A content distribution system and method based on cloud service

Abstract

The present invention relates to the technical field of content distribution, and specifically to a content distribution system and method based on cloud services, wherein the system includes a real-time compression module, a task scheduling module, a bandwidth adaptation module, and a content merging module. The present invention effectively improves the real-time compression rate of data by dynamically adjusting compression parameters to cope with different bandwidths and device capabilities, thereby reducing the time required for transmission and optimizing the use of network resources. The performance data of cloud service edge nodes is intelligently analyzed to make task scheduling more accurate, and task priorities can be adjusted in real time, thereby reducing network congestion during peak hours, significantly improving the flexibility and efficiency of data processing, and dynamically adjusting the data shard size in combination with real-time network performance data, so that the system can maintain stability when the bandwidth fluctuates, thereby providing users with a homogeneous service experience under different network conditions, optimizing the data shard merging logic, and reducing data arrival delay.

Description

Technical Field

[0001] The present invention relates to the technical field of content distribution, and in particular to a content distribution system and method based on cloud services. Background Art

[0002] Content distribution technology focuses on how to effectively distribute data and content from the source point to multiple receiving points around the world. It covers content distribution networks (CDNs), edge computing, caching mechanisms, load balancing, and network optimization technologies. These technologies work together to reduce latency between content providers and end users, increase data transmission rates, optimize bandwidth usage, and ensure that content can be delivered and accessed quickly and reliably even under high demand conditions. Content distribution provides a faster and more stable access experience by copying content to widely distributed servers and redirecting user requests to the nearest servers.

[0003] Among them, the content distribution system of cloud services refers to a system that uses a distributed server network to effectively distribute content to the user end. It has a wide range of uses, such as in video streaming services, web page acceleration, software updates and other application scenarios. The content distribution system uses localized storage data and intelligent routing technology to ensure that users can receive data from the nearest server, thereby reducing latency and improving the speed and efficiency of content acquisition. It is crucial for processing large amounts of network traffic.

[0004] In scenarios with unstable networks or extremely uneven geographical distribution of users, traditional systems cannot effectively adapt to changes, resulting in service interruptions or performance degradation, especially in areas far from data centers. Traditional systems perform poorly in data compression and instant response. In application scenarios that require rapid processing and transmission of large amounts of real-time streaming content, the lack of efficient real-time data processing capabilities leads to increased latency. Existing task scheduling and resource allocation appear rigid under high load conditions and lack the ability to dynamically optimize based on real-time loads. This is particularly evident during large-scale online activities or peak usage periods. The lack of flexible task management and priority adjustment affects the overall quality of distribution services. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a content distribution system and method based on cloud services.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a content distribution system based on cloud service, the system comprising:

[0007] The real-time compression module measures the data stream size of content distribution based on the current cloud service bandwidth and user device capabilities, adjusts LZMA compression parameters, calculates the real-time data compression rate, and compresses instant streaming media and real-time data service content to obtain compression efficiency data streams;

[0008] The task scheduling module monitors the CPU and memory usage data of edge nodes in the cloud service based on the compression efficiency data stream, analyzes node performance and task load in real time, reorders content distribution tasks, adjusts key task priorities, and obtains scheduling performance optimization information;

[0009] The bandwidth adaptation module analyzes the current network bandwidth and bandwidth trend based on the scheduling performance optimization information and the cloud service network performance data, and dynamically adjusts the data fragment size of content distribution to match the current network fluctuations and obtain dynamic bandwidth adjustment parameters;

[0010] The content merging module calculates the arrival delay between content pieces based on the dynamic bandwidth adjustment parameters, optimizes the combination logic of data pieces, performs content piece merging, optimizes the user-side content receiving process, and obtains the receiving process performance index.

