free rate, and adjusts the priority of the nodes with occupancy rates exceeding the threshold to generate an optimized transmission path sequence;

[0051] The migration performance monitoring module monitors the running state of the target virtual machine and the network latency in real time based on the optimized transmission path sequence, compares the difference value between the node memory usage rate and the block storage read / write rate, corrects the performance deviation and records the performance status, and generates the monitoring result after migration completion.

[0052] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0053] In the present invention, by proportionally allocating the processing capabilities of virtual machines, the weighted average block storage read / write rate, and the network latency time, the performance indicators of complex resources are uniformly processed in a matrix manner, improving the accuracy and efficiency of resource optimization. Combining the performance weight differences and the cumulative calculation of path parameters, the expression of performance dependencies is optimized, effectively avoiding the problem of parameter ambiguity in resource scheduling, and strengthening the visualization and operability of dependency relationships. By dynamically adjusting the transmission efficiency distribution and priority reallocation of nodes, the elastic control ability of the transmission path is enhanced, solving the problem of resource transmission bottlenecks at high-load nodes. For the real-time status monitoring and dynamic expansion adjustment of target-end resources, the node performance deviation and operation load deviation are controlled within a reasonable range, ensuring the stability and data consistency of the resource migration process. This optimization process based on dynamic calculation and step-by-step recursive adjustment realizes the flexible scheduling and efficient migration of resources in a complex multi-cloud environment, while significantly improving the accuracy of resource integration and the controllability of business operations. The overall solution has significant gains in performance optimization, resource matching, and migration reliability compared with the prior art, and can meet the dynamic requirements of complex application scenarios in a multi-cloud environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a schematic diagram of the workflow of the present invention;

[0055] Figure 2 is a flowchart of the preliminary resource optimization data set in the present invention;

[0056] Figure 3 is a flowchart of the resource performance dependency weight value in the present invention;

[0057] Figure 4 is a flowchart of the resource mapping optimization path set in the present invention;

[0058] Figure 5 is a flowchart of the resource priority migration sequence list in the present invention;

[0059] Figure 6 is a flowchart of the performance status table of the migrated completed nodes in the present invention;

[0060] Figure 7 is a flowchart of the monitoring result of the migrated completed in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0062] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0063] Embodiment 1: Please refer to Figure 1 , the present invention provides a technical solution: a cross-cloud resource migration method, including the following steps:

[0064] S1: Based on the processing capacity of the virtual machine instance, the block storage read and write rate, and the network latency time, proportionally allocate the processing capacity of the virtual machine instance, perform a weighted average on the block storage read and write rate and the network latency time, integrate them into a two-dimensional matrix, sort and eliminate low-threshold nodes, and generate a preliminary resource preference dataset;

[0065] S2: Based on the preliminary resource preference dataset, extract the performance weight difference between the virtual machine and the private subnet, accumulate the distributed file dependency path parameters to the node path, generate a resource performance dependency weight value, and fill it into the mapping relationship graph path set, perform an optimal connection screening on the critical path, and generate a resource mapping optimization path set;

[0066] S3: Based on the resource mapping optimization path set, analyze the transmission latency of the virtual machine and the bandwidth occupancy of the load balancer, match the memory idle rate, recursively adjust the transmission efficiency distribution value, reallocate the transmission priority for nodes exceeding the critical point, and iteratively adjust to generate a resource priority migration sequence list;

[0067] S4: Based on the resource priority migration sequence list, real-time update the running status and network latency of the target virtual machine, compare the memory usage rate and the read and write rate, perform resource expansion on nodes where the difference value exceeds the range, correct the performance offset value of the target node to within the standard range, generate a migration completion node performance status table, and monitor and adjust the node running load to generate a migration completion monitoring result.

[0068] The preliminary resource optimization dataset includes processing capacity metrics, storage weighting metrics, and network latency metrics. The resource performance dependency weight values include node performance differences and dependency path weighting. The resource mapping optimization path set includes mapping relationship graph filling and path optimization results. The resource priority migration sequence list includes transmission efficiency adjustment and transmission priority setting. The performance status table of migrated nodes includes performance status update and performance standard calibration. The monitoring results after migration include monitoring data analysis and load adjustment feedback.

