Multi-task processing method and system in computing network integration environment

Abstract

The invention relates to the technical field of task processing, in particular to a multi-task processing method and system in a computing network integration environment, which comprises the following steps of: acquiring a CPU peak value and a memory retention time length in three periods of a task, calculating a resource concentration ratio, generating a load compact record, judging that a concentration ratio sequence is progressively increased and a task exceeding a threshold value is listed, and judging a schedulable window in combination with node peak interval fluctuation, calculating a resource adaptation ratio to generate matching mapping, and rejecting a high-delay link to complete task scheduling. According to the method, the CPU occupancy rate peak value and the memory duration time of the task are quantitatively analyzed in three continuous resource allocation periods, the high-access task of which the resource demand tends to increase progressively is accurately identified, and the adaptation degree between the task demand and the node resource is verified in real time in combination with node resource adaptation ratio value analysis; by constructing a round-trip delay variation ratio sequence and calculating a continuous period difference value, unstable links are eliminated in time, and more reliable and efficient task data transmission is realized.

Description

Technical Field

[0001] The present invention relates to the field of task processing technology, and in particular to a multi-task processing method and system in a computing-network fusion environment. Background Art

[0002] The field of task processing technology primarily involves the scheduling, allocation, execution, and resource coordination of multiple processing tasks within a computing system or network environment. Its core goal is to improve task execution efficiency, shorten response time, and increase system throughput and resource utilization. This area encompasses task modeling, task partitioning, priority assignment, scheduling strategies, task migration mechanisms, as well as fault tolerance and load balancing solutions. Typical applications include distributed systems, cloud computing platforms, high-performance computing systems, and edge computing environments. Task processing also needs to consider the dynamic and heterogeneous nature of the system, thereby developing appropriate scheduling and resource allocation strategies to achieve multi-task parallel processing and optimal resource allocation.

[0003] The multi-tasking method in a converged computing and network environment refers to a solution for efficiently allocating and executing multiple processing tasks within an architecture that deeply integrates computing and network resources. This method aims to coordinate computing nodes and network resources to achieve optimal scheduling and execution path selection for tasks within a distributed, multi-layered computing system, thereby improving the efficiency and reliability of overall business processing. This method is suitable for large-scale task processing scenarios requiring computing and network collaboration, such as multi-access edge computing, intelligent manufacturing, and smart cities.

[0004] Traditional processing methods only formulate scheduling and resource allocation strategies based on static task information or resource usage in isolated cycles, and fail to effectively identify the task resource demand trend and the node resource fluctuation period characteristics, resulting in the frequent disconnection between the task resource demand growth trend and the node resource allocation strategy, causing resource idleness or task overload. The matching of tasks and node resources lacks refined resource adaptation analysis, and tasks are assigned to nodes with inappropriate resources, exacerbating resource waste and low execution efficiency. The static detection and evaluation of link transmission delay lacks a periodic dynamic change monitoring mechanism, resulting in failure to detect link delay fluctuations exceeding expectations in a timely manner, seriously affecting the efficiency and reliability of task data transmission. For example, task data delays or transmission interruptions frequently occur in smart manufacturing or edge computing, reducing overall processing performance. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a multi-task processing method and system in a computing-network fusion environment.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a multi-task processing method in a computing-network fusion environment, comprising the following steps: S1: Obtain the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture over three consecutive resource allocation cycles, calculate the resource usage concentration of the tasks in each cycle, and generate a compact record set of cycle loads; S2: Based on the periodic load compact record set, the length of the monotonically increasing segment of the sequence is calculated according to the resource usage concentration sequence of the task, and compared with the set resource density threshold. If it is increased for three consecutive cycles and exceeds the threshold, the task is listed as having a high access tendency and a priority task list is obtained; S3: Based on the prioritized task list, extract the CPU peak period in each cycle, calculate the interval between two adjacent peaks of the node, calculate the degree of sequence fluctuation, determine whether the fluctuation is lower than the set stability threshold, and generate a node scheduling adaptation segment list; S4: Based on the node scheduling adaptation segment list and the task request parameters, the resource adaptation ratio is calculated. If the resource adaptation ratio is greater than the set resource adaptation threshold, it is marked as a successful match, and a multi-task matching node mapping table is obtained.

[0007] As a further solution of the present invention, the periodic load compact record set includes task periodic resource scheduling continuity parameters, unit period peak resource occupation factor and period average resource load balancing index; the priority task list specifically includes task access demand trend label, period priority identification status value and density trend stability flag item; the node scheduling adaptation segment list includes periodic fluctuation mean factor, periodic standard deviation stability mark and adjustable segment boundary index; the multi-task matching node mapping table specifically refers to the node resource remaining capacity index, task resource demand mapping relationship and matching task identification comparison number.