[0011] The present invention is improved in that the step of measuring the data stream size of the content distribution is specifically as follows:

[0012] Based on the current cloud service bandwidth and user device capabilities, monitor the size of data received by the user device during the target time period, collect real-time transmission information of data packets, and obtain original data flow records;

[0013] Extract the total amount of data flow per minute from the original data flow record, and calculate the average size of the data flow in the time period;

[0014] According to the average size of the data flow, the flow changes during peak and non-peak hours are analyzed using the formula:

[0015]

[0016] Calculate the deviation of the data flow size , we get the data flow volatility, where For minute data traffic, is the average data flow size, is the sample size.

[0017] The present invention is improved in that the step of acquiring the compression efficiency data stream is specifically as follows:

[0018] Setting the compression level and dictionary size of the LZMA compression algorithm, determining the optimal combination of compression parameters, performing multiple compression tests, comparing the size and processing time of compressed files under different parameter settings, selecting a parameter configuration that matches the current data type, and generating an optimized compression parameter set;

[0019] Based on the optimized compression parameter set, the live streaming media and real-time data service content are compressed, and the data size before and after compression is measured, using the formula:

[0020]

[0021] Calculating the compression ratio , and obtain the compression efficiency data stream, where is the original data size, Is the compressed data size.

[0022] The present invention is improved in that the real-time analysis steps of the node performance and task load are specifically as follows:

[0023] Based on the compression efficiency data stream, automatically collect CPU and memory usage data of each node of the cloud service to generate real-time performance data;

[0024] Applying data cleaning technology to process the real-time performance data, removing abnormal values ​​or erroneous records, and verifying the consistency and availability of the data to obtain a processed performance data set;

[0025] Based on the processed performance data set, a load analysis is performed using the formula:

[0026]

[0027] Computing resource utilization , generate node performance evaluation results, where is the current resource usage, is the average resource usage, is the total available resources.

[0028] The present invention is improved in that the step of obtaining the scheduling performance optimization information is specifically as follows:

[0029] Analyze nodes with extremely high or low resource utilization and identify key nodes that need to reallocate tasks;

[0030] According to the current task load and performance data of the key nodes, tasks are transferred from high-load nodes to low-load nodes, and task priorities are adjusted for load balancing to generate a task allocation plan;

[0031] Based on the task allocation plan, the formula is adopted:

[0032]

[0033] Optimize task allocation and obtain scheduling performance optimization information ,in, Representation Node The load capacity, Representative Node The task weight factor, Is a node Total processing capacity, is the total number of nodes participating in the calculation.

[0034] The present invention is improved in that the analysis steps of the current network bandwidth and bandwidth trend are specifically as follows:

[0035] Based on the scheduling performance optimization information, data noise is removed and key performance indicators are extracted to obtain cleaned network performance data;

[0036] Based on the cleaned network performance data, the current status and predicted trend of bandwidth are analyzed using the formula:

[0037]

[0038] Get bandwidth usage trend index ,in, Indicates the currently measured bandwidth usage, represents the variability of bandwidth usage, is a coefficient used to adjust the model's sensitivity to bandwidth usage, is the adjustment coefficient.

[0039] The present invention is improved in that the step of obtaining the dynamic bandwidth adjustment parameter is specifically as follows:

[0040] Collect real-time traffic and bandwidth data of the current network, analyze and evaluate network traffic fluctuations, and generate traffic and bandwidth fluctuation information;

[0041] Based on the traffic and bandwidth fluctuation information, the formula is adopted:

[0042]

[0043] Get dynamic bandwidth adjustment parameters ,in, is the optimized data sharding baseline size, is the bandwidth usage trend index, is the network bandwidth fluctuation rate, and It is the coefficient that adjusts the influence of bandwidth trend and volatility on data sharding adjustment.

[0044] The present invention is improved in that the step of obtaining the receiving process performance indicator is specifically as follows:

[0045] Based on the dynamic bandwidth adjustment parameters, logs of received data are collected from the user end, including receiving time, data volume and error rate, and abnormal and erroneous data in the logs are removed to generate cleaned log data;

[0046] Based on the cleaned log data, extract and calculate key performance indicators, including average receiving rate and data integrity, and generate performance analysis records;

[0047] The data in the performance analysis record is evaluated using the formula:

[0048]

[0049] Computing performance indicators , and obtain the receiving process performance index, where Represents the average receiving rate, represents the data integrity index, The number of errors recorded in the log.