[0069] Please refer to Figure 2 , and the specific steps for obtaining the preliminary resource optimization dataset are as follows:

[0070] S111: Based on the processing capacity, block storage read / write rate, and network latency time of virtual machine instances, proportionally allocate the processing capacity of virtual machine instances. By comparing the load conditions and required resources of different virtual machine instances, generate the processing capacity weight value of virtual machine instances;

[0071] First, it is necessary to measure the actual processing capacity of each virtual machine instance in detail, involving monitoring parameters such as the CPU utilization rate, memory occupancy rate, disk I / O rate, and network bandwidth usage of the virtual machine. Through the comprehensive evaluation of indicators, the total processing capacity value of each virtual machine instance can be obtained. The measurement method can use resource monitoring tools (such as the top command or vmstat tool) to obtain the resource occupancy situation in real time and record it. Obtain the total processing capacity of all virtual machine instances. By calculating the ratio of the processing capacity value of each virtual machine to the total processing capacity value, obtain the processing capacity ratio of each virtual machine instance. For accuracy, the actual resource usage situation of the virtual machine instance needs to be considered during calculation, and multiple samplings are required. The processing capacity ratio of each virtual machine instance will be used as the basis for the proportion of this instance in resource allocation. It is also necessary to set a certain threshold to identify instances with weak processing capacity or overload, ensure the balance and stability of resource allocation, and will be adjusted according to the priority or task characteristics of the virtual machine instance to ultimately ensure the optimal configuration of resources. Finally, generate the processing capacity weight value of the virtual machine instance, which will be used to guide the dynamic scheduling of virtual machine resources and the load balancing strategy.

[0072] S112: Based on the processing capacity weight value of virtual machine instances, perform a weighted average calculation on the block storage read / write rate and network latency time, using the formula:

[0073] ;

[0074] Generate the weighted block storage and network latency weight values;

[0075] Among them, is the weighted average value, is the weight coefficient of the block storage read / write rate, is the weight coefficient of the network latency time, is the read / write rate of the block storage, is the network latency time;

[0076] The benefit of the formula is that by taking the weighted average of the read / write rate of the block storage and the network latency time, the impacts of these two parameters on the virtual machine performance can be considered simultaneously, ensuring the reasonable allocation of resources;

[0077] First, it is necessary to obtain the read / write rate of the block storage and the network latency time of the actual values. The read / write rate of the block storage can be obtained through hardware monitoring tools or the APIs provided by the cloud platform. The network latency time is measured through network testing tools (such as the ping command), or obtained through the network monitoring interface of the cloud platform. After monitoring and calculation, the read / write rate of the block storage , the network latency time are obtained, and the weight coefficients and are set. These two coefficients are the results obtained through virtual machine performance testing and historical data analysis;

[0078] According to the formula, substitute the values and perform the calculation: ;

[0079] First, calculate the part of the read / write rate of the block storage: ;

[0080] Then, calculate the network latency part: ;

[0081] Next, add these two parts and divide by the sum of the weight coefficients: ;

[0082] The result shows that the comprehensive weight value obtained through weighted calculation is 28002.1213, representing the comprehensive influence degree of the block storage and the network latency. This weight value will be used for priority sorting and resource scheduling in the subsequent virtual machine resource allocation to ensure the optimal configuration and utilization of resources.

[0083] S113: Integrate the weighted weight values of the block storage and the network latency with the processing capacity weight into a two-dimensional matrix, sort according to the weighted results, screen out the virtual machine instances that meet the conditions, eliminate the instances below the threshold, and generate a preliminary resource optimization dataset;

[0084] First, for each virtual machine instance, a two-dimensional matrix is generated based on its processing capacity weight and the weighted storage and network latency weights. Each row of the matrix represents the relevant data of a virtual machine instance, where the first column represents the processing capacity weight of the virtual machine instance, and the second column represents the corresponding weighted value of block storage and network latency. The processing capacity weight of the virtual machine instance is obtained through the aforementioned calculation, and the weighted value of block storage and network latency is obtained by weighting with the weights in the formula. After the matrix is generated, each virtual machine instance needs to be sorted in descending order according to the comprehensive weight value to ensure that resource allocation preferentially meets the needs of high-performance virtual machine instances. After sorting, according to the set threshold, virtual machine instances with stronger processing capabilities and excellent performance are screened, and more resources are allocated to them. When screening, the threshold can be adjusted according to the priority, task nature, and resource occupancy of the virtual machine instance to achieve a more flexible resource scheduling effect. Based on the screening results, a preliminary resource optimization dataset is generated. The dataset will include the processing capacity weight and the weighted storage and network latency weight of each virtual machine instance, as well as resource allocation suggestions, which serve as the basis for subsequent virtual machine resource scheduling decisions.

