Software development method and system based on Internet and cloud computing

Abstract

The invention provides a software development method and system based on Internet and cloud computing. According to the method, time series data fed back by software development requirements are collected in real time through the Internet, multi-dimensional analysis is conducted through a cloud computing platform, the association strength of time distribution characteristics of requirement change frequency and user emotion parameters is extracted, and an evolution graph containing requirement priority distribution characteristics and change path topology is generated; dynamically distributing an optical signal transmission channel matched with the priority based on the weight of the map node, and carrying out priority weight marking on the original time sequence data; and finally, dynamically binding a cloud computing resource allocation proportion in combination with the weight value of the marked data and the topological node connection strength, and generating a task scheduling instruction. Through cross-layer cooperation of the demand evolution graph and the photon integrated circuit, accurate dynamic matching of the software development demand priority and the cloud computing resources is realized. According to the technical scheme provided by the invention, the efficiency and real-time performance of software development can be improved.

Description

Technical Field

[0001] The present application relates to the field of software development technology, and in particular to a software development method and system based on the Internet and cloud computing. Background Art

[0002] In current software development practices, dynamic demand changes have become the norm, especially in internet products, enterprise applications, and cloud services. Users expect rapid iteration, continuous delivery, and the ability to flexibly respond to changes in the market and business environment. These application scenarios require development methods that are highly flexible and collaborative, supporting remote team collaboration, real-time demand adjustments, rapid version releases, and flexible resource scheduling. Furthermore, they must implement visual management of the development process, continuous assurance of code quality, and integrated deployment across cross-platform environments to improve overall delivery efficiency and system stability.

[0003] To address these technical requirements, some companies are already adopting cloud-based continuous integration and delivery platforms based on the concept of integrated development and operations collaboration, combined with agile project management tools for software development. This solution leverages cloud-native architecture to automate the entire development, testing, and deployment process, using containerization technology to achieve environmental consistency. It also leverages Kanban systems and user story maps to dynamically decompose and prioritize requirements. Furthermore, it incorporates automated testing and pipeline mechanisms to ensure that each code change is rapidly verified and released, thereby improving team responsiveness and delivery quality, and initially meeting the software development management requirements for some dynamic demand scenarios.

[0004] However, this existing solution still has obvious limitations. First, it relies on relatively fixed process templates. When faced with extremely frequent changes in requirements, conflicts are likely to arise between task scheduling and version merging strategies, leading to increased integration costs. Secondly, the current automation tool chain lacks the ability to understand the business intent behind the changes in requirements, making it difficult to proactively adapt to changing trends. A large amount of manual intervention is still required for rule configuration and decision-making. In addition, team collaboration mechanisms are mostly based on centralized platforms, which lack support for distributed, multi-role parallel development, affecting the efficiency and transparency of cross-regional and cross-organizational collaboration. Therefore, there is an urgent need for a smarter and more adaptable software development method based on the Internet and cloud computing to meet the complex challenges of evolving requirements. Summary of the Invention

[0005] The present application provides a software development method and system based on the Internet and cloud computing, which is used to solve the problems of low efficiency and poor real-time performance of software development in the prior art.

[0006] In a first aspect, the present application provides a software development method based on the Internet and cloud computing, comprising:

[0007] Collecting time series data of software development demand feedback in real time through the Internet and transmitting the time series data to a cloud computing platform, wherein the time series data includes demand change frequency parameters and user comment sentiment tendency parameters;

[0008] Performing multi-dimensional feature analysis on the time series data through the cloud computing platform, extracting the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generating a software development demand evolution map including demand priority distribution characteristics and demand change path topology;

[0009] Based on the node weights in the software development demand evolution graph, dynamically allocating optical signal transmission channels adapted to the demand priority distribution characteristics;

[0010] performing priority weight marking processing on the time series data according to the mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics to generate marked time series data;

[0011] According to the priority weight in the marked time series data and the node connection strength of the demand change path topology, the priority weight is dynamically bound to the allocation ratio of cloud computing resources to generate a software development task scheduling instruction.

[0012] Optionally, dynamically binding the priority weights in the marked time series data with the node connection strengths of the demand change path topology to the allocation ratio of cloud computing resources to generate software development task scheduling instructions includes:

[0013] Establishing a mapping relationship table between the priority weights in the marked time series data and the types of cloud computing resources, wherein the types of cloud computing resources include low-latency computing resource types and high-throughput storage resource types;

[0014] According to the node connection strength of the required change path topology, a corresponding type combination of cloud computing resources is matched from the mapping relationship table, and an initial allocation ratio is generated based on the initial proportion of the low-latency computing resource type and the high-throughput storage resource type in the type combination;

[0015] Based on the distribution density of the priority weights, dynamically modifying the resource type ratio in the initial allocation ratio to generate a modified allocation ratio;

[0016] The matching degree between the revised allocation ratio and the node connection strength is calculated, and the revised allocation ratio whose matching degree meets the preset conditions is associated with the software development task execution order corresponding to the priority weight and stored to generate a software development task scheduling instruction.

[0017] Optionally, establishing a mapping relationship table between priority weights in the marked time series data and types of cloud computing resources includes:

[0018] Dividing the priority weights in the marked time series data into a high priority weight value interval and a low priority weight value interval according to a numerical range;

[0019] Based on the task completion time parameter and the data storage capacity parameter extracted from the marked time series data, a first correspondence between the high priority weight value interval and the low latency computing resource type in the cloud computing resources is established, and a second correspondence between the low priority weight value interval and the high throughput storage resource type in the cloud computing resources is established;

[0020] The first corresponding relationship and the second corresponding relationship are combined into a mapping relationship table in the order of the numerical ranges of the priority weights.

[0021] Optionally, dynamically revising the resource type proportion in the initial allocation ratio based on the distribution density of the priority weights to generate a revised allocation ratio includes:

[0022] Calculating dynamic adjustment coefficients for low-latency computing resource types and high-throughput storage resource types based on a ratio of distribution densities of a high-priority weight numerical interval to a low-priority weight numerical interval in the priority weights;

[0023] Proportionally adjusting the proportion of the low-latency computing resource type in the initial allocation ratio based on the dynamic adjustment coefficient to generate a first adjusted proportion, and reversely compensating the proportion of the high-throughput storage resource type in the initial allocation ratio based on the dynamic adjustment coefficient to obtain a second adjusted proportion;

[0024] The first adjusted ratio and the second adjusted ratio are merged. When the merged result meets the preset resource allocation interval, the merged result is output as the revised allocation ratio. When the merged result exceeds the preset resource allocation interval, the merged result scaled proportionally according to the preset ratio is output as the revised allocation ratio.

[0025] Optionally, performing multi-dimensional feature analysis on the time series data through the cloud computing platform, extracting the time distribution characteristics of the requirement change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generating a software development requirement evolution map including requirement priority distribution characteristics and requirement change path topology, including:

[0026] Dividing the time series data into time segments through the cloud computing platform, and extracting the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user comment sentiment tendency parameter based on the time segments;

[0027] Superimposing the time distribution feature and the correlation strength, and dynamically allocating a demand priority weight value according to a proportional relationship between the time distribution feature and the correlation strength in the superimposed result to generate a demand priority distribution feature;

[0028] Based on the demand priority weight values ​​in the demand priority distribution characteristics, constructing connection relationships between adjacent time segments in the demand change path, and continuously integrating the connection relationships according to time sequence and intensity distribution to generate a demand change path topology;

[0029] The demand priority distribution characteristics and the demand change path topology are jointly mapped through the cloud computing platform to generate a software development demand evolution map.

[0030] Optionally, superimposing the time distribution feature and the association strength, and dynamically allocating a demand priority weight value according to a proportional relationship between the time distribution feature and the association strength in the superimposed result to generate a demand priority distribution feature, includes:

[0031] Superimposing the characteristic value of each time segment in the time distribution feature with the correlation value of the corresponding time segment in the correlation strength to generate a combined metric value of the time segment;

[0032] determining a proportional relationship between a characteristic value of the time distribution feature and the correlation value based on a distribution interval of the combined metric value, and mapping the proportional relationship into an interaction intensity of the time segment;

[0033] Convert the interaction strength into a dynamic adjustment factor according to a preset ratio, and generate an initial weight value for each time segment in combination with the characteristic value of the corresponding time segment. Adjust the initial weight value proportionally according to the interval range of the initial weight value to obtain the demand priority weight value;

[0034] Based on the time series arrangement order of the demand priority weight values, the difference in weight values ​​of adjacent time segments is corrected for continuity to obtain the demand priority distribution characteristics.

[0035] Optionally, performing priority weight marking processing on the time series data according to the mapping relationship between the optical signal transmission channel and the demand priority distribution feature to generate marked time series data includes:

[0036] determining configuration parameters of the optical signal transmission channel according to a matching relationship between node weights in the software development requirement evolution graph and weight values ​​in the requirement priority distribution feature;

[0037] According to the mapping relationship between the configuration parameters and the weight values, the parameter change of the configuration parameters is converted into the priority weight tag value of the corresponding time segment in the time series data, and the priority weight tag value is dynamically associated with the data flow of the corresponding time segment in the time series data to generate marked time series data.