[0050] A content distribution method based on cloud services, wherein the content distribution method based on cloud services is performed based on the above-mentioned content distribution system based on cloud services, and comprises the following steps:

[0051] S1: Based on the current cloud service bandwidth and user device capabilities, measure the content distribution data stream size, adjust LZMA compression parameters, calculate the real-time data compression rate, compress instant streaming media and real-time data service content, and obtain the compression efficiency data stream;

[0052] S2: Based on the compression efficiency data stream, monitor the CPU and memory usage data of edge nodes in the cloud service, analyze node performance and task load in real time, reorder content distribution tasks, and obtain scheduling performance optimization information;

[0053] S3: Based on the scheduling performance optimization information and in combination with the cloud service network performance data, the current network bandwidth and bandwidth trend are analyzed, and the data fragment size of the content distribution is dynamically adjusted to obtain a dynamic bandwidth adjustment parameter;

[0054] S4: Based on the dynamic bandwidth adjustment parameter, calculate the arrival delay between content slices, optimize the combination logic of data slices, perform content slice merging, and generate merging efficiency data;

[0055] S5: According to the combined efficiency data, content distribution is adjusted, the user-side content receiving process is optimized, and the transmission efficiency and stability of data in different network environments are verified to obtain a receiving process performance index.

[0056] Compared with the prior art, the advantages and positive effects of the present invention are:

[0057] In the present invention, by dynamically adjusting compression parameters to cope with different bandwidths and device capabilities, the real-time compression rate of data is effectively improved, thereby reducing the time required for transmission and optimizing the use of network resources. Intelligent analysis of the performance data of cloud service edge nodes makes task scheduling more accurate, and can adjust task priorities in real time, reducing network congestion during peak hours, significantly improving the flexibility and efficiency of data processing, and dynamically adjusting the data shard size in combination with real-time network performance data allows the system to maintain stability when the bandwidth fluctuates, thereby providing users with a homogeneous service experience under different network conditions, optimizing the data slice merging logic, reducing data arrival delays, and improving the overall speed and efficiency of data transmission, ensuring that users have a seamless and efficient experience when receiving large-capacity data. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A module diagram of a content distribution system based on cloud services is proposed for the present invention;

[0059] Figure 2 A flow chart for measuring the data stream size of content distribution in the present invention;

[0060] Figure 3 A flowchart of obtaining the compression efficiency data stream in the present invention;

[0061] Figure 4 A flowchart of real-time analysis of node performance and task load in the present invention;

[0062] Figure 5 A flowchart for obtaining scheduling performance optimization information in the present invention;

[0063] Figure 6 The present invention is a flowchart for analyzing the current network bandwidth and bandwidth trend;

[0064] Figure 7 A flowchart of obtaining dynamic bandwidth adjustment parameters in the present invention;

[0065] Figure 8 The figure is a flowchart for obtaining the receiving process performance indicator in the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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 intended to limit the present invention.

[0067] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are 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 therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. Example

[0068] See also Figure 1 The present invention provides a technical solution: a content distribution system based on cloud services comprises:

[0069] The real-time compression module measures the data stream size of content distribution based on the current cloud service bandwidth and user device capabilities, adjusts LZMA compression parameters, calculates the real-time data compression rate, and compresses instant streaming media and real-time data service content to obtain compression efficiency data streams;

[0070] The task scheduling module monitors the CPU and memory usage data of edge nodes in the cloud service based on the compression efficiency data stream, analyzes node performance and task load in real time, reorders content distribution tasks, adjusts the priority of key tasks, and obtains scheduling performance optimization information;

[0071] The bandwidth adaptation module analyzes the current network bandwidth and bandwidth trend based on the scheduling performance optimization information and the cloud service network performance data, and dynamically adjusts the data fragment size of content distribution to match the current network fluctuations and obtain dynamic bandwidth adjustment parameters;

[0072] The content merging module calculates the arrival delay between content pieces based on dynamic bandwidth adjustment parameters, optimizes the combination logic of data pieces, and performs content piece merging to optimize the user-side content receiving process and obtain the receiving process performance indicators.