[0085] Please refer to Figure 3 , the steps for obtaining the resource performance dependence weight value are specifically as follows:

[0086] S211: Based on the preliminary resource optimization dataset, extract the performance index data between the virtual machine and the private subnet, perform a ratio analysis on the virtual machine network traffic and the subnet traffic, analyze the differences between the traffic ratio and the throughput and latency parameters, and generate the performance weight difference value between the virtual machine and the private subnet;

[0087] Collect the input traffic and output traffic of the virtual machine network traffic separately for each time period, call the network performance monitoring tool to record the throughput data, and further obtain the throughput stability value for each time period by analyzing the throughput volatility. For the acquisition of the latency parameter, through multi-point monitoring of the network latency, the latency value is calculated in segments according to the time segment and its segment mean is extracted. The throughput and latency parameters are processed in a normalized manner and unified to the same scale. Then, according to the relative changes between the throughput and the latency parameter, calculate their difference value matrix. Select the main difference parameters in the difference value matrix for weight allocation, use the allocated weights to weight each difference, and accumulate the weighted results of each difference item by item to form the performance weight difference value between the virtual machine and the private subnet. By further comparing the performance difference change trends of the virtual machine in different time segments.

[0088] S212: Based on the performance weight difference value between the virtual machine and the private subnet, call the path parameter set of the distributed file, identify the node path weight and bandwidth occupancy ratio in the path, and superimpose the influence of the path dependence times. Use the formula:

[0089] ;

[0090] Calculate the node path dependency weight;

[0091] Among them, represents the node path dependency weight, represents the path weight of the th node in the path, represents the bandwidth occupancy ratio of the th node in the path, represents the cumulative number of path dependencies, represents the total number of nodes in the path, is the weight adjustment coefficient;

[0092] The benefit of the formula is that by introducing the path dependency count and the weight adjustment coefficient , it can quantify the dependency relationship between path nodes, and combined with the bandwidth occupancy ratio reflect the contribution degree of the node to the overall path performance, thereby improving the flexibility and applicability of path weight calculation;

[0093] Suppose the number of path nodes , the path weights of each node , , , , , the node bandwidth occupancy ratio , , , , , the path dependency count , , , , , the weight adjustment coefficient , the total number of path nodes ;

[0094] The first step: Calculate the comprehensive weight of each node and substitute it into the formula:

[0095] Node 1: ;

[0096] Node 2: ;

[0097] Nodes 3, 4, and 5 are calculated successively to get , , ;

[0098] The second step: Accumulate the comprehensive weights of all nodes: ;

[0099] Step 3: Calculate the node path dependency weight: ;

[0100] The result shows that the node path dependency weight value is 1.6298, which reflects the comprehensive influence degree of the nodes in the path in terms of bandwidth occupancy ratio, path dependency times, etc. This value can be used for further analysis of path performance.

[0101] S213: Based on the node path dependency weight, combined with the performance weight difference value between the virtual machine and the private subnet, perform a normalization operation on the weight data of both, analyze the weight dependency of the overall resource performance, and generate a resource performance dependency weight value;

[0102] Perform a normalization process on the node path weight value and the performance weight difference value between the virtual machine and the private subnet. Represent the two weight values in a matrix according to the weight sum and difference dimensions. By comparing the corresponding weight difference distribution in the matrix, calculate the normalized value of the weight distribution of the virtual machine and the private subnet in each path node, accumulate the normalized values to generate a resource performance dependency weight value, classify the degree of dependence of the resource performance through the above weight value, and organize the resource performance dependency weight value into the final structured data.

[0103] Please refer to Figure 4 , the specific steps for obtaining the resource mapping optimization path set are as follows:

[0104] S221: Based on the resource performance dependency weight value, extract each performance parameter in the resource set. By comparing the performance values between resources, determine the weight difference between each pair of resources, and classify and count the weight differences to generate a resource weight mapping node set;

[0105] When extracting each performance parameter in the resource set, group them according to the category of the resource performance parameter. The values of each group of performance parameters are extracted by calling the resource performance monitoring interface one by one to generate a performance value set in sequence. In the process of comparing the performance values between resources, the weight value of the dependent performance parameter is used as the comparison basis, and the weight difference between each pair of performance parameters between resources is calculated in sequence. The calculation method is completed by setting the relative weight value formula for each group of performance parameters, and the performance difference results are recorded. In the classification and statistics, resources with similar performance difference values are grouped into the same category to generate a weight difference statistical matrix. Each row represents the range of performance parameters of a certain type of resource, and each column records the weight difference results of the resources belonging to this category. Through the results recorded in the difference statistical matrix, classify the weight differences between resources to generate a weight mapping node set. Each node in the node set is used to represent the mapping relationship of the dependency weight between resources, and number the weight connection points between resources, which can be used for subsequent path screening and optimization processes and support the weight correlation analysis between resources.