[0008] As a further solution of the present invention, the steps of obtaining the periodic load compact record set are specifically as follows: S101: Obtain the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture within three consecutive resource allocation cycles. Based on the task trigger time, end time, and resource usage segment identifier in the task scheduling log, extract the CPU usage sequence and memory occupancy sequence corresponding to the task within each cycle unit time period to establish a basic dataset of periodic resource usage corresponding to the task. S102: Based on the periodic resource usage basic data set, extract the maximum value of the CPU usage sequence of each task in a unit time period as the CPU occupancy peak value, call the length of the time period in which the memory occupancy sequence is continuously not less than a set threshold, calculate the resource usage interval concentration of the tasks in the period, and generate a resource concentration expression value list; S103: Based on the resource concentration expression value list, according to the resource interval concentration degree values ​​of the task in three consecutive cycles, a weighted average operation is performed on the numerical sequence of the same task in each cycle, and combined with the task existence identification and execution segment coverage in the cycle, the periodic load convergence intensity of the task is obtained, and a compact record set of the periodic load is established.

[0009] As a further solution of the present invention, the step of obtaining the priority task list is specifically as follows: S201: Based on the periodic load compact record set, extract the resource usage concentration value of each task in three consecutive scheduling cycles, arrange them into a period sequence in chronological order, perform a greater than relationship judgment on adjacent items in the period sequence, and count the number of consecutive segments that meet the relationship, obtain the length of the resource usage growth trend of the task in the corresponding period segment, and generate the resource concentration incremental segment length value; S202: Based on the incremental segment length value of the resource concentration, according to the growth segment length of each task, a set resource density threshold is obtained, the resource usage concentration sequence of the task in three cycles is called, a greater than relationship judgment is performed on each value, the number of items that meet the threshold condition is counted and compared with the growth segment length, the task identifiers that meet the three-cycle increment and are above the threshold are obtained, and a resource trend-compliant task number set is generated; S203: Call the task number set that meets the resource trend, filter the original task list according to the task identification information, filter out task items that do not meet the continuous growth and high resource density conditions, obtain a task set that meets the access tendency judgment requirements, and establish a priority task list.

[0010] As a further solution of the present invention, the step of obtaining the node scheduling adaptation segment list is specifically as follows: S301: Based on the prioritized task list, according to the assigned node corresponding to each task, the number of CPU instruction cycles completed and the memory usage percentage of the node in the last three scheduling cycles are collected, and the time period corresponding to the maximum value is extracted from the CPU usage sequence in each cycle and recorded as the periodic CPU peak period, thereby generating a periodic peak time index set; S302: Based on the period peak time index set, calculate the interval on the time axis according to two adjacent CPU peak time periods and sequentially construct a time interval sequence, call the node memory usage percentage sequence in the corresponding period on the adjacent interval sequence, and calculate to obtain the scheduling period stability scale value; S303: According to the scheduling cycle stability scale value, determine whether it is lower than the set stability threshold range, obtain the cycle index segment that meets the scheduling conditions, mark it as a schedulable window, extract the cycle time period range consistent with the task mapping relationship, and establish a node scheduling adaptation segment list.

[0011] As a further solution of the present invention, the formula for calculating and obtaining the scheduling period stability scale value is specifically: ; in, represents the scheduling period stability scale value, Representative The normalized value of the CPU peak time point, Representative The normalized value of the CPU peak time point, Representative The memory usage percentage of the node in the scheduling cycle, Represents the normalized time interval divided by the mean of the corresponding memory indicator. Represents the total number of scheduling cycles.

[0012] As a further solution of the present invention, the step of obtaining the multi-task matching node mapping table is specifically as follows: S401: Based on the node scheduling adaptation segment list, extract the executable state interval of each node in the task request cycle, collect the current remaining number of CPU cores and unallocated memory capacity of each node, record the node resource status information, and establish a node adjustable resource parameter set; S402: Based on the node adjustable resource parameter set, the requested number of CPU cores and the requested memory capacity of each task are obtained, and the ratios of these values ​​with the remaining number of CPU cores and the remaining memory capacity of the node are calculated. The node task response delay value and the duration of the occupancy period are integrated for joint adjustment. The resource adaptation strength value between the task and the node is calculated and compared with the set resource adaptation threshold. Task-node pairs that exceed the threshold are screened to obtain a resource adaptation matching result. The formula for calculating the resource adaptation strength value between the task and the node is as follows: ; in, Indicates the The resource adaptation strength value of each task, Indicates the The number of CPU cores requested by each task, Indicates the The number of CPU cores currently available for allocation to each node. Indicates the The memory capacity requested by each task, Indicates the The current remaining allocatable memory capacity of each node, Indicates the The number of cycles that a task occupies, Indicates the The normalized response delay value of a task under the assigned node; S403: calling the resource adaptation matching result, summarizing and arranging the numbers, parameters and matching status between tasks and nodes according to the task node pair identifier, eliminating matching combinations that do not meet the conditions, and establishing a multi-task matching node mapping table.