[0073] The compression efficiency data stream includes compression rate, data integrity index, and transmission efficiency. The scheduling performance optimization information includes task response time, resource allocation efficiency, and processing priority. The dynamic bandwidth adjustment parameters include real-time bandwidth utilization, adjusted segment size, and response time change results. The receiving process performance indicators include data arrival rate, merge execution time, and receiving integrity rating.

[0074] See also Figure 2 ,The specific steps for measuring the data stream size of content distribution are:

[0075] Based on the current cloud service bandwidth and user device capabilities, monitor the size of data received by the user device during the target time period, collect real-time transmission information of data packets, and obtain original data flow records;

[0076] Collect data within a specific time period, monitor and record the data size received by each user device in real time. The data collection process involves real-time monitoring of the network interface and the capture and recording of data packets. Set the monitored network interface parameters, such as IP address and port number, and then capture all data packets passing through the interface, record the size and timestamp of each data packet for further data analysis. Through recording, the data flow can be preliminarily classified and quantified, such as sorting the data packets in chronological order and calculating the total data flow or average data flow within a time window. Such quantification results will provide basic data for subsequent data flow analysis, thereby obtaining the original data flow record.

[0077] Extract the total amount of data flow per minute from the original data flow record and calculate the average size of the data flow in the time period;

[0078] The total data flow per minute is extracted from the collected raw data flow records. This process requires further processing and analysis of the records of the original data packets. All records are sorted by timestamp, and then the data is grouped and counted by minute to calculate the total size of all data packets per minute. The processing process requires precise time matching and data integration technology to ensure the accuracy and representativeness of the data. Through data processing, the data flow per minute and the average data flow size can be obtained.

[0079] According to the average size of data flow, analyze the traffic changes during peak and non-peak hours, using the formula:

[0080]

[0081] Calculate the deviation of the data flow size , we get the data flow volatility, where For minute data traffic, is the average data flow size, is the sample size;

[0082] The data traffic monitored within one minute was 100MB, 150MB, and 120MB;

[0083] Calculate the average:

[0084]

[0085] Substitute into the formula to calculate:

[0086]

[0087] It indicates that the volatility of data flow is 25MB. This result shows that the volatility of data flow size is small and network data transmission is relatively stable, which is of great significance for monitoring and optimizing network performance.

[0088] See also Figure 3 , the specific steps for obtaining the compression efficiency data stream are:

[0089] Set the compression level and dictionary size of the LZMA compression algorithm, determine the optimal combination of compression parameters, perform multiple compression tests, compare the size and processing time of compressed files under different parameter settings, select the parameter configuration that matches the current data type, and generate the optimized compression parameter set;

[0090] When conducting compression testing, it is necessary to set and adjust the parameters of multiple compression algorithms. The process includes selecting compression level and dictionary size as the main variables, recording the size of compressed files and processing time under each parameter configuration, and identifying the parameter combination that can best improve compression efficiency through comparative analysis. It not only involves parameter adjustment, but also detailed records of each experimental result and subsequent data analysis. Through the steps, the parameter configuration suitable for the current data type is determined, and an optimized compression parameter set is generated.

[0091] Based on the optimized compression parameter set, the live streaming media and real-time data service content are compressed, and the data size before and after compression is measured using the formula:

[0092]

[0093] Calculating the compression ratio , and obtain the compression efficiency data stream, where is the original data size, is the size of the compressed data;

[0094] The following data was collected:

[0095] Original data size 500MB, compressed data size is 350MB, and we get:

[0096]

[0097] It indicates that the compression efficiency is 30%. The result shows that the compression operation reduces the data volume by 30%, indicating that the compression parameters are set correctly and effectively reduce the data volume, reducing the burden of storage and transmission.