[0038] In a second aspect, the present application provides a software development system based on the Internet and cloud computing, comprising:

[0039] The acquisition module collects time series data of software development demand feedback in real time through the Internet and transmits the time series data to the cloud computing platform. The time series data includes demand change frequency parameters and user comment sentiment tendency parameters;

[0040] An extraction module performs multi-dimensional feature analysis on the time series data through the cloud computing platform, extracts the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generates a software development demand evolution map including demand priority distribution characteristics and demand change path topology;

[0041] an allocation module that dynamically allocates optical signal transmission channels adapted to the demand priority distribution characteristics based on the node weights in the software development demand evolution graph;

[0042] a marking module, performing priority weight marking processing on the time series data according to a mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics, and generating marked time series data;

[0043] The generation module dynamically binds the priority weights in the marked time series data with the node connection strength of the demand change path topology according to the priority weights and generates software development task scheduling instructions.

[0044] In a third aspect, an embodiment of the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a software development method based on the Internet and cloud computing as described in the first aspect above.

[0045] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a software development method based on the Internet and cloud computing as described in the first aspect.

[0046] In an embodiment of the present application, time series data of software development demand feedback is collected in real time through the Internet, and the time series data is transmitted to a cloud computing platform, the time series data including a demand change frequency parameter and a user comment sentiment tendency parameter; the time series data is subjected to a multi-dimensional feature analysis by the cloud computing platform, the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user comment sentiment tendency parameter are extracted, and a software development demand evolution map including a demand priority distribution feature and a demand change path topology is generated; based on the node weights in the software development demand evolution map, an optical signal transmission channel adapted to the demand priority distribution feature is dynamically allocated; according to the mapping relationship between the optical signal transmission channel and the demand priority distribution feature, the time series data is subjected to priority weight marking processing to generate marked time series data; according to the priority weight in the marked time series data, combined with the node connection strength of the demand change path topology, the priority weight is dynamically bound to the allocation ratio of cloud computing resources to generate a software development task scheduling instruction.

[0047] The technical solution of this application has the following beneficial effects:

[0048] This application collects demand change frequency and user comment sentiment parameters in real time through the Internet to ensure the timeliness and integrity of demand feedback data and provide original data support for dynamic analysis. Multi-dimensional feature extraction is performed on time series data to construct demand priority distribution characteristics and change path topology to achieve quantitative expression and visual presentation of dynamic evolution of demand. Based on the dynamic allocation of adaptive optical signal transmission channels based on node weights, the physical properties of photonic integrated circuits are used to achieve the optimal matching of demand priority and physical resources. Through the mapping rules of optical signal channels and priority distribution, the original time series data is converted into structured data with weight labels, and a standardized interactive interface between data layer and physical layer resources is established. Combined with the priority weight and the connection strength of topological nodes, the cloud computing resource allocation ratio is dynamically bound to generate scheduling instructions that take into account both efficiency and stability.

[0049] Furthermore, a mapping table between priority weights and resource types is established, matching resource type combinations based on node connection strength to generate an initial allocation ratio. A matching calculation verifies the compatibility of the modified ratio with the topology strength, ultimately generating scheduling instructions. Through dynamic mapping and matching verification mechanisms, resource allocation ratios are precisely adjusted in real time based on demand priority and topology strength, breaking through traditional static allocation models. This ensures maximum resource utilization while meeting the rigid requirements of millisecond-level response and high concurrent loads in agile development scenarios.

[0050] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0052] Figure 1 A flowchart of a software development method based on the Internet and cloud computing provided by the present application is shown;

[0053] Figure 2 A schematic diagram of the structure of a software development system based on the Internet and cloud computing provided by the present application is shown;

[0054] Figure 3 A schematic structural diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0055] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0056] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0057] Research has found that in current agile software development scenarios, demand dynamics change frequently and unpredictably, and traditional technical solutions have significant limitations in dealing with complex demand evolution. Existing solutions that combine machine learning-based demand forecasting models with cloud computing resource scheduling frameworks have achieved partial automation, but still rely on static model training and preset rules, resulting in insufficient real-time performance, weak correlation, and rigid resource scheduling. For example, when generating demand path topology, existing solutions often use containerized expansion strategies, but do not combine the mapping relationship between node weights and optical signal channels, resulting in a disconnect between resource allocation ratios and demand priorities, ultimately affecting development efficiency and resource utilization. In addition, existing technologies for feature extraction of time series data are limited to single-dimensional analysis, and fail to construct a comprehensive evolutionary map that includes priority distribution and changing path topology, further limiting the accuracy of task scheduling.

[0058] In response to the above-mentioned defects, this application proposes a software development method based on the Internet and cloud computing. This solution collects time series data of demand feedback in real time through the Internet, uses the cloud computing platform to perform multi-dimensional feature analysis, extracts time distribution characteristics and correlation strength, and generates an evolutionary map containing demand priority distribution and change path topology. Furthermore, based on the weight of the graph node, the optical signal transmission channel that adapts to the demand priority is dynamically allocated, and the time series data is weighted by the mapping relationship between the optical signal channel and the priority. Finally, the weight of the labeled data is combined with the node connection strength to dynamically bind the cloud computing resource allocation ratio to generate task scheduling instructions. This solution solves the problems of response lag, weak correlation and rigid resource allocation in the existing technology through real-time data-driven demand evolution modeling and optical signal-level resource adaptation mechanism, significantly improves demand response efficiency and resource utilization, and provides a dynamic optimization solution for complex agile development scenarios.

[0059] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0060] Figure 1 A flowchart of a software development method based on the Internet and cloud computing is provided for the embodiment of the present application. Figure 1 As shown, the method includes:

[0061] 101. Collecting time series data of software development demand feedback in real time through the Internet, and transmitting the time series data to a cloud computing platform, wherein the time series data includes a demand change frequency parameter and a user comment sentiment tendency parameter;

[0062] In this step, the time series data refers to parameters including demand change frequency and user review sentiment tendency.

[0063] The demand change frequency parameter refers to the quantitative demand dynamics, which includes three sub-parameters: the number of changes, the change time interval, and the change amplitude.

[0064] The sentiment tendency parameter of user reviews refers to the numerical score generated through sentiment analysis, where positive values ​​indicate positive feedback and negative values ​​indicate negative feedback.

[0065] The Internet is a large-scale network system composed of computer networks around the world connected by common protocols, enabling global sharing of information.

[0066] The cloud computing platform is a computing, storage and network resource service provided based on the Internet. It abstracts hardware resources into "cloud resources" that can be used on demand through virtualization technology.

[0067] In an embodiment of the present application, first, the change log in the demand management system is collected in real time through a tracking tool, and the open interface of the social media platform is synchronously called to obtain the original text of the user comments; the change log is parsed in time series, and the number of demand changes, the standard deviation of the interval time and the change amplitude within the preset time window are counted to generate a structured demand change frequency parameter; secondly, the user comment text is subjected to sentiment analysis, and the positive and negative sentiment scores of each comment are calculated using a pre-trained sentiment classification model, and the proportion of positive emotions and the intensity of negative sentiment fluctuations in each time window are counted to generate the user comment sentiment tendency parameter; the two types of parameters are encapsulated as time series data and transmitted to the cloud computing platform in real time through the message queue middleware.

[0068] In actual applications on e-commerce platforms, users submit change requests through online forms, such as "add nighttime sales" or "optimize the payment process." They also leave feedback on social media and in-app comment sections, such as "nighttime sales are too slow" or "payment system frequently freezes." The system captures this data in real time using a distributed crawler framework, employing protocols to ensure data security. A sliding window algorithm is used to calculate the frequency of change requests, dynamically calculating the number of new requests per hour and incorporating an exponential smoothing model to predict future trends. A sentiment analysis model classifies comments, such as "nighttime sales are too slow," as having negative sentiment. Finally, the extracted frequency values ​​and sentiment scores are packaged and encrypted for transmission to a cloud computing platform for storage.

[0069] 102. Perform multi-dimensional feature analysis on the time series data through the cloud computing platform to extract the time distribution characteristics of the requirement change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generate a software development requirement evolution map including requirement priority distribution characteristics and requirement change path topology;

[0070] In this step, the time distribution characteristics refer to the fluctuation pattern of the demand change frequency on the time axis, including the peak distribution pattern and interval characteristics.

[0071] Association strength refers to the overlap density and conflict intensity of positive and negative emotional parameters on the time axis.

[0072] The demand priority distribution feature refers to the node weight set generated by weighted fusion of demand changes and emotional conflicts.

[0073] The requirement change path topology is a graph structure that shows the evolution relationship of requirements. Nodes represent requirements and edges represent evolution paths.