[0106] S222: Mapping a node set based on resource weights, by screening node and path weight combinations, analyzing the accumulated path weight values one by one, calculating the dependency strength of the associated resource node combination, matching the calculation results with the weight values, and generating a key resource path set;

[0107] By extracting the node weight values of the resource weight mapping node set, the connection path of each node is analyzed in turn, and the dependency strength analysis model is set according to the cumulative value of the connection weight between nodes. When screening the node and path combination, nodes with larger weight values are given priority, and in the combined path, path segments with similar weights are gradually screened. The calculation basis of the path cumulative value is the item-by-item superposition result of the connection node weights. When calculating the dependency strength, the cumulative total value and average value of the node weights of the path are analyzed one by one. For the path segments with significant weight differences between nodes, the weight ratios of the associated resource nodes are further calculated to determine the dependency strength of the node combination. After mapping the measurement results to the weight value interval, a key resource path set is generated according to the weight value. The resource path with higher weight is taken as the key path given priority. The paths in the set are sorted from high to low according to the weight value, providing basic support for subsequent path optimization.

[0108] S223: Based on the key resource path set, by comparing the dependency weight relationship of nodes in the path, the path segment with the optimal weight value is selected and connected, the integrity of the connected path is verified and the structure is adjusted, the optimized path is added to the mapping set, and the resource mapping optimized path set is obtained;

[0109] When comparing the dependency weight relationship of nodes in the path, each path in the key resource path set is extracted one by one. By splitting the path, the weight value of each node in the path is recorded, and the weight difference of adjacent nodes in the path segment is calculated. When screening the optimal path segment, the path segment with smaller weight difference and higher cumulative value is given priority. A more stable path combination is formed through connection processing. When verifying the integrity of the connected path, the total weight value of the path and the uniformity of the node weight distribution are calculated to ensure that the path structure can meet the integrity requirements of resource dependency relationships in cross-cloud resource migration. In the process of adjusting the path structure, the path segments with large weight differences are split and recombined until the path weight distribution tends to be uniform. After the optimized path is added to the mapping set, the resource mapping optimized path set is obtained, which will be used as the core dependency structure of cross-cloud resource migration to guide the scheduling and path planning of resource migration and achieve efficient deployment and migration of resources.

[0110] See also Figure 5 , the specific steps for obtaining the resource priority migration sequence table are:

[0111] S311: Based on the optimized path set of resource mapping, extract the virtual machine transmission delay data, calculate the average delay value of each path, and conduct a coupling analysis with the bandwidth occupancy value of the subnet load balancer to evaluate the transmission delay characteristics. Select the paths that exceed the target threshold, record the bandwidth occupancy ratio, and generate the analysis result of transmission delay and bandwidth occupancy;

[0112] Collect the transmission delay data from the virtual machine to the subnet. Record the delay time of each path through multiple measurements to obtain a continuous data set of delay times. Divide the data into multiple time periods and calculate the average delay value of each path within each time period. At the same time, for the delay data of each path, further calculate its fluctuation amplitude, identify the paths with large fluctuations as the target paths to be further analyzed. Conduct a bandwidth occupancy measurement on the target paths, record the bandwidth value allocated by the subnet load balancer for the paths and its actual usage ratio. According to the ratio difference between the actual bandwidth occupancy and the allocated bandwidth value, combined with the delay time data, evaluate the transmission efficiency of the virtual machine on different paths, calculate the transmission delay characteristic values of each path, and generate the analysis result of transmission delay and bandwidth occupancy based on the delay characteristic values and bandwidth occupancy ratio, providing input data and evaluation basis for subsequent path optimization and priority allocation.

[0113] S312: Based on the analysis result of transmission delay and bandwidth occupancy, evaluate the relationship between path transmission load and memory occupancy, recursively adjust the transmission efficiency distribution value, using the formula:

[0114] ;

[0115] Generate the distribution adjustment result of path transmission efficiency;

[0116] Among them, represents the transmission efficiency distribution value, is the memory free rate of the th path, is the path bandwidth occupancy weight, is the path load value, is the adjustment coefficient, is the total weight sum of path transmission priorities, represents the total number of paths;

[0117] The advantage of the formula is that by combining the memory free rate and the path bandwidth occupancy weight, it refines the analysis of resource utilization among paths, and introduces the absolute value of the difference in path load values. Combined with the square root operation, it weakens the influence of abnormal path loads on the calculation results, making the evaluation result of the transmission efficiency distribution smoother;

[0118] Assume the total number of paths , the memory free rate of the first path , the memory free rate of the second path and the memory free rate of the third path and the memory free rate of the fourth path and the corresponding bandwidth occupancy weight is , , , , and the path load value is , , , , the adjustment coefficient , the total path priority ;

[0119] Calculate the comprehensive value of each path:

[0120] Path 1: ;

[0121] Path 2: ;

[0122] Path 3: ;

[0123] Path 4: ;

[0124] Accumulate the comprehensive values and calculate the transmission efficiency distribution value: ;

[0125] This result indicates that the transmission efficiency distribution value is , which reflects the comprehensive transmission performance of the current path under the conditions of bandwidth utilization, memory free rate, and load distribution. This value can be used for the next adjustment of transmission priorities.