[0013] As a further embodiment of the present invention, the method further comprises the following steps: S5: Based on the multi-task matching node mapping table, a round-trip delay variation ratio sequence is constructed according to the transmission links of the nodes, and the difference is calculated with the corresponding position of the previous cycle sequence. If the difference is greater than the round-trip delay variation threshold, the path is eliminated and the task data transmission processing is carried out using the remaining links to obtain task scheduling execution information; The task scheduling execution information includes a link stability status mark, a path connectivity level classification item, and a node path screening index set.

[0014] As a further solution of the present invention, the step of obtaining the task scheduling execution information is specifically as follows: S501: Based on the multi-task matching node mapping table and the established task and node pairing information, extract the transmission link of each task-dependent node, collect the minimum round-trip delay and maximum round-trip delay of the link in the past three monitoring cycles, and construct a round-trip delay ratio sequence in each cycle to obtain a round-trip delay variation ratio sequence set; S502: Calling the round-trip delay variation ratio sequence set, selecting corresponding positions of the current cycle and the previous cycle, calculating the difference between the corresponding ratios of the two cycles, calling a set round-trip delay variation threshold, comparing the difference, marking links greater than the threshold as unstable paths, and generating link fluctuation status marking information; S503: Call the link fluctuation status mark information, remove the corresponding link-task pair from the multi-task matching node mapping table, extract the remaining matching task corresponding links as data transmission paths and complete task data forwarding, and establish task scheduling execution information.