[0074] In this application example, the cloud computing platform first uses a time series decomposition algorithm to isolate the temporal distribution characteristics of the demand change frequency parameter. Then, dynamic time warping is used to calculate the correlation strength between user review sentiment and the demand change frequency parameter. Subsequently, a clustering algorithm is used to divide demand priorities into hierarchies, construct a demand change path topology, and finally integrate and generate a visual demand evolution map.

[0075] After receiving time series data, the e-commerce platform's cloud computing platform first uses a time series decomposition algorithm to break down demand change frequency into periodic holiday demand surges, long-term growth trends, and residuals. This model then predicts future demand changes. The Pearson correlation coefficient is used to analyze the correlation between negative user sentiment and demand frequency, finding that demand change frequency increases significantly when negative sentiment scores are low. Subsequently, demand priorities are calculated and clustered into high-priority demands. A graph neural network constructs a demand change path topology, with node weights representing priority scores and edge weights representing the co-occurrence frequency of change events, identifying key demand nodes. The resulting evolutionary graph identifies high-priority demands and their dependent paths.

[0076] 103. Based on the node weights in the software development requirement evolution graph, dynamically allocate an optical signal transmission channel adapted to the requirement priority distribution characteristics;

[0077] In this step, the node weight refers to the quantitative priority value of the node in the demand priority distribution feature.

[0078] Dynamic allocation is a common term in many fields, usually referring to the flexible adjustment of resource allocation strategies according to real-time demand or environmental changes during operation to achieve optimal efficiency or adaptability.

[0079] Optical signal transmission channel refers to a dynamically reconfigurable physical communication channel in a photonic integrated circuit that supports wavelength division multiplexing.

[0080] In an embodiment of the present application, first, the weight information of each node is extracted from the constructed software development demand evolution map. These nodes represent different functional modules or task units, and their weights reflect the importance of the corresponding demand in the current development stage and the expected resource consumption. Then, according to the node weight distribution characteristics, a dynamic resource scheduling algorithm is used to quantitatively analyze the demand priority and identify a set of task nodes on a high-priority, critical path. Subsequently, under the cloud computing network architecture, the data communication requirements involved in these high-priority tasks are mapped to the underlying physical transmission layer, and a matching evaluation is performed based on the bandwidth capacity and current load status of the optical signal transmission channel. Finally, based on the matching results, an optical signal transmission channel with corresponding service quality assurance is dynamically allocated to the critical task nodes to ensure the data interaction efficiency and system responsiveness of high-priority demands during the development process, thereby realizing a closed-loop control process from demand evolution analysis to communication resource allocation.

[0081] During a major e-commerce promotion, the demand evolution graph showed that the "nighttime promotion feature" demand had a high priority score, placing it on a critical change path. The system prioritized optical signal channels based on node weights, assigning high-priority requests to specific wavelengths. A controller monitored channel load in real time and, in conjunction with a dynamic bandwidth allocation algorithm, increased the default bandwidth to meet the low-latency transmission requirements of high-priority requests. A wavelength selective switch dynamically switched transmission paths, ensuring data bypassed congested nodes. Forward error correction technology also reduced the bit error rate. An optical spectrum analyzer monitored channel quality in real time, ensuring that high-priority request data was transmitted quickly to the development team, while lower-priority requests were transmitted at lower rates via lower-priority channels, creating a differentiated transmission strategy.

[0082] 104. Perform priority weight marking processing on the time series data according to the mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics to generate marked time series data;

[0083] In this step, the mapping relationship refers to the correspondence rule between the physical characteristics of the optical channel and the priority weight.

[0084] Priority weight labeling is a technique that quantifies the priority and weight of tasks, data, or resources and encodes them into identifiable labels or identifiers to guide the system's decision-making sequence and resource allocation when processing complex tasks.

[0085] Labeled time series data refers to a structured dataset with priority labels attached to the original time series data.

[0086] In this embodiment, a mapping relationship between wavelength ranges and demand priority distribution characteristics is first established based on the physical characteristics of the allocated optical signal transmission channel. Priority weighting is then applied to each data segment in the original time series data, with the tag value determined by the weight interval matched in the mapping table. The tagged time series data is then sorted by timestamp and written into a time series database to generate tagged time series data.

[0087] During nighttime promotions on e-commerce platforms, the mapping rules between optical signal channels and demand priorities are automatically matched to demand data. Because these nighttime promotion requests occupy high channels, they are tagged with a high-priority tag. Fields are inserted into the original time series data to create structured tagged data. A hash check ensures data integrity, and blockchain technology writes the tagged data to a distributed ledger to prevent tampering. Smart contracts trigger subsequent resource allocation processes, automatically initiating resource scheduling instructions when a high-priority tag is detected. The tagged time series data is then verified and transmitted to the cloud computing platform.

[0088] 105. Based on the priority weight in the marked time series data and the node connection strength of the demand change path topology, the priority weight is dynamically bound to the allocation ratio of cloud computing resources to generate a software development task scheduling instruction.

[0089] In this step, the node connection strength refers to the weight of the edge in the requirement change path topology, which represents the strength of the evolutionary relationship between requirements.

[0090] Priority weight refers to the value obtained by comprehensively evaluating the importance and urgency of tasks or requirements through quantitative indicators during task scheduling or resource allocation.

[0091] Software development task scheduling instructions refer to commands or policies that are automatically generated or manually configured during the development process to control the order of task execution, resource allocation, and dependencies based on priority weights and system resource status.

[0092] Dynamic binding refers to a mechanism in which the resource allocation ratio is dynamically adjusted according to the priority weight and connection strength.

[0093] In the embodiment of the present application, a mapping table between priority weights and cloud computing resources is first constructed. The initial resource allocation ratio is calculated based on the sum of the node connection strengths in the required change path topology. The initial ratio is dynamically adjusted based on the distribution density of the priority weights within the time window. The matching degree between the adjusted ratio and the node connection strength is verified by correlation coefficient calculation. If the preset conditions are met, a resource allocation instruction is generated; otherwise, the ratio is readjusted. The final output is a software development task scheduling instruction containing instance type, quantity, and execution order.

[0094] During a major e-commerce promotion, the resource allocation for nighttime promotions, which fell on the critical path, was dynamically adjusted from an initial low allocation to a high allocation, triggering a scheduling instruction to expand the cluster. A reinforcement learning algorithm monitored the cloud computing platform's memory and storage usage in real time. If it detected insufficient resources for high-priority tasks, it automatically increased resource allocation. The container orchestration engine dynamically expanded the deployment scale based on the scheduling instructions, ensuring the development team could complete feature iterations quickly. The service mesh implemented traffic control and load balancing, ensuring that resources for high-priority tasks were not preempted by lower-priority tasks, ultimately forming a closed-loop optimized resource scheduling process.

[0095] In summary, steps 101 to 105 achieve precise matching of demand priorities with physical resources by collecting demand data in real time and constructing an evolution graph, combined with dynamic allocation of optical signal channels by photonic integrated circuits. Furthermore, through weighted labeling and dynamic binding, high-priority demands are prioritized for low-latency resources, while global resource allocation is optimized based on topological connectivity strength. This reduces system response latency and improves resource utilization, effectively supporting agile development and real-time operations and maintenance under high-concurrency demands.

[0096] In order to solve the problem of insufficient matching accuracy between cloud computing resource allocation and dynamic demand in agile development scenarios, the solution analyzes the priority weights in the marked time series data, combines the node connection strength of the demand change path topology, establishes a matching relationship table between priority and resource type, and dynamically adjusts the resource allocation ratio. A dynamic correction mechanism is introduced to optimize resource allocation based on the priority distribution density to ensure that critical tasks obtain sufficient computing and storage support, while improving resource utilization efficiency and task scheduling accuracy. In some embodiments, step 105 dynamically binds the priority weights to the allocation ratio of cloud computing resources based on the priority weights in the marked time series data and the node connection strength of the demand change path topology to generate software development task scheduling instructions, including:

[0097] 201. Establish a mapping relationship table between priority weights in the marked time series data and types of cloud computing resources, wherein the types of cloud computing resources include low-latency computing resource types and high-throughput storage resource types;

[0098] In step 201, priority weight refers to a metric that quantifies the importance of a task, typically calculated from factors such as task urgency, dependencies, and resource consumption. Low-latency computing resource types are suitable for tasks that are sensitive to response time. High-throughput storage resource types are suitable for tasks that require large amounts of data reading and writing. A mapping table is a rule table that associates priority weights with cloud computing resource types, such as "high-priority tasks to low-latency computing resources."

[0099] In this embodiment of the application, dynamic mapping rules are established by extracting the priority weight and timestamp data of the task and combining it with the characteristics of the cloud computing resource type. First, the marked time series data is obtained from the task management system and the characteristics are analyzed. Second, the cloud computing resource classification standard is defined based on the task requirements. For example, the low-latency computing resource type is suitable for real-time processing scenarios, and the high-throughput storage resource type is suitable for batch storage scenarios. Finally, a decision tree is used to analyze historical task data to determine the association between priority weight and resource type, and a mapping relationship table is generated.