[0098] See also Figure 4 ,The real-time analysis steps of node performance and task load are as follows:

[0099] Based on the compression efficiency data stream, automatically collect the CPU and memory usage data of each node of the cloud service to generate real-time performance data;

[0100] By implementing monitoring at the edge nodes, ensuring real-time data updates, capturing the CPU and memory usage data of each node, and involving the deployment of technology, the monitoring parameters are set to adapt to the specific needs of different nodes. Correct configuration directly affects the effectiveness and accuracy of data collection. The real-time and accuracy of data are the key to evaluating node performance. The generated real-time performance data records will provide the basis for the next step of data processing and analysis.

[0101] Apply data cleaning technology to process real-time performance data, remove outliers or erroneous records, and verify the consistency and availability of data to obtain a processed performance data set;

[0102] The collected real-time performance data is cleaned and standardized to remove outliers or erroneous records caused by sensor errors, data transmission errors, etc. This step ensures the accuracy and reliability of data analysis. The preprocessed data set will unify the format and reduce noise, laying the foundation for more in-depth data analysis. Standardized performance data is easier to identify and process, ensuring the efficiency and accuracy of data analysis.

[0103] Based on the processed performance data set, load analysis is performed using the formula:

[0104]

[0105] Computing resource utilization , generate node performance evaluation results, where is the current resource usage, It is the average value of resource usage, which is used to reflect the fluctuation of node usage. is the total available resources;

[0106] Current resource usage of the node is 150, the average resource usage is 100, the total resource capacity is 200, substitute it into the formula to calculate:

[0107]

[0108] The results show that the current resource utilization of the node is relatively stable and the load is low. By considering the volatility of resource usage for load rate calculation, more dynamic and adaptive decision support can be provided for resource scheduling, which helps to optimize resource allocation and improve overall performance.

[0109] See also Figure 5,The specific steps for obtaining scheduling performance optimization information are:

[0110] Analyze nodes with extremely high or low resource utilization and identify key nodes that need to reallocate tasks;

[0111] By monitoring resource utilization, especially those nodes with extremely high CPU and memory utilization of 80% or extremely low 20%, the nodes that need priority attention are automatically marked. Relying on continuous performance monitoring and real-time data analysis technology, the data is compared with historical performance standards to identify nodes that deviate from the normal operating range. The tasks of the nodes are re-evaluated to decide whether resources need to be reallocated or the task queue needs to be adjusted, so as to ensure that all nodes can operate in the optimal state and avoid resource waste or overload.

[0112] According to the current task load and performance data of key nodes, tasks are transferred from high-load nodes to low-load nodes, and task priorities are adjusted for load balancing to generate a task allocation plan;

[0113] An in-depth analysis is performed on the marked key nodes. Based on the current task load and performance data, a decision is made to transfer some tasks from high-load nodes to low-load nodes. Task priorities are dynamically adjusted to ensure that tasks are evenly distributed to each node. In this process, models are used to predict node load and optimize task scheduling, including queue theory models and machine learning prediction models. Task reordering and priority adjustment are performed based on real-time performance analysis results to ensure that each node is not overloaded due to too many tasks or idle due to too few tasks. The generated task allocation plan takes into account the operating capabilities and response time of each node.

[0114] Based on the task allocation plan, the formula is adopted:

[0115]

[0116] Optimize task allocation and obtain scheduling performance optimization information ,in, Representation Node The load capacity, Representative Node The task weight factor, Is a node Total processing capacity, is the total number of nodes participating in the calculation;

[0117] The following data were monitored:

[0118] Number of nodes , node load , task weight factor , the total processing capacity of the node ;

[0119] Calculate the numerator:

[0120]

[0121] Calculate the denominator:

[0122]

[0123] Apply the formula to calculate:

[0124]

[0125] The result shows that the optimization index is 0.153. The larger the value, the better the scheduling performance optimization, which reflects the improved efficiency of system load and resource utilization after reallocating tasks and adjusting priorities.