[0126] S313: Based on the distribution adjustment result of path transmission efficiency, reassign priorities to the path nodes whose transmission efficiency exceeds the critical point, recursively adjust the node weights item by item according to the priorities, sort according to node criticality and load priority, determine the migration priority, and generate a resource priority migration sequence list;

[0127] Select the path nodes with transmission efficiency lower than the distribution threshold, calculate the difference between the load of the low-efficiency path and the load of the remaining nodes item by item, reassign the transmission priority of the low-efficiency path according to the load difference value, and at the same time adjust the path weights of the nodes according to the reassigned priorities. By gradually iterating to optimize the load distribution and priority distribution, finally sort all paths by priority and output the migration priority sequence, prioritize the paths with higher transmission efficiency distribution at the front of the sequence list, and generate a resource priority migration sequence list according to the sorting result.

[0128] Please refer to Figure 6, the steps for obtaining the performance status table of the migration completed nodes are specifically as follows:

[0129] S411: Based on the resource priority migration sequence table, monitor the running status of the virtual machines at the target end in real time, collect the network latency value and CPU occupancy rate of the target virtual machine, combine the node resource metrics to record the change data, normalize the network latency and CPU occupancy rate, and obtain the real-time performance status of the target virtual machine;

[0130] Collect and record the monitoring data of the network latency and CPU occupancy rate. Divide the monitoring process into multiple time segments to ensure data integrity and the continuity of the time series, so as to accurately capture the trend of the running status changing over time. For the collected network latency data, calculate the average latency value within each time segment, and at the same time calculate the latency fluctuation range to identify the time segments with larger fluctuations. In the CPU occupancy rate monitoring, also calculate the average value of the occupancy rate changes in different time segments, and record the maximum and minimum values of the occupancy rate in each time segment to further analyze the CPU load stability. Normalize the average value data of the network latency and CPU occupancy rate to make the two have a unified scale for subsequent calculation and analysis. Compare the normalized network latency and CPU occupancy rate data with the predefined target performance standard values respectively, calculate the performance offset value of each time segment, and record the distribution trend of the time segment offset values. Sort out the performance status change rules of the target virtual machine in different time segments to generate the real-time performance status of the target virtual machine, providing basic data support for subsequent performance correction and resource expansion adjustment.

[0131] S412: Based on the real-time performance status of the target virtual machine, compare the memory usage rate and the read / write rate, judge the change of the average memory occupancy through the difference value, calculate the performance deviation by combining the read / write rate fluctuation range, correct the difference value, and adjust the resource expansion parameters, using the formula:

[0132] ;

[0133] Generate the performance deviation correction result of the target virtual machine;

[0134] Among them, represents the performance deviation value, is the current memory usage rate, is the target memory usage rate, is the current read / write rate, is the target read / write rate;

[0135] The advantage of the formula is that it simultaneously evaluates the deviations of the memory usage rate and the read / write rate. By combining the square operation and smoothing processing, it avoids misjudgment caused by the fluctuation of a single parameter, and further improves the robustness and accuracy of the performance deviation calculation;

[0136] Set the current memory usage rate of the target virtual machine , the target memory usage rate , the current read / write rate , the target read / write rate ;

[0137] First step, calculate the square of the difference in memory usage rate: ;

[0138] Second step, calculate the absolute value of the difference in read / write rate: ;

[0139] Third step, substitute the calculation result into the formula: ;

[0140] This result indicates that the performance deviation value , indicating that there is a certain deviation between the memory usage rate and the read / write rate of the current virtual machine, and it is necessary to further analyze and adjust the resource expansion parameters to reduce the performance deviation value.

[0141] S413: Based on the performance deviation correction result of the target virtual machine, perform resource expansion adjustment on the nodes exceeding the standard range, correct the real-time performance status of the expanded nodes, normalize the adjusted performance value to the standard range, summarize the node performance correction status, and generate a performance status table for the migrated completed nodes;

[0142] Perform dynamic resource adjustment on all nodes whose difference values exceed the standard range. First, detect the memory usage rate and read / write rate of the nodes exceeding the standard range, respectively record the time segments with the largest difference values, and perform resource expansion operations on the performance data corresponding to the time segments. By allocating more memory resources, reduce the memory usage rate deviation value. At the same time, adjust the read / write rate by optimizing disk performance or increasing network bandwidth, so that the read / write rate gradually approaches the target value. For the resource expansion operation of each node, monitor the change trend of its performance offset value in real time. After each round of adjustment operation, recalculate the current performance deviation value, and gradually optimize the performance indicators of the nodes. When the performance offset values of all nodes return to the standard range, normalize the performance status of the nodes, organize the performance status data of the corrected nodes, and record and summarize the performance status of all nodes as a performance status table for the migrated completed nodes, providing data support for subsequent resource configuration and migration strategy optimization.