[0100] 202. According to the node connection strength of the required change path topology, a corresponding type combination of cloud computing resources is matched from the mapping relationship table, and an initial allocation ratio is generated based on the initial proportion of the low-latency computing resource type and the high-throughput storage resource type in the type combination;

[0101] In step 202, node connection strength measures the closeness of inter-task dependencies, such as a value calculated by the number of task collaborations and the amount of data interaction. Type combination refers to the combination of different resource types based on task requirements during cloud computing resource allocation. The initial allocation ratio refers to the initial resource type allocation ratio determined based on the mapping table and node connection strength.

[0102] In an embodiment of the present application, first, the topology of the demand change path and its dependencies are extracted, a task association network is constructed, and a graph neural network is used to calculate the node connection strength. For example, the connection strength of high-frequency data interaction tasks is higher. Subsequently, a corresponding combination of cloud computing resource types is matched for each task according to a mapping relationship table, and the proportion of low-latency computing resource types and high-throughput storage resource types is adjusted based on the connection strength. For example, the proportion of low-latency resources shared by high-connection-strength tasks is increased. Finally, the initial allocation ratio is generated.

[0103] 203. Based on the distribution density of the priority weights, dynamically modify the resource type ratio in the initial allocation ratio to generate a modified allocation ratio;

[0104] In step 203, the distribution density of priority weights describes the degree of concentration of task priority weights along the timeline, such as periods of high-priority bursts. Dynamic correction involves adjusting the initial allocation ratio based on the distribution density, such as increasing the proportion of low-latency resources during high-density periods. Resource type ratios refer to the proportion of different resource types within the total resources during resource allocation. Allocation ratios are the process of further refining resource type ratios to specific tasks or users, specifically the actual proportion of resource types allocated to each task.

[0105] In this embodiment, a sliding window algorithm is first used to calculate the distribution density of high, medium, and low priority weights per unit time. For example, during a major promotion, the proportion of high-priority tasks surges. Secondly, a correction rule is formulated based on this change in distribution density. Finally, the correction rule is combined with the initial allocation ratio to generate a revised allocation ratio.

[0106] 204. Calculate the matching degree between the revised allocation ratio and the node connection strength, associate and store the revised allocation ratio whose matching degree meets the preset conditions with the software development task execution order corresponding to the priority weight, and generate a software development task scheduling instruction.

[0107] In step 204, the matching calculation refers to measuring the degree of compatibility between the revised allocation ratio and the node connection strength, for example, by quantifying the difference using Euclidean distance or cosine similarity. The preset condition refers to a matching threshold, which is used to determine whether the revised allocation ratio is feasible. The software development task execution sequence refers to the logical or chronological order in which various development tasks are executed during the software development process, according to established processes or rules. Software development task scheduling instructions are instructions or rules that dynamically plan and control the execution time, resource allocation, and priority of development tasks through technical means or tools.

[0108] In an embodiment of the present application, first, the Euclidean distance or cosine similarity is used to quantify the matching degree between the corrected allocation ratio and the node connection strength, for example, the difference between the corrected ratio and the connection strength is calculated. Secondly, a preset threshold is set, and if the matching degree meets the standard, the corrected allocation ratio is accepted, otherwise the corrected allocation ratio is iteratively adjusted until the condition is met. Finally, the successfully matched corrected allocation ratio is associated and stored with the execution order of the software development task in order of priority, and a software development task scheduling instruction is generated. For example, low-latency resources are preferentially allocated to process high-priority tasks, thereby achieving coordinated optimization of resource allocation and task scheduling.

[0109] Here's a specific example:

[0110] During major e-commerce promotions, the platform needs to handle a massive volume of orders and inventory synchronization tasks. The platform prioritizes order tasks, such as orders with successful payments having a higher priority weight and unpaid orders having a lower priority weight. A mapping table is established, with high-priority tasks corresponding to a higher proportion of low-latency resources. The topology of the demand change path is analyzed, revealing that the node connection strengths for payment orders and inventory deduction tasks are high. A low-latency resource combination is matched and an initial allocation ratio is generated. A significant increase in the density of high-priority tasks during the promotion period is detected, and the ratio is adjusted to a higher proportion of low-latency resources. The compatibility of the revised allocation ratio with the node connection strength is verified. Once the matching degree is met, a scheduling instruction is generated, prioritizing the allocation of low-latency resources to process high-priority orders to ensure real-time response.

[0111] In summary, steps 201 to 204 maximize resource utilization, optimize task response speed, improve adaptability to complex scenarios, and enhance automated decision-making capabilities by dynamically binding priority weights and cloud computing resource allocation ratios, combined with demand change path topology analysis and real-time task density detection.

[0112] To further improve the resource scheduling efficiency of e-commerce platforms, the solution divides priority weights into high and low intervals, and combines parameters such as task completion time and data storage capacity to establish corresponding relationships between high-priority tasks and low-latency resources, and low-priority tasks and high-throughput resources. Ultimately, a unified mapping relationship table is formed, providing clear structure and logically rigorous data support for subsequent resource scheduling, improving the rationality and adaptability of resource allocation. In some embodiments, the establishment of a mapping relationship table between the priority weights in the marked time series data and the types of cloud computing resources in step 201 includes:

[0113] 301. Divide the priority weights in the marked time series data into a high priority weight value interval and a low priority weight value interval according to a numerical range;

[0114] In step 301, priority weighting is used to quantify the criticality of tasks based on business needs, categorizing them into high-priority and low-priority groups by setting numerical ranges. This categorization is based on a comprehensive assessment of task timeliness, resource consumption, and business value.

[0115] In an embodiment of the present application, the system extracts priority weight tag values ​​by collecting the marked time series data of various tasks in the e-commerce scenario. The calculation of the weight tag value is based on business rules. For example, the weight of real-time transaction tasks includes factors such as payment success rate and user waiting time, while the weight of offline tasks is related to data throughput and processing timeliness. By analyzing the distribution of historical data, the dynamic quantile method is used to determine the threshold for dividing the priority weight interval. For example, during a promotional activity, the system detects that the weight tag values ​​of order payment tasks are concentrated in a higher interval, then this interval is defined as a high priority weight value interval, and the remaining tasks are low priority weight value intervals. During the division process, the system introduces a sliding window mechanism to recalculate the threshold every five minutes to ensure adaptation to traffic fluctuations. At the same time, the task completion time and storage capacity are normalized by the feature engine to avoid the influence of single indicator deviation on the accuracy of interval division.

[0116] 302. Based on the task completion time parameter and the data storage capacity parameter extracted from the marked time series data, establish a first correspondence between the high-priority weight value interval and the low-latency computing resource type in the cloud computing resources, and establish a second correspondence between the low-priority weight value interval and the high-throughput storage resource type in the cloud computing resources;

[0117] In step 302, the data storage capacity parameter refers to the total amount of data that a storage device or system can accommodate, typically measured in bytes. Low-latency computing refers to a computing model that completes data processing and response in an extremely short time, typically measured in nanoseconds or milliseconds. High-throughput storage resource types refer to storage systems that can efficiently handle large-scale concurrent data reads and writes, and are suitable for data-intensive scenarios.

[0118] In an embodiment of the present application, the resource scheduler monitors the performance indicators of various nodes in the cloud computing cluster in real time. For high-priority tasks, the system combines its task completion time parameters to screen out low-latency nodes whose current load is lower than the warning value, and establishes a strong binding relationship to obtain a first corresponding relationship. For example, the order payment task needs to ensure that the payment link delay is short, and the system allocates computing nodes equipped with solid-state drives and high-speed networks for it. For low-priority tasks, according to their data storage capacity parameters, high-throughput storage resource types with high disk throughput and parallel processing capabilities are matched to obtain a second corresponding relationship. At the same time, the system adopts a flexible quota mechanism. When the utilization rate of the high-priority task resource pool exceeds a certain amount, it automatically triggers the temporary degradation and release of low-priority task resources to ensure the supply of core business resources.

[0119] 303. Combine the first corresponding relationship and the second corresponding relationship into a mapping relationship table in the order of the numerical ranges of the priority weights.

[0120] In step 303, priority weight refers to a numerical indicator used to quantify the importance of different tasks or data in resource allocation or task scheduling. A mapping relationship table is a structured data table used to define association rules between different entities or resources.