[0015] A multi-task processing system in a computing-network fusion environment, wherein the multi-task processing system in the computing-network fusion environment is used to implement the multi-task processing method in the computing-network fusion environment. The system includes: The cycle load analysis module obtains the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture over three consecutive resource allocation cycles, calculates the resource usage concentration of tasks in each cycle, and generates a compact set of cycle load records; The task priority evaluation module calculates the length of the monotonically increasing segment of the sequence based on the periodic load compact record set and the resource usage concentration sequence of the task, and compares it with the set resource density threshold. If the sequence increases for three consecutive cycles and exceeds the threshold, the task is listed as having a high access tendency and a priority task list is obtained; The node scheduling analysis module extracts the CPU peak period in each cycle based on the prioritized task list, calculates the interval between two adjacent peaks of the node, calculates the degree of sequence fluctuation, determines whether the fluctuation is lower than the set stability threshold, and generates a node scheduling adaptation segment list; The task matching evaluation module calculates the resource adaptation ratio based on the node scheduling adaptation segment list and the task request parameters. If the resource adaptation ratio is greater than the set resource adaptation threshold, it is marked as a successful match and a multi-task matching node mapping table is obtained; The scheduling execution module constructs a round-trip delay variation ratio sequence based on the multi-task matching node mapping table and the transmission link of the node, and calculates the difference with the corresponding position of the previous cycle sequence. If the difference is greater than the round-trip delay variation threshold, the path is eliminated and the remaining links are used to implement task data transmission processing to obtain task scheduling execution information.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by quantitatively analyzing the CPU occupancy peak and memory duration of the task in three consecutive resource allocation cycles, the resource utilization concentration is obtained, and high-access tasks with increasing resource requirements are accurately identified. Tasks can be prioritized and scheduling accuracy can be improved. The sequence fluctuation degree of the node CPU peak period interval is calculated to determine the stable schedulable window of node resources, effectively avoiding the mismatch between node resource scheduling and task requirements. Combined with the node resource adaptation ratio analysis, the adaptation between task requirements and node resources is verified in real time to ensure efficient matching of tasks and node resources, avoiding resource waste and task delays. In terms of task transmission link selection, by constructing a round-trip delay change ratio sequence and calculating the difference between consecutive cycles, unstable links are promptly eliminated, achieving more reliable and efficient task data transmission, and improving the stability, accuracy and resource utilization efficiency of task processing in the overall computing-network integration environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1It is a schematic diagram of the workflow of the present invention; Figure 2 This is a detailed flow chart of S1 of the present invention; Figure 3 This is a detailed flow chart of S2 of the present invention; Figure 4 This is a detailed flow chart of S3 of the present invention; Figure 5 This is a detailed flow chart of S4 of the present invention; Figure 6 This is a detailed flow chart of S5 of the present invention; Figure 7 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0020] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0021] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0022] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0023] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0024] See also Figure 1 The present invention provides a technical solution: a multi-task processing method in a computing-network fusion environment, comprising the following steps: S1: Obtain the CPU usage, memory usage, and task scheduling logs of registered tasks in the computing-network convergence architecture over three consecutive resource allocation cycles. Based on the peak CPU usage and memory retention time of each task within a unit time period, calculate the resource usage concentration of the task in each cycle and generate a compact record set of periodic loads. S2: Based on the periodic load compact record set, the length of the monotonically increasing segment of the sequence is calculated according to the resource usage concentration sequence of the task in three scheduling cycles, and compared with the set resource density threshold. If the sequence is increasing for three consecutive cycles and exceeds the threshold, the task is listed as having a high access tendency and a priority task list is obtained; S3: Based on the prioritized task list, according to the task assignment node, the number of CPU instruction cycles completed and the memory usage percentage of the node in the last three cycles are called, the CPU peak period in each cycle is extracted, the interval between two adjacent peaks of the node is calculated, a peak interval sequence is constructed, and the degree of sequence fluctuation is calculated to determine whether the fluctuation is lower than a set stability threshold. If so, the cycle segment is marked as a schedulable window, and a node scheduling adaptation segment list is generated; S4: Based on the node scheduling adaptation segment list and the node's executable state interval, extract the remaining schedulable CPU cores and allocatable memory capacity within the task request period, and calculate the resource adaptation ratio according to the task request parameters. If the resource adaptation ratio is greater than a set resource adaptation threshold, it is marked as a successful match, and a multi-task matching node mapping table is obtained; S5: Based on the multi-task matching node mapping table, extract the minimum round-trip delay and the maximum round-trip delay of the link in the last three monitoring cycles according to the node's transmission link, construct a round-trip delay change ratio sequence, and calculate the difference with the corresponding position of the previous cycle sequence. If the difference is greater than the round-trip delay change threshold, mark the link as unstable, eliminate the path, and use the remaining links to implement task data transmission processing to obtain task scheduling execution information; The periodic load compact record set includes task cycle resource scheduling continuity parameters, unit cycle peak resource occupation factor and cycle average resource load balancing index; the priority task list specifically includes task access demand trend label, period priority identification status value and density trend stability flag item; the node scheduling adaptation segment list includes period fluctuation mean factor, period standard deviation stability mark and adjustable segment boundary index; the multi-task matching node mapping table specifically refers to the node resource remaining capacity index, task resource demand mapping relationship and matching task identification comparison number; the task scheduling execution information includes link stability status mark, path connectivity level classification item and node path screening index set.