[0143] Please refer to Figure 7 , the specific steps for obtaining the monitoring result of the migrated completed are as follows:

[0144] S421: Based on the performance status table of the migration completion nodes, extract the node performance parameter data, read and screen and sort out the operation metrics node by node, conduct time series analysis on the CPU occupancy rate, memory usage rate, and network traffic values, and generate a node performance status record table;

[0145] When extracting the node performance parameter data, read the performance parameters of each migrated node in sequence, call the performance monitoring interface item by item according to the record order in the node performance status table, collect the core operation metrics such as the CPU occupancy rate, memory usage rate, and network traffic of each node, sort the collected performance parameters indexed by the time stamp, classify and organize the performance data of the nodes according to the time series. After reading the operation metrics node by node, based on the time series analysis method, calculate the change trend of each performance parameter, determine the index fluctuation range and abnormal peak value. In the analysis, for the monitoring records of the CPU occupancy rate, every 10 minutes is used as a sampling interval, count the average occupancy value and screen out the abnormal points exceeding the preset threshold. For the memory usage rate, according to the memory allocation strategy of the node running tasks, record the instantaneous peak value and the continuous occupancy status. The network traffic data is statistically analyzed separately by upload and download components, generate a time series curve graph of the network traffic of each node. Through the data sorting and analysis of each node, generate a node performance status record table, including the time series data of the performance metrics of each node and the abnormal point marks, providing a basic basis for subsequent load analysis and adjustment.

[0146] S422: Based on the node performance status record table, analyze the node load occupancy situation, mark and classify the nodes exceeding the load threshold, transfer the tasks of the high-load nodes to the low-load nodes, evaluate the performance metrics of the low-load nodes, verify the adjusted node load parameters, and obtain the node load adjustment result;

[0147] Based on the data in the node performance status record table, analyze the load occupancy of each node, comprehensively score the CPU occupancy rate, memory usage rate, and network traffic of the node, calculate the current load status of the node, and mark and classify the nodes with load score values exceeding the threshold as high-load nodes. After marking, select low-load nodes as the target nodes for task transfer, and dynamically allocate some running tasks on the high-load nodes to the low-load nodes. During the transfer process, evaluate the performance indicators of the low-load nodes one by one, especially paying attention to the immediate change values of CPU and memory and the adjustment of network throughput after the transfer tasks, to ensure that the low-load nodes are still within the controllable load range after receiving the tasks. When verifying the adjusted node load parameters, collect the CPU occupancy rate and memory usage rate of the nodes after task transfer in real time by calling the monitoring interface again, record the load change trend, and ensure that the adjusted node operation parameters meet the performance standards of the migration tasks. Finally, obtain the node load adjustment results, and archive the performance parameters of the nodes before and after load adjustment to lay a foundation for the operation optimization of the migrated resources.

[0148] S423: Based on the node load adjustment results, monitor the adjusted nodes in real time, capture the running load and performance parameter data, compare the load stability and performance performance item by item, record the dynamic information after node migration, summarize the monitoring data, and generate the monitoring results after migration completion;

[0149] When monitoring the adjusted nodes in real time, continuously collect the load parameters and performance data of the node operation, record the core indicators such as CPU occupancy rate, memory usage rate, and network traffic at the minute level, and capture the dynamic change values of each parameter in real time. When comparing the load stability item by item, use the node parameters in the load adjustment results as the benchmark, compare the current monitoring data with the adjustment results, and check whether the load change tends to be stable, focusing on the long-term impact of the migration task on the node performance. In the comparison of performance performance, record the completion time, response delay, and resource occupancy of the running tasks of each node, generate a performance fluctuation curve graph, identify the time periods of abnormal fluctuations and the corresponding load change data. When recording the dynamic information after node migration, statistically analyze each performance parameter of the node by time segment, and classify and organize it to form the historical trend of performance performance. After summarizing the monitoring data, generate the monitoring results after migration completion by analyzing the load change of each node before and after the migration task. The monitoring results include the load adjustment of the node, the performance change trend, and the evaluation of the task migration effect.