[0121] In an embodiment of the present application, the data orchestration module integrates the priority intervals generated by the first corresponding relationship with the mapping rules of the second corresponding relationship. First, the intervals are sorted in descending order according to the priority weights to form a tree-like hierarchical structure to obtain a mapping relationship table. For example, the high-priority interval serves as the root node, and the low-priority interval serves as the child node, and the child node can inherit some resource attributes of the parent node. Secondly, a specific resource type and quota are associated with each interval. When a certain type of resource fails, it automatically switches to an alternative resource pool according to the backup strategy. In addition, the mapping relationship table supports dynamic expansion. When a new resource type is added, the mapping relationship is automatically updated through the label matching mechanism without manual intervention.

[0122] Here's a specific example:

[0123] During a major promotion, an e-commerce platform's order payment service's weight tag value jumped to a higher priority range due to a surge in traffic. Based on the rules of Phase 1, the system assigned it to a low-latency computing resource pool, ensuring that millions of transaction requests per second could be completed quickly. Simultaneously, user behavior log analysis tasks, which were in a lower priority range, were dispatched to a high-throughput storage cluster for batch processing. When a sudden peak in the payment service caused resource shortages, the Phase 3 mapping table triggered a resource borrowing mechanism, temporarily allocating some offline task resources to the payment service. After the peak subsided, capacity expansion was automatically triggered to fill the gap in the offline resource pool. Through these three coordinated phases, the platform ensured the stability of the payment chain while improving overall resource utilization.

[0124] In summary, steps 301 to 303 achieve intelligent and flexible resource scheduling for e-commerce platforms by dynamically dividing task priority intervals, establishing dual-track resource mapping rules, and building a global relationship table. Core businesses are guaranteed low-latency resources, non-critical tasks fully utilize high-throughput resources, resource allocation conflicts are reduced, and the idle rate is significantly reduced. At the same time, the adaptive mechanism of the mapping table effectively responds to traffic fluctuations and hardware failures, improving system robustness. In actual applications, after adopting this method, an e-commerce platform has improved resource scheduling efficiency during peak hours and reduced infrastructure costs, providing sustainable optimization support for complex business scenarios.

[0125] In order to solve the problem of resource scramble and idleness during peak business hours caused by the rigid resource allocation of e-commerce platforms, the solution calculates the dynamic adjustment coefficient of resource types based on the distribution density of priority weights, and accordingly enhances and compensates the initial allocation ratio. At the same time, a boundary constraint mechanism for the resource allocation interval is introduced, and proportional scaling is automatically implemented when the limit is exceeded to ensure the security of resource regulation and the stability of system operation, so that resource allocation is more in line with the execution characteristics of actual development tasks. In some embodiments, the distribution density of the priority weights described in step 203 is used to dynamically correct the proportion of resource types in the initial allocation ratio to generate a corrected allocation ratio, including:

[0126] 401. Calculate dynamic adjustment coefficients for low-latency computing resource types and high-throughput storage resource types based on a ratio of distribution densities of a high-priority weight value interval to a low-priority weight value interval in the priority weights;

[0127] In step 401, the dynamic adjustment coefficient is a numerical factor that reflects the difference in the distribution density of high-priority and low-priority tasks, guiding the dynamic adjustment of resource allocation. The distribution density ratio refers to the ratio of the number of tasks in the high-priority task weight range to the number of tasks in the low-priority range, reflecting the intensity of competition for resources between the two types of tasks.

[0128] In an embodiment of the present application, the resource scheduling system counts the distribution of the number of tasks in the high-priority and low-priority intervals in the current task queue in real time. For example, at the beginning of a promotion, the number of high-priority tasks such as order payment surges, and its distribution density ratio may reach three to one. The system adopts a sliding window mechanism to calculate the ratio every two minutes and convert it into a dynamic adjustment coefficient through a nonlinear function. When the density of high-priority tasks is significantly higher than that of low-priority tasks, the coefficient increases exponentially, and vice versa, the fluctuation is suppressed by a logarithmic function. In the specific calculation, the system introduces an elastic buffer factor to avoid coefficient jumps due to instantaneous task surges. For example, when the ratio exceeds the preset threshold, the system automatically enables the historical mean smoothing algorithm, and weights the current coefficient with the average of the previous five minutes to ensure the stability of the adjustment process.

[0129] 402. Adjust the proportion of the low-latency computing resource type in the initial allocation ratio based on the dynamic adjustment coefficient to generate a first adjusted proportion, and reversely compensate the proportion of the high-throughput storage resource type in the initial allocation ratio based on the dynamic adjustment coefficient to obtain a second adjusted proportion.

[0130] In step 402, the initial allocation ratio is a pre-set baseline allocation ratio of low-latency resources to high-throughput resources. The first adjustment ratio is a positive adjustment of the low-latency resources based on the dynamic coefficient, and the second adjustment ratio is a negative compensation for the high-throughput resources. Both must meet the total resource conservation constraint.

[0131] In an embodiment of the present application, the system uses the initial allocation ratio as a benchmark, with a higher proportion of low-latency resources and a lower proportion of high-throughput resources, to dynamically adjust the coefficient to the proportion of low-latency resources. For example, when the coefficient is relatively small, the proportion of low-latency resources increases, while the proportion of high-throughput resources decreases through a reverse compensation mechanism. During the compensation process, the system performs a secondary correction based on real-time resource load data. If it is detected that there are idle resources in the high-throughput node cluster, it is allowed to exceed the compensation lower limit and temporarily borrow some resources to support low-latency tasks. The adjustment process adopts a dual-queue verification mechanism. The main queue performs the proportion adjustment, and the standby queue simulates the resource utilization rate after adjustment. If the simulation result shows that the resource exceeds the limit, the proportion rollback mechanism is triggered.

[0132] 403. Merge the first adjusted ratio and the second adjusted ratio. When the merged result meets the preset resource allocation interval, output the merged result as a revised allocation ratio. When the merged result exceeds the preset resource allocation interval, output the merged result after geometric scaling according to the preset ratio as a revised allocation ratio.

[0133] In step 403, the preset resource allocation range is the minimum resource allocation range required to ensure basic system functionality. Scaling refers to reducing the resource allocation by the same amount as the resource type if the adjustment exceeds the range. The allocation ratio refers to the preset ratio between low-latency computing resources and high-throughput storage resources in the cloud computing resource pool, which guides the initial resource scheduling strategy.

[0134] In an embodiment of the present application, the data arbitration module combines the first adjustment ratio and the second adjustment ratio, and compares them with the preset interval that low-latency resources must not be less than half and high-throughput resources must not be too low. If the merging result is compliant, it is directly sent to the resource scheduler; if it exceeds the interval, the scaling engine is started. For example, after adjustment, the proportion of low-latency resources is high and the proportion of high-throughput resources is low, exceeding the lower limit of high-throughput resources. The system scales the two types of resources synchronously according to the excess range. During the scaling process, the system gives priority to ensuring the resource supply of high-priority tasks and fills the gap through the elastic resource pool. The revised allocation ratio is submitted to each node through distributed transactions to ensure that global resources are viewed as Figure 1 Consistency.

[0135] Here's a specific example:

[0136] During a major sales event on an e-commerce platform, a midnight flash sale triggered a significant increase in the density of high-priority tasks. After calculating a dynamic adjustment factor, the system adjusted the initial allocation ratio (low-latency resources predominantly, high-throughput resources secondary) to a preponderance of low-latency resources and a minimal share of high-throughput resources. Because the proportion of high-throughput resources fell below the preset minimum guarantee, the scaling engine adjusted the allocation to moderately adjust the low-latency resource ratio back to a higher range and increase the high-throughput resources above the baseline guarantee. During execution, the resource scheduler detected underutilized resource pools in the high-throughput node cluster and temporarily allocated some of these idle resources to low-latency tasks, resulting in a final allocation ratio where low-latency resources were near full capacity and high-throughput resources maintained at the minimum operating threshold. After the sales peak, the system automatically restored resource allocation to the initial baseline and fully returned the temporarily allocated resources to the high-throughput cluster to handle the backlog of log analysis tasks.

[0137] In summary, steps 401 to 403 achieve precise and adaptive resource allocation for e-commerce platforms through dynamic perception of task distribution density, elastic adjustment of resource ratios, and intelligent scaling mechanisms. Core businesses receive sufficient low-latency resource guarantees during traffic peaks, non-critical tasks flexibly utilize idle resources, and overall resource utilization is significantly improved. After a leading e-commerce company applied this method, the resource scheduling response speed during major promotions was greatly improved, resource contention conflicts were significantly reduced, and annual operation and maintenance costs were effectively reduced. In particular, in instantaneous high-concurrency scenarios, the system demonstrated excellent elastic scalability and stability.