[0126] See also Figure 6 , the specific steps for analyzing the current network bandwidth and bandwidth trend are:

[0127] Based on the scheduling performance optimization information, data noise is removed and key performance indicators are extracted to obtain cleaned network performance data;

[0128] Integrate performance data from the cloud service network with scheduling performance optimization information, conduct in-depth analysis to identify key performance indicators and point out potential bottleneck problems, and ensure that the information collected from multiple data sources can provide accurate basic data for subsequent performance evaluation. The basic data is cleaned and verified through data processing technology to remove errors and anomalies, ensuring the reliability and representativeness of the data. The generated network performance analysis report will be used to evaluate the current status of the network and predict its future performance trends.

[0129] Based on the cleaned network performance data, analyze the current status and forecast trend of bandwidth using the formula:

[0130]

[0131] Get bandwidth usage trend index ,in, Indicates the currently measured bandwidth usage, represents the variability of bandwidth usage, is a coefficient used to adjust the model's sensitivity to bandwidth usage, is the adjustment factor used to weight the effect of variability in bandwidth usage in the overall trend analysis;

[0132] The following data were monitored:

[0133] Mbps;

[0134] Mbps;

[0135] and ;

[0136] Calculation process:

[0137]

[0138] The result shows that the bandwidth trend index is 11.5. The value helps analyze the stability and potential fluctuations of current network bandwidth usage and provides a basis for formulating reasonable bandwidth management strategies.

[0139] See also Figure 7 , the specific steps for obtaining the dynamic bandwidth adjustment parameters are:

[0140] Collect real-time traffic and bandwidth data of the current network, analyze and evaluate network traffic fluctuations, and generate traffic and bandwidth fluctuation information;

[0141] By monitoring network traffic and bandwidth data in real time, combined with known network usage trends, analyzing network performance fluctuations and traffic changes, it involves data collection and preliminary processing to ensure accurate and real-time network status information, and conduct preliminary bandwidth fluctuation analysis based on the data. By calculating the average value and fluctuation rate of data traffic, a comprehensive report on network traffic and bandwidth fluctuations is generated to ensure that the data transmission strategy matches the actual network status.

[0142] Based on the traffic and bandwidth fluctuation information, the formula is used:

[0143]

[0144] Get dynamic bandwidth adjustment parameters ,in, is the optimized data sharding baseline size, is the bandwidth usage trend index, is the network bandwidth fluctuation rate, and It is the coefficient that adjusts the influence of bandwidth trend and volatility on data sharding adjustment;

[0145] The following data was collected:

[0146] bytes;

[0147] Indicates that the bandwidth growth trend is 5%;

[0148] It means the bandwidth fluctuation rate is 10%;

[0149] and .

[0150] Substitute into the formula to calculate:

[0151]

[0152] The result shows that after adjustment according to the current network status, the dynamic bandwidth adjustment parameter is 1343.75 bytes, which helps to improve the flexibility and efficiency of network data processing.

[0153] See also Figure 8 , the specific steps for obtaining the receiving process performance indicator are:

[0154] Based on the dynamic bandwidth adjustment parameters, the logs of received data are collected from the user end, including the receiving time, data volume and error rate, and the abnormal and erroneous data in the records are removed to generate cleaned log data;

[0155] The cleaned log data set provides basic data for subsequent performance analysis, so that the receiving process can be evaluated and optimized more accurately. The data set directly affects the accuracy of performance evaluation and the reliability of performance indicators. Through the steps, the specific process of data collection and cleaning can be clarified, and the integrity and accuracy of the data can be ensured.

[0156] Based on the cleaned log data, extract and calculate key performance indicators, including average receiving rate and data integrity, and generate performance analysis records;

[0157] Based on the cleaned log data, key performance indicators such as average receiving rate and data integrity are extracted and calculated. Through data analysis methods, the mean and standard deviation of the indicators are calculated, and performance analysis records are generated. The records show the key performance indicators in the receiving process and provide a basis for evaluating the overall performance. At the same time, the execution of this step ensures that the necessary statistical information can be accurately obtained from the log data, providing the necessary data support for performance evaluation.