[0150] The cross-cloud resource migration system is used to execute the above cross-cloud resource migration method. The system includes:

[0151] The node screening module calculates the weighted value of the block storage read / write rate and the network latency time based on the processing capacity of the virtual machine instance, the block storage read / write rate, and the network latency time, evaluates the three parameters by combining them proportionally, and performs node sorting to screen out the nodes with scores greater than the set threshold to obtain a high-score node set;

[0152] Based on the high-score node set, the path dependency module calculates the weights of the distributed file dependency path node attributes, accumulates the file path length to the node weight value, and screens the path by combining the performance differences between the virtual machine and the subnet to generate an optimized result of the path weight;

[0153] Based on the optimized result of the path weight, the transmission priority module analyzes the transmission latency and bandwidth occupancy ratio of the nodes, re-analyzes the distribution of the transmission efficiency by combining the node memory free rate, and adjusts the priority of the nodes with occupancy ratios exceeding the threshold to generate an optimized transmission path sequence;

[0154] Based on the optimized transmission path sequence, the migration performance monitoring module monitors the running state of the target virtual machine and the network latency in real time, compares the difference value between the node memory usage rate and the block storage read / write rate, corrects the performance deviation and records the performance state to generate a migration completion monitoring result.

[0155] The above is only the preferred embodiment of the present invention, and it does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A cross-cloud resource migration method, characterized in that, It includes the following steps: S1: Based on the processing capacity of the virtual machine instance, the read / write rate of the block storage, and the network latency time, proportionally allocate the processing capacity of the virtual machine instance, perform a weighted average on the read / write rate of the block storage and the network latency time, integrate them into a two-dimensional matrix, sort and eliminate low-threshold nodes, and generate a preliminary resource optimization dataset; The specific steps for obtaining the preliminary resource optimization dataset are as follows: S111: Based on the processing capacity of the virtual machine instance, the read / write rate of the block storage, and the network latency time, proportionally allocate the processing capacity of the virtual machine instance. By comparing the load conditions and required resources of different virtual machine instances, generate the processing capacity weight value of the virtual machine instance; S112: Based on the processing capacity weight value of the virtual machine instance, perform a weighted average calculation on the read / write rate of the block storage and the network latency time, using the formula: ； Generate the weighted block storage and network latency weight value; wherein, is the weighted average value, is the weight coefficient of the block storage read / write rate, is the weight coefficient of the network latency time, is the block storage read / write rate, is the network latency time; S113: Integrate the weighted block storage and network latency weight value with the processing capacity weight into a two-dimensional matrix. Each row of the matrix represents the relevant data of a virtual machine instance, where the first column represents the processing capacity weight of the virtual machine instance, and the second column represents the corresponding weighted value of the block storage and network latency. Sort according to the weighted results, select the virtual machine instances that meet the conditions, and eliminate the instances below the threshold to generate a preliminary resource optimization dataset; S2: Based on the preliminary resource optimization dataset, extract the performance weight difference between the virtual machine and the private subnet, accumulate the distributed file dependency path parameters to the node path, generate the resource performance dependency weight value, and perform an optimal connection selection on the critical path to generate a resource mapping optimization path set; The specific steps for obtaining the resource performance dependency weight value are as follows: S211: Based on the preliminary resource optimization dataset, extract the performance index data between the virtual machine and the private subnet, perform a ratio analysis on the virtual machine network traffic and the subnet traffic, analyze the differences between the traffic ratio and the throughput and latency parameters, and generate the performance weight difference value between the virtual machine and the private subnet; S212: Based on the performance weight difference value between the virtual machine and the private subnet, call the path parameter set of the distributed file, identify the node path weight and bandwidth ratio in the path, and superimpose the influence of the path dependency times, using the formula: ； Calculate the node path dependency weight; Among them, represents the node path dependence weight, represents the path weight of the th node in the path, represents the bandwidth occupancy ratio of the th node in the path, represents the cumulative number of path dependencies, represents the total number of nodes in the path, is the weight adjustment coefficient; S213: Based on the node path dependency weight, combine the performance weight difference value between the virtual machine and the private subnet, perform a normalization operation on the weight data of the two, analyze the weight dependency of the overall resource performance, and generate the resource performance dependency weight value; The specific steps for obtaining the resource mapping optimization path set are as follows: S221: Based on the resource performance dependency weight value, extract each performance parameter in the resource set. By comparing the performance values between resources, determine the weight difference between each pair of resources, and classify and statistically analyze the weight differences to generate a resource weight mapping node set; S222: Based on the resource weight mapping node set, by screening the node and path weight combinations, analyze each path weight cumulative value one by one, measure the dependency strength of the associated resource node combinations, and match the measurement results with the weight values to generate a critical resource path set; S223: Based on the set of critical resource paths, by comparing the node dependency weight relationships within the paths, filter out the path segments with the optimal weight values, connect and process them, verify the integrity of the connected paths and adjust the structure, split and recombine the path segments with large weight differences until the path weight distribution tends to be uniform, add the optimized paths to the mapping set, and obtain the resource mapping optimized path set; S3: Based on the resource mapping optimized path set, analyze the transmission latency of the virtual machines and the bandwidth occupancy of the load balancers, match the memory idle rate, recursively adjust the transmission efficiency distribution value, reassign the transmission priorities for the nodes exceeding the critical point, and iteratively adjust to generate a resource priority migration sequence list; The specific steps for obtaining the resource priority migration sequence list are as follows: S311: Based on the resource mapping optimized path set, extract the virtual machine transmission latency data, calculate the average latency value of each path, and perform a coupled analysis with the bandwidth occupancy value of the private subnet load balancer to evaluate the transmission latency characteristics, select the paths exceeding the target threshold, record the bandwidth occupancy ratio, and generate the analysis result of transmission latency and bandwidth occupancy; S312: Based on the analysis result of transmission latency and bandwidth occupancy, evaluate the relationship between path transmission load and memory occupancy, recursively adjust the transmission efficiency distribution value, and use the formula: ； Generate the distribution adjustment result of path transmission efficiency; Among them, represents the transmission efficiency distribution value, is the memory free rate of the th path, is the path bandwidth occupancy weight, is the path load value, is the adjustment coefficient, is the total sum of path transmission priority weights, represents the total number of paths; S313: Based on the distribution adjustment result of path transmission efficiency, reassign the priorities for the path nodes with transmission efficiency exceeding the critical point, recursively adjust the node weights item by item according to the priorities, sort according to node criticality and load priority, determine the migration priorities, and generate a resource priority migration sequence list; S4: Based on the resource priority migration sequence list, update the running status and network latency of the target virtual machines in real time, compare the memory usage rate and read / write rate, perform resource expansion on the nodes with difference values exceeding the range, correct the performance offset value of the target nodes to within the standard range, generate a performance status table of the migrated completed nodes, monitor and adjust the node running load, and generate a monitoring result of the migrated completed; The specific steps for obtaining the performance status table of the migrated completed nodes are as follows: S411: Based on the resource priority migration sequence list, monitor the running status of the target virtual machines in real time, collect the network latency value and CPU occupancy rate of the target virtual machines, record the change data in combination with the node resource metrics, and normalize the network latency and CPU occupancy rate to obtain the real-time performance status of the target virtual machines; S412: Based on the real-time performance status of the target virtual machines, compare the memory usage rate and read / write rate, judge the change of the average memory occupancy through the difference value, calculate the performance deviation in combination with the fluctuation range of the read / write rate, correct the difference value, and adjust the resource expansion parameters, using the formula: ； Generate the performance deviation correction result of the target virtual machines; Among them, represents the performance deviation value, is the current memory usage rate, is the target memory usage rate, is the current read / write rate, is the target read / write rate; S413: Based on the performance deviation correction result of the target virtual machines, perform resource expansion adjustment on the nodes exceeding the standard range, correct the real-time performance status of the expanded nodes, normalize the adjusted performance value to the standard range, summarize the node performance correction status, and generate a performance status table of the migrated completed nodes.