[0138] The solution generates a comprehensive influence index by extracting the temporal distribution characteristics of the demand change frequency and the correlation strength of the user comment sentiment tendency, dynamically assigns demand priority weights, and constructs a demand change path topology. Finally, the priority distribution characteristics and the path topology are jointly mapped to comprehensively depict the entire process of demand evolution, providing knowledge support and visual guidance for resource scheduling and task sorting. In some embodiments, step 102 involves performing a multi-dimensional feature analysis on the time series data through the cloud computing platform, extracting the temporal distribution characteristics of the demand change frequency parameters and the correlation strength of the user comment sentiment tendency parameters, and generating a software development demand evolution map containing the demand priority distribution characteristics and the demand change path topology, including:

[0139] 501. Divide the time series data into time segments through the cloud computing platform, and extract the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user review sentiment tendency parameter based on the time segments;

[0140] In step 501, time segmentation involves dividing the business cycle into continuous time units based on user behavior fluctuations. This is used to capture the phased characteristics of demand changes. Time distribution characteristics reflect the intensity and evolutionary trends of demand change requests within different time periods. Correlation strength refers to the causal relationship between user review sentiment and demand changes.

[0141] In the embodiment of the present application, the cloud computing platform collects demand change records and user comment data in e-commerce scenarios and segments them according to natural time windows. For example, a cycle is divided into several time segments, each of which contains a certain amount of data. The feature extraction engine simultaneously analyzes two dimensions. In the time distribution dimension, the submission frequency, type distribution, and cross-departmental collaboration dependencies of demand change requests in each segment are counted to obtain time distribution features; in the emotional association dimension, the semantic analysis model is used to calculate the correlation strength between keywords in user comments and specific demands.

[0142] 502. Superimpose the time distribution feature and the correlation strength, and dynamically assign a demand priority weight value based on a proportional relationship between the time distribution feature and the correlation strength in the superposition result to generate a demand priority distribution feature;

[0143] In step 502, the demand priority weight is a comprehensive indicator that quantifies the demand's urgency and business value. It is dynamically calculated by combining the temporal distribution characteristics with the emotional association strength. The demand priority distribution characteristics describe the weight ranking and resource utilization ratio of different demands in the global business cycle.

[0144] In the embodiment of the present application, the priority allocation module normalizes the superimposed feature matrix. For example, during the promotional burst period, the demand for payment system expansion has a significantly higher weight than other demands due to its dense time distribution and associated negative user comments. The system introduces an adaptive balancing strategy. When the time distribution characteristics of a demand account for too high a proportion, it automatically reduces the calculation weight of its emotional association strength to avoid a single dimension dominating the priority determination. The final generated demand priority distribution characteristics are presented in the form of a topological graph, where the node size indicates the weight value and the edge weight indicates the dependency relationship between demands.

[0145] 503. Based on the demand priority weight values ​​in the demand priority distribution characteristics, construct connection relationships between adjacent time segments in the demand change path, continuously integrate the connection relationships according to the time sequence and intensity distribution, and generate a demand change path topology;

[0146] In step 503, the demand change path refers to the evolution trajectory of the demand in the time dimension, and the connection relationship describes the logical association of the change in demand priority in adjacent time segments. Continuous integration refers to the construction of a global evolution network across segments based on time sequence and intensity distribution.

[0147] In an embodiment of the present application, the path analysis engine connects the demand priority distribution of each time segment in series. For example, the demand for optimizing the loading of the product details page during the warm-up period, and the demand for expanding the distributed cache due to the surge in user visits during the outbreak period, form a connection relationship between adjacent time segments in the demand change path. The system filters weakly associated paths through a strength threshold and only retains core links with demand priority weight values ​​higher than the critical value. For complex scenarios, a directed acyclic graph structure is used to express the priority inheritance and conflict resolution logic to ensure the traceability and explainability of the demand change path topology.

[0148] 504. The demand priority distribution characteristics and the demand change path topology are jointly mapped through the cloud computing platform to generate a software development demand evolution map.

[0149] In step 504, the requirement change path represents the evolution of requirements over time, and the connections describe the logical associations between changes in requirement priorities in adjacent time segments. The software development requirement evolution graph is a visual knowledge graph that integrates priority distribution and path topology. It is used to describe the dynamic evolution patterns and resource mapping relationships throughout the entire requirement lifecycle.

[0150] In an embodiment of the present application, the software development demand evolution map generation module maps the priority distribution characteristics to node attributes, maps the demand change path topology to edge relationships, and presents them through a three-dimensional space-time coordinate system. For example, high-frequency demands are represented in the software development demand evolution map as coarse-grained links that run through multiple time segments. The color depth of the nodes indicates the strength of the emotional association, and the size indicates the resource consumption. The software development demand evolution map supports interactive exploration. Clicking on a node allows you to drill down to view the associated user review summary, demand document, and implementation status. At the same time, the system has a built-in prediction model that recommends resource pre-allocation plans for future segments based on historical evolution laws.

[0151] Here's a specific example:

[0152] During its annual promotional cycle, an e-commerce platform analyzed demand change data by stratifying it into multiple time segments. During the pre-launch period, user comments focused on "high search latency." The system identified demand changes during this period as primarily focused on expanding search services, with peak sentiment intensity, and assigned it the highest priority. During the peak period, pressure on the payment chain increased dramatically. Demand changes were optimized with distributed transactions, forming a cross-segment dependency path with the pre-launch search demand. The graph generation module mapped the two into parallel links with competing resources and marked warning areas requiring server resource coordination. Based on the graph's prompts, the operations team preemptively divided elastic computing resources into dedicated search and payment pools to ensure core demand conflicts. After the promotional event, the graph revealed a sharp increase in the weight of logistics query demand during the final payment period. The platform adjusted its subsequent version planning accordingly, prioritizing logistics interface optimization as a key task for the next phase.

[0153] In summary, steps 501 to 504 achieve deep coupling between e-commerce demand management and user feedback through time-segment feature extraction, dynamic priority allocation, path topology construction, and graph visualization. The platform can accurately identify the evolutionary paths of high-value demands, avoid fragmented resource allocation, and shorten the demand response cycle. In actual applications, after applying this method to a leading e-commerce platform, the efficiency of version iteration increased, the accuracy of matching demand with user pain points increased, and resource waste was reduced, especially during large-scale promotional activities. This provides systematic support for software engineering management in complex business scenarios.

[0154] In order to further improve the accuracy and business scenario adaptability of the dynamic allocation of demand priorities on e-commerce platforms, the solution generates a combined metric value by superimposing and analyzing the time distribution characteristics and the sentiment tendency parameters of user comments, and converts it into a dynamic adjustment factor to generate a preliminary weight value. It is then normalized and the trend of adjacent segments is smoothed to eliminate the impact of short-term fluctuations, ensuring that the demand priority distribution characteristics have good continuity and stability, and realizing closed-loop reasoning from external data to internal weights. In some embodiments, the time distribution characteristics and correlation strength are superimposed in step 502, and the demand priority weight value is dynamically allocated according to the proportional relationship between the time distribution characteristics and the correlation strength in the superimposed result to generate the demand priority distribution characteristics, including:

[0155] 601. Superimpose the characteristic value of each time segment in the time distribution feature and the correlation value of the corresponding time segment in the correlation strength to generate a combined metric value of the time segment;

[0156] In step 601, the combined metric is a comprehensive indicator that integrates the temporal distribution eigenvalues ​​and the strength of the user sentiment correlation. It is used to quantify the global impact of a demand change within a single time segment. The eigenvalues ​​reflect the intensity and evolutionary trend of demand changes over time, while the correlation value indicates the strength of the causal relationship between user sentiment and demand changes.

[0157] In an embodiment of the present application, the cloud computing platform performs multi-dimensional data fusion on the divided time segments. For example, during the warm-up period of a promotional activity, the system extracts the characteristic value of the change frequency of the product search optimization demand during that period, and simultaneously analyzes the correlation value of the negative emotions in the user comments. Through a nonlinear superposition algorithm, the two are mapped into a combined metric value. When the characteristic value and the correlation value are both higher than the average level, the combined metric value increases exponentially; if only a single dimension is prominent, a weighted accumulation strategy is adopted. A dynamic smoothing mechanism is introduced in the superposition process to avoid metric value jumps due to instantaneous data fluctuations. For example, when an abnormally high-frequency demand change occurs in a certain period, the system automatically calls the historical data for the same period to perform mean correction to ensure the stability of the combined metric value.

[0158] 602. Determine a proportional relationship between the characteristic value of the time distribution characteristic and the correlation value based on the distribution interval of the combined metric value, and map the proportional relationship to the interaction intensity of the time segment;

[0159] In step 602, the interaction strength describes the transmission effect of demand changes and the resource competition relationship between adjacent time segments. It is dynamically determined by the distribution range ratio of the combined metric value. The distribution range refers to the proportion of the current segment's combined metric value to the total value of the global time period, reflecting its influence on the overall business evolution.

[0160] In an embodiment of the present application, the system establishes a horizontal comparative relationship between the combined metric values ​​​​between time segments. For example, the proportion of the combined metric value of the big promotion outbreak period to the total value of the whole period is significantly different from the distribution interval ratio of the warm-up period, and the interaction intensity between the two is marked as a strong correlation level. The system constructs a circular buffer to store the proportional relationship between the characteristic value and the proportional value of each segment, and uses a gradient descent algorithm to calculate the weight transfer coefficient between segments. For demands with resource competition, the system additionally introduces a conflict resolution factor to reduce the intensity calculation deviation caused by resource preemption. The final generated interaction intensity is used to identify potential conflicts and collaborative paths for the evolution of demand across time periods.