[0158] The data in the performance analysis record is evaluated using the formula:

[0159]

[0160] Computing performance indicators , and obtain the receiving process performance index, where Represents the average receiving rate, represents the data integrity index, is the number of errors recorded in the log;

[0161] The following data was collected:

[0162] MB / s;

[0163] ;

[0164] ;

[0165] Substituting into the formula, we get:

[0166]

[0167] The results show that despite the existence of errors, the overall reception efficiency remains at a high level, which can effectively reflect the performance and stability of the data reception process and is a key indicator for optimizing the user-side content reception process.

[0168] A content distribution method based on cloud service comprises the following steps:

[0169] S1: Based on the current cloud service bandwidth and user device capabilities, measure the content distribution data stream size, adjust LZMA compression parameters, calculate the real-time data compression rate, compress instant streaming media and real-time data service content, and obtain the compression efficiency data stream;

[0170] S2: Based on the compression efficiency data stream, monitor the CPU and memory usage data of edge nodes in the cloud service, analyze node performance and task load in real time, reorder content distribution tasks, and obtain scheduling performance optimization information;

[0171] S3: Based on the scheduling performance optimization information and combined with the cloud service network performance data, the current network bandwidth and bandwidth trend are analyzed, and the data fragment size of the content distribution is dynamically adjusted to obtain the dynamic bandwidth adjustment parameters;

[0172] S4: Based on the dynamic bandwidth adjustment parameters, calculate the arrival delay between content slices, optimize the combination logic of data slices, perform content slice merging, and generate merging efficiency data;

[0173] S5: According to the merged efficiency data, adjust the content distribution, optimize the user-side content receiving process, and verify the transmission efficiency and stability of the data in different network environments to obtain the receiving process performance index.

[0174] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A content distribution system based on cloud services, characterized in that: The system comprises: The real-time compression module measures the data stream size of content distribution based on the current cloud service bandwidth and user device capabilities, adjusts LZMA compression parameters, calculates the real-time data compression rate, and compresses instant streaming media and real-time data service content to obtain compression efficiency data streams; The task scheduling module monitors the CPU and memory usage data of edge nodes in the cloud service based on the compression efficiency data stream, analyzes node performance and task load in real time, reorders content distribution tasks, adjusts key task priorities, and obtains scheduling performance optimization information; The bandwidth adaptation module analyzes the current network bandwidth and bandwidth trend based on the scheduling performance optimization information and the cloud service network performance data, and dynamically adjusts the data fragment size of content distribution to match the current network fluctuations and obtain dynamic bandwidth adjustment parameters; The content merging module calculates the arrival delay between content pieces based on the dynamic bandwidth adjustment parameters, optimizes the combination logic of data pieces, and performs content piece merging, optimizes the user-side content receiving process, and obtains the receiving process performance index; The steps of obtaining the dynamic bandwidth adjustment parameter are specifically as follows: Collect real-time traffic and bandwidth data of the current network, analyze and evaluate network traffic fluctuations, and generate traffic and bandwidth fluctuation information; Based on the traffic and bandwidth fluctuation information, the formula is adopted: Get dynamic bandwidth adjustment parameters ,in, is the optimized data sharding baseline size, is the bandwidth usage trend index, is the network bandwidth fluctuation rate, and It is the coefficient that adjusts the influence of bandwidth trend and volatility on data sharding adjustment; The steps for obtaining the receiving process performance indicator are specifically as follows: Based on the dynamic bandwidth adjustment parameters, logs of received data are collected from the user end, including receiving time, data volume and error rate, and abnormal and erroneous data in the logs are removed to generate cleaned log data; Based on the cleaned log data, extract and calculate key performance indicators, including average receiving rate and data integrity, and generate performance analysis records; The data in the performance analysis record is evaluated using the formula: Computing performance indicators , and obtain the receiving process performance index, where Represents the average receiving rate, represents the data integrity index, The number of errors recorded in the log.