2. The cross-cloud resource migration method according to claim 1, wherein The specific steps for obtaining the monitoring result of the migrated completed are as follows: S421: Extract node performance parameter data based on the migration completion node performance status table, read and filter the operation metrics node by node, perform time series analysis on the CPU occupancy rate, memory usage rate, and network traffic value, and generate a node performance status record table; S422: Based on the node performance status record table, analyze the node load occupancy, mark and classify the nodes exceeding the load threshold, transfer the high-load node tasks to the low-load nodes, evaluate the performance metrics of the low-load nodes, verify the adjusted node load parameters, and obtain the node load adjustment result; S423: Based on the node load adjustment result, monitor the adjusted nodes in real time, capture the operation load and performance parameter data, compare the load stability and performance performance item by item, record the dynamic information after node migration, summarize the monitoring data, and generate a migration completion monitoring result.

3. A cross-cloud resource migration system, characterized in that, According to the cross-cloud resource migration method described in any one of claims 1-2, the system includes: The node screening module calculates the weighted value of the block storage read / write rate and the network latency time based on the processing capacity of the virtual machine instance, the block storage read / write rate, and the network latency time, evaluates the three parameters in proportion, sorts the nodes, and screens the nodes with scores greater than the set threshold to obtain a high-score node set; The path dependency module calculates the weights of the distributed file dependency path node attributes based on the high-score node set, accumulates the file path length to the node weight value, and screens the path by combining the performance differences between the virtual machine and the subnet to generate a path weight optimization result; The transmission priority module analyzes the transmission latency and bandwidth occupancy ratio of the nodes based on the path weight optimization result, re-analyzes the transmission efficiency distribution by combining the node memory free rate, and adjusts the priority of the nodes with occupancy rates exceeding the threshold to generate an optimized transmission path sequence; The migration performance monitoring module monitors the running status of the target virtual machine and the network latency in real time based on the optimized transmission path sequence, compares the difference value between the node memory usage rate and the block storage read / write rate, corrects the performance deviation and records the performance status, and generates a migration completion monitoring result.

Images