[0161] 603. Convert the interaction strength into a dynamic adjustment factor according to a preset ratio, and generate an initial weight value for each time segment in combination with the characteristic value of the corresponding time segment. Adjust the initial weight value proportionally according to the interval range of the initial weight value to obtain a demand priority weight value.

[0162] In step 603, the dynamic adjustment factor is a correction coefficient generated based on the strength of the interaction. It is used to balance the local weight of the feature value within a time segment with the global weight of the cross-segment impact. The initial weight value is the original priority score without considering cross-time correlation. The proportional adjustment refers to the compliance calibration of the initial value based on the business constraints.

[0163] In an embodiment of the present application, the priority engine converts the interaction intensity into an adjustment factor. For example, when the intensity of a certain time segment is identified as a core node, its adjustment factor will be amplified, so that the initial weight value of the demand change in the segment obtains additional gain. During the adjustment process, the system combines the resource pool capacity threshold to impose a secondary constraint on the initial weight value. If the adjusted demand priority weight value exceeds the current resource supply capacity, the degradation rule is triggered, and the weight is gradually reduced according to the importance of the demand type. For example, the demand for expansion of the payment system during the promotional outbreak period has an excessively high weight due to the gain of the adjustment factor, but is limited by the size of the server cluster. The system automatically adjusts its weight back to a supportable range, while reserving a buffer margin for associated demands.

[0164] 604. Based on the temporal arrangement order of the demand priority weight values, perform continuity correction on the weight value differences of adjacent time segments to obtain demand priority distribution characteristics.

[0165] In step 604, continuity correction is a technical measure to eliminate priority jumps between adjacent time segments, ensuring the coherence of the demand evolution path through a time series smoothing algorithm. Weight difference refers to the magnitude of the change in priority of the same demand between adjacent segments, reflecting the severity of the business focus shift.

[0166] In an embodiment of the present application, the timing analysis module performs convolution filtering on the demand priority weight value. For example, the weight value of the optimization demand of a certain product recommendation algorithm is low during the warm-up period. After entering the outbreak period, the weight increases sharply due to the decrease in user click-through rate, and the difference between the two exceeds the reasonable range. The system calculates the mean through a sliding window, performs Laplace smoothing on the mutation point, and generates a continuously increasing correction curve. At the same time, combined with the correction curve, a cross-segment weight association rule is established to continuously correct the difference in weight values ​​of adjacent time segments. The corrected demand priority distribution characteristics are stored in the form of a weight curve with a timestamp for downstream resource scheduling decisions.

[0167] Here's a specific example:

[0168] An e-commerce platform divides its annual promotion cycle into three core time periods: the warm-up period, the explosive growth period, and the final payment period. During the warm-up period, users reported slow loading of product detail pages. The system calculated a high combined metric value for this period, and marked the interaction strength with the explosive growth period as moderate. After dynamic adjustment factor enhancement, the initial weight of detail page optimization demand increased significantly. However, due to server resource limitations, the ratio was adjusted, and the final weight was set at the upper limit of the operational range. After entering the explosive growth period, the payment system faced a surge in pressure, and its combined metric value dominated. The system adjusted the payment demand weight to the highest globally, while retaining the residual weight of the warm-up period detail page demand through continuous correction to achieve a smooth transition. During the final payment period, the weight of logistics query demand increased rapidly due to a surge in user comments urging orders. The system automatically generated a priority evolution path connecting the explosive growth period and the final payment period, guiding the operations team to deploy cache resources in advance.

[0169] In summary, steps 601 to 604, through combined metric fusion, cross-period intensity mapping, dynamic weight calculation, and continuity correction, achieve precise quantification of e-commerce demand priorities and ensure their evolutionary consistency. The system automatically identifies time windows for high-value demand, balancing resource conflicts between short-term urgent needs and long-term strategic requirements, significantly improving the efficiency of version iteration and troubleshooting.

[0170] In order to further improve the collaborative efficiency of resource scheduling and software development demand evolution of e-commerce platforms, the solution determines the configuration parameters of the optical signal transmission channel by comparing the node weights and priority distribution characteristics, and converts its changes into priority weight tag values ​​in the time series data, and binds it to the data stream to generate tagged time series data. This method improves the utilization rate of network communication resources, enhances the system's responsiveness to high-priority tasks, and realizes the deep integration and collaborative management of computing, network, and task priorities. In some embodiments, the time series data is subjected to priority weight tagging processing according to the mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics in step 104 to generate tagged time series data, including:

[0171] 701. Determine configuration parameters of the optical signal transmission channel according to a matching relationship between node weights in the software development requirement evolution graph and weight values ​​in the requirement priority distribution feature;

[0172] In step 701, the optical signal transmission channel configuration parameters are a set of physical layer attributes that describe the bandwidth, latency, and fault tolerance of the data transmission path. These parameters are used to ensure the transmission quality of data streams corresponding to high-priority requirements. The node weight-to-requirement priority matching relationship refers to the mapping rules between the node importance levels in the demand evolution graph and the task urgency in the time series data.

[0173] In an embodiment of the present application, the system dynamically adjusts the configuration parameters of the optical signal transmission channel by analyzing the node attributes of the demand evolution map and combining the priority distribution characteristics of the current time segment. For example, when the software development demand evolution map shows that the weight of a certain commodity recommendation algorithm optimization demand node is significantly higher than that of other nodes, the system allocates a high-bandwidth, low-latency channel to the data stream corresponding to the demand and enables dual-link redundant transmission. An adaptive load balancing strategy is introduced during the configuration process. If it is detected that multiple high-weight nodes are competing for the same physical channel, the system automatically splits the wavelength resources according to the priority ratio to ensure that the core demand exclusively occupies the main channel and the secondary demand shares the backup channel. At the same time, additional fault-tolerant buffers are configured for key nodes based on historical fault records to avoid data flow interruptions due to hardware anomalies.

[0174] 702. According to the mapping relationship between the configuration parameters and the weight values, the parameter change of the configuration parameters is converted into the priority weight tag value of the corresponding time segment in the time series data, and the priority weight tag value is dynamically associated with the data flow of the corresponding time segment in the time series data to generate marked time series data.

[0175] In step 702, the priority weight tag value is a dynamic identifier embedded in the time series data, indicating the processing priority of the data flow in the current time segment. Dynamic data flow association involves binding the tag value with the underlying transmission channel, computing resources, and storage policy to form an end-to-end priority guarantee link.

[0176] In an embodiment of the present application, the data tagging engine captures the configuration parameter changes of the optical channel in real time and converts them into priority weight tag values ​​at the granularity of a time segment. For example, when the channel bandwidth required for capacity expansion of a payment system is increased to a peak level, the system generates the highest-level tag value for the payment transaction data in the corresponding time segment and automatically associates it with a dedicated computing node cluster. A two-layer verification mechanism is adopted in the tagging process. The main tagging layer records the real-time adjustment results, and the auxiliary tagging layer stores the historical configuration version. When a parameter out-of-bounds or resource conflict is detected, it automatically rolls back to the most recent stable version. At the same time, the priority weight tag value is bound to the data flow of the upstream and downstream service links through the data stream orchestrator to obtain the tag time series data, such as reserving cache space for high tag value data streams and triggering the abnormal retry mechanism first.

[0177] Here's a specific example:

[0178] During a major promotion on an e-commerce platform, the demand evolution graph showed that the node weight of the inventory synchronization service reached its highest global peak during the rush sale period. Based on this, the system allocated an ultra-high-bandwidth optical channel to this service and converted the channel parameter changes into the highest priority tag value during the rush sale period. All inventory query requests were embedded with this tag at the transport layer and routed to a cluster of compute nodes equipped with solid-state drive acceleration for processing. During peak hours that evening, the user behavior analysis service triggered channel parameter adjustments due to an increase in node weight. While ensuring that the inventory service tag value remained unchanged, the system assigned it a secondary channel and generated a medium-priority tag. Secondary priority data streams were automatically deferred while the compute nodes were queued, but could be preemptively transmitted during periods of channel inactivity. After the event ended, all tag values ​​returned to baseline levels, and resource allocation returned to a steady state.

[0179] In summary, steps 701 to 702 achieve precise matching of resource supply and demand evolution on the e-commerce platform through dynamic adaptation of optical channel parameters and embedding of data stream labels. High-priority demands are guaranteed deterministic transmission, data stream processing latency is significantly reduced, and the fault tolerance of key services is significantly enhanced. In actual applications, after a leading e-commerce platform adopted this method, the data transmission efficiency of core services during the promotion period was improved to an industry-leading level, and business interruptions caused by resource competition were reduced to zero, providing systematic technical support for service quality assurance in ultra-large-scale concurrent scenarios.