2. The cloud service-based content distribution system according to claim 1, characterized in that: The steps for measuring the data stream size of the content distribution are specifically as follows: Based on the current cloud service bandwidth and user device capabilities, monitor the size of data received by the user device during the target time period, collect real-time transmission information of data packets, and obtain original data flow records; Extract the total amount of data flow per minute from the original data flow record, and calculate the average size of the data flow in the time period; According to the average size of the data flow, the flow changes during peak and non-peak hours are analyzed using the formula: Calculate the deviation of the data flow size , we get the data flow volatility, where For minute data traffic, is the average data flow size, is the sample size.

3. The cloud service-based content distribution system according to claim 1, characterized in that: The steps of obtaining the compression efficiency data stream are specifically as follows: Setting the compression level and dictionary size of the LZMA compression algorithm, determining the optimal combination of compression parameters, performing multiple compression tests, comparing the size and processing time of compressed files under different parameter settings, selecting a parameter configuration that matches the current data type, and generating an optimized compression parameter set; Based on the optimized compression parameter set, the live streaming media and real-time data service content are compressed, and the data size before and after compression is measured, using the formula: Calculating the compression ratio , get the compression efficiency data stream, where is the original data size, Is the compressed data size.

4. The cloud service-based content distribution system according to claim 1, characterized in that: The real-time analysis steps of the node performance and task load are specifically as follows: Based on the compression efficiency data stream, automatically collect CPU and memory usage data of each node of the cloud service to generate real-time performance data; Applying data cleaning technology to process the real-time performance data, removing abnormal values ​​or erroneous records, and verifying the consistency and availability of the data to obtain a processed performance data set; Based on the processed performance data set, a load analysis is performed using the formula: Computing resource utilization , generate node performance evaluation results, where is the current resource usage, is the average resource usage, is the total available resources.

5. The cloud service-based content distribution system according to claim 1, characterized in that: The steps for obtaining the scheduling performance optimization information are specifically as follows: Analyze nodes with extremely high or low resource utilization and identify key nodes that need to reallocate tasks; According to the current task load and performance data of the key nodes, tasks are transferred from high-load nodes to low-load nodes, and task priorities are adjusted for load balancing to generate a task allocation plan; Based on the task allocation plan, the formula is adopted: Optimize task allocation and obtain scheduling performance optimization information ,in, Representation Node The load capacity, Representative Node The task weight factor, Is a node Total processing capacity, is the total number of nodes participating in the calculation.

6. The cloud service-based content distribution system according to claim 1, characterized in that: The analysis steps of the current network bandwidth and bandwidth trend are specifically as follows: Based on the scheduling performance optimization information, data noise is removed and key performance indicators are extracted to obtain cleaned network performance data; Based on the cleaned network performance data, the current status and predicted trend of bandwidth are analyzed using the formula: Get bandwidth usage trend index ,in, Indicates the currently measured bandwidth usage, represents the variability of bandwidth usage, is a coefficient used to adjust the model's sensitivity to bandwidth usage, is the adjustment coefficient.

7. A content distribution method based on cloud service, characterized in that: The content distribution system based on cloud service according to any one of claims 1 to 6 is implemented, comprising the following steps: Based on the current cloud service bandwidth and user device capabilities, the content distribution data stream size is measured, the LZMA compression parameters are adjusted, the real-time data compression rate is calculated, and the instant streaming media and real-time data service content are compressed to obtain the compression efficiency data stream; Based on the compression efficiency data stream, the CPU and memory usage data of edge nodes in the cloud service are monitored, node performance and task load are analyzed in real time, content distribution tasks are reordered, and scheduling performance optimization information is obtained; Based on the scheduling performance optimization information, combined with the cloud service network performance data, the current network bandwidth and bandwidth trend are analyzed, and the data fragment size of the content distribution is dynamically adjusted to obtain a dynamic bandwidth adjustment parameter; Based on the dynamic bandwidth adjustment parameters, calculate the arrival delay between content slices, optimize the combination logic of data slices, perform content slice merging, and generate merging efficiency data; According to the combined efficiency data, content distribution is adjusted, the user-side content receiving process is optimized, and the transmission efficiency and stability of data in different network environments are verified to obtain the receiving process performance index.

Images