[0180] Figure 2 The present invention provides a schematic diagram of a software development system based on the Internet and cloud computing. Figure 2 As shown, the system includes:

[0181] The acquisition module 21 collects time series data of software development demand feedback in real time through the Internet and transmits the time series data to the cloud computing platform. The time series data includes demand change frequency parameters and user comment sentiment tendency parameters;

[0182] Extraction module 22 performs multi-dimensional feature analysis on the time series data through the cloud computing platform, extracts the time distribution characteristics of the requirement change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generates a software development requirement evolution map including requirement priority distribution characteristics and requirement change path topology;

[0183] An allocation module 23 dynamically allocates optical signal transmission channels adapted to the demand priority distribution characteristics based on the node weights in the software development demand evolution graph;

[0184] The marking module 24 performs priority weight marking processing on the time series data according to the mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics to generate marked time series data;

[0185] The generation module 25 dynamically binds the priority weights in the marked time series data with the node connection strength of the demand change path topology, and generates software development task scheduling instructions.

[0186] Figure 2 The software development system based on the Internet and cloud computing can execute Figure 1 The implementation principles and technical effects of the Internet- and cloud-computing-based software development method described in the illustrated embodiment are not further elaborated. The specific manner in which each module and unit performs operations in the Internet- and cloud-computing-based software development system described in the aforementioned embodiment has been described in detail in the related embodiments of the method and will not be further elaborated here.

[0187] In one possible design, Figure 2 The software development system based on the Internet and cloud computing of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0188] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0189] The processing component 32 is used for the above Figure 1 The embodiment provides a software development method based on the Internet and cloud computing.

[0190] The processing component 32 may include one or more processors to execute computer instructions to perform all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0191] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0192] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0193] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0194] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0195] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0196] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment is a software development method based on the Internet and cloud computing.

[0197] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0198] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0199] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0200] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A software development method based on the Internet and cloud computing, characterized in that: include: Collecting time series data of software development demand feedback in real time through the Internet and transmitting the time series data to a cloud computing platform, wherein the time series data includes demand change frequency parameters and user comment sentiment tendency parameters; Performing multi-dimensional feature analysis on the time series data through the cloud computing platform, extracting the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generating a software development demand evolution map including demand priority distribution characteristics and demand change path topology; Based on the node weights in the software development demand evolution graph, dynamically allocating optical signal transmission channels adapted to the demand priority distribution characteristics; performing priority weight marking processing on the time series data according to the mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics to generate marked time series data; According to the priority weight in the marked time series data and the node connection strength of the demand change path topology, the priority weight is dynamically bound to the allocation ratio of cloud computing resources to generate a software development task scheduling instruction.

2. The method according to claim 1, characterized in that The step of dynamically binding the priority weights in the marked time series data with the node connection strengths of the demand change path topology and the allocation ratio of cloud computing resources to generate software development task scheduling instructions includes: Establishing a mapping relationship table between the priority weights in the marked time series data and the types of cloud computing resources, wherein the types of cloud computing resources include low-latency computing resource types and high-throughput storage resource types; According to the node connection strength of the required change path topology, a corresponding type combination of cloud computing resources is matched from the mapping relationship table, and an initial allocation ratio is generated based on the initial proportion of the low-latency computing resource type and the high-throughput storage resource type in the type combination; Based on the distribution density of the priority weights, dynamically modifying the resource type ratio in the initial allocation ratio to generate a modified allocation ratio; The matching degree between the revised allocation ratio and the node connection strength is calculated, and the revised allocation ratio whose matching degree meets the preset conditions is associated with the software development task execution order corresponding to the priority weight and stored to generate a software development task scheduling instruction.

3. The method according to claim 2, characterized in that The step of establishing a mapping relationship table between priority weights in the marked time series data and types of cloud computing resources includes: Dividing the priority weights in the marked time series data into a high priority weight value interval and a low priority weight value interval according to a numerical range; Based on the task completion time parameter and the data storage capacity parameter extracted from the marked time series data, a first correspondence between the high priority weight value interval and the low latency computing resource type in the cloud computing resources is established, and a second correspondence between the low priority weight value interval and the high throughput storage resource type in the cloud computing resources is established; The first corresponding relationship and the second corresponding relationship are combined into a mapping relationship table in the order of the numerical ranges of the priority weights.

4. The method according to claim 2, characterized in that The dynamically revising the resource type proportion in the initial allocation ratio based on the distribution density of the priority weights to generate a revised allocation ratio includes: Calculating dynamic adjustment coefficients for low-latency computing resource types and high-throughput storage resource types based on a ratio of distribution densities of a high-priority weight numerical interval to a low-priority weight numerical interval in the priority weights; Proportionally adjusting the proportion of the low-latency computing resource type in the initial allocation ratio based on the dynamic adjustment coefficient to generate a first adjusted proportion, and reversely compensating the proportion of the high-throughput storage resource type in the initial allocation ratio based on the dynamic adjustment coefficient to obtain a second adjusted proportion; The first adjusted ratio and the second adjusted ratio are merged. When the merged result meets the preset resource allocation interval, the merged result is output as the revised allocation ratio. When the merged result exceeds the preset resource allocation interval, the merged result scaled proportionally according to the preset ratio is output as the revised allocation ratio.

5. The method according to claim 1, wherein The multi-dimensional feature analysis of the time series data is performed on the cloud computing platform to extract the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user comment sentiment tendency parameter, and generate a software development demand evolution map containing demand priority distribution characteristics and demand change path topology, including: Dividing the time series data into time segments through the cloud computing platform, and extracting the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user comment sentiment tendency parameter based on the time segments; Superimposing the time distribution feature and the correlation strength, and dynamically allocating a demand priority weight value according to a proportional relationship between the time distribution feature and the correlation strength in the superimposed result to generate a demand priority distribution feature; Based on the demand priority weight values ​​in the demand priority distribution characteristics, constructing connection relationships between adjacent time segments in the demand change path, and continuously integrating the connection relationships according to time sequence and intensity distribution to generate a demand change path topology; The demand priority distribution characteristics and the demand change path topology are jointly mapped through the cloud computing platform to generate a software development demand evolution map.

6. The method according to claim 5, characterized in that The step of superimposing the time distribution feature and the correlation strength, and dynamically allocating a demand priority weight value according to a proportional relationship between the time distribution feature and the correlation strength in the superimposed result to generate a demand priority distribution feature includes: Superimposing the characteristic value of each time segment in the time distribution feature with the correlation value of the corresponding time segment in the correlation strength to generate a combined metric value of the time segment; determining a proportional relationship between a characteristic value of the time distribution feature and the correlation value based on a distribution interval of the combined metric value, and mapping the proportional relationship into an interaction intensity of the time segment; Convert the interaction strength into a dynamic adjustment factor according to a preset ratio, and generate an initial weight value for each time segment in combination with the characteristic value of the corresponding time segment. Adjust the initial weight value proportionally according to the interval range of the initial weight value to obtain the demand priority weight value; Based on the time series arrangement order of the demand priority weight values, the difference in weight values ​​of adjacent time segments is corrected for continuity to obtain the demand priority distribution characteristics.

7. The method according to claim 1, characterized in that The step of performing priority weight marking processing on the time series data according to the mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics to generate marked time series data includes: determining configuration parameters of the optical signal transmission channel according to a matching relationship between node weights in the software development requirement evolution graph and weight values ​​in the requirement priority distribution feature; According to the mapping relationship between the configuration parameters and the weight values, the parameter change of the configuration parameters is converted into the priority weight tag value of the corresponding time segment in the time series data, and the priority weight tag value is dynamically associated with the data flow of the corresponding time segment in the time series data to generate marked time series data.

8. A software development system based on the Internet and cloud computing, characterized in that: include: The acquisition module collects time series data of software development demand feedback in real time through the Internet and transmits the time series data to the cloud computing platform. The time series data includes demand change frequency parameters and user comment sentiment tendency parameters; An extraction module performs multi-dimensional feature analysis on the time series data through the cloud computing platform, extracts the time distribution characteristics of the demand change frequency parameter and the correlation strength of the user review sentiment tendency parameter, and generates a software development demand evolution map including demand priority distribution characteristics and demand change path topology; an allocation module that dynamically allocates optical signal transmission channels adapted to the demand priority distribution characteristics based on the node weights in the software development demand evolution graph; a marking module, performing priority weight marking processing on the time series data according to a mapping relationship between the optical signal transmission channel and the demand priority distribution characteristics, and generating marked time series data; The generation module dynamically binds the priority weights in the marked time series data with the node connection strength of the demand change path topology according to the priority weights and generates software development task scheduling instructions.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a software development method based on the Internet and cloud computing as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the software development method based on the Internet and cloud computing as claimed in any one of claims 1 to 7 is implemented.

Images