[0025] See also Figure 2 The specific steps for obtaining the periodic load compact record set are: S101: Obtain the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture within three consecutive resource allocation cycles. Based on the task trigger time, end time, and resource usage segment identifier in the task scheduling log, extract the CPU usage sequence and memory occupancy sequence corresponding to the task within each cycle unit time period to establish a basic dataset of periodic resource usage corresponding to the task. Obtain the CPU usage, memory occupancy and task scheduling logs of registered tasks in the computing-network fusion architecture within three consecutive resource allocation cycles. First, perform field extraction on the task scheduling log, parse the trigger time and end time of each task item by item, and extract the scheduling start and end time of the task in each resource allocation cycle. Match the task number field and the cycle time axis index in the log, map the task event to the time period within a fixed cycle, and then perform resource data extraction on the time period within each mapping cycle. Extract the CPU usage data within the unit time period (such as 1 second) from the structured CPU usage record to form a time series, and perform time alignment with the memory usage record in the same cycle to extract the memory occupancy sequence value at the corresponding time point. If there is a task segment execution identifier in the log, it is necessary to calculate the time series based on the scheduling time. The CPU and memory data are partitioned separately, and the data of the scheduling interruption period is eliminated. Only the resource usage records within the continuous scheduling time are retained. The CPU and memory usage are further aggregated by task number to ensure that each task forms an independent resource usage sequence in different cycles. In the example, if a task taskA runs from 0 seconds to 5 seconds in the first cycle, the CPU usage is 42%, 51%, 47%, 53%, 49%, and 50%, and the memory usage is 62%, 63%, 63%, 64%, 65%, and 64%, then the resource sequence recorded in its first cycle is the above data. In this way, batch processing is performed on all tasks and the task ID, cycle number, CPU usage sequence, and memory usage sequence are structured and combined to finally establish the basic dataset of cycle resource usage corresponding to the task. S102: Based on the periodic resource usage basic data set, extract the maximum value of the CPU usage sequence of each task in a unit time period as the CPU occupancy peak value, call the length of the time period in which the memory occupancy sequence is continuously not less than a set threshold, calculate the resource usage interval concentration of the tasks in the period, and generate a resource concentration expression value list; Based on the basic data set of periodic resource usage, we first perform a maximum value recognition operation on the CPU usage sequence of each task in each cycle, select the peak CPU usage from the time series and record the corresponding value and time point, and traverse the memory occupancy sequence in the cycle. The memory value on each time slice is compared with the preset memory threshold. If the continuous values ​​are all greater than the memory threshold, they are counted into the resource maintenance segment, and the continuous duration is counted as the length of the memory occupancy duration. The memory threshold is set to , which is set based on the average memory usage of all tasks in the current system scheduling cycle Plus the maximum tolerance fluctuation value of system scheduling ,Right now: ,in, is the average memory usage of all tasks in the period, For a fixed system dispatch safety margin (e.g. 5%), for example, when , ,but If the memory sequence of a task within 5 seconds is 62%, 63%, 63%, 64%, 65%, and 64%, then the length of the continuous segment is 6 seconds. Based on this, two parameters are constructed for each task in each cycle: CPU peak value and memory usage duration. These parameters are used as resource concentration description indicators for the task in that cycle. They are further combined and output according to task dimensions to form a task resource expression structure. In this example, if taskA has a CPU peak value of 53% and a memory maintenance period of 6 seconds in cycle 1, its resource concentration description parameters are (53, 6). Each task is processed sequentially to generate the resource expression sequence of all tasks within three cycles, and finally a list of resource concentration expression values ​​is generated. S103: Based on the resource concentration expression value list, according to the resource interval concentration degree values ​​of the task in three consecutive cycles, a weighted average operation is performed on the numerical sequence of the same task in each cycle, and combined with the task existence identification and execution segment coverage in the cycle, the periodic load convergence intensity of the task is obtained, and a compact record set of the periodic load is established.

[0026] According to the resource interval concentration value of each task in the resource concentration expression value list in three consecutive cycles, the resource expression value of each task in the three cycles is first weighted and integrated, and the CPU peak value and memory duration are weighted averaged respectively. The weights used in the weighted operation are recorded as , by the task in The execution continuity of a cycle is determined if the effective running time of a task in the cycle is not less than the average scheduling time of the task. 80%, then set , otherwise set to This rule is designed to distinguish the weight difference between complete scheduling and interrupted tasks. In the example, if task A has a continuous running time greater than 1 in cycles 1 and 2, , and cycle 3 is interrupted, then its CPU peak weighted average is: ; in, For the CPU peak of the period, The weight value of the corresponding cycle is used, and a similar weighted operation is performed on its memory maintenance length. At the same time, it is determined whether the coverage ratio of the task in the cycle execution segment exceeds 50%. If it is less than this coverage ratio, its cycle weight is automatically set to 0 to prevent the abnormal resource usage caused by the incomplete task from affecting the overall judgment. Finally, the weighted average result is structured to form a cycle resource concentration description structure in the task dimension. This structure will serve as one of the basic reflection indicators of resource behavior stability. The task number and the weighted CPU concentration value and memory concentration value are output, and a compact record set of cycle load is finally established.

[0027] See also Figure 3 The steps for obtaining the priority task list are as follows: S201: Based on the periodic load compact record set, extract the resource usage concentration value of each task in three consecutive scheduling cycles, arrange them into a period sequence in chronological order, perform a greater than relationship judgment on adjacent items in the period sequence, and count the number of consecutive segments that meet the relationship, obtain the length of the resource usage growth trend of the task in the corresponding period segment, and generate the resource concentration incremental segment length value; Based on the compact record set of periodic load, the resource usage concentration value of each task in three consecutive scheduling cycles is extracted. First, the scheduling cycle number corresponding to the task is parsed, and the resource concentration value in cycles 1, 2, and 3 is extracted from the record set with the task number as the primary key. The extracted values ​​are constructed into a one-dimensional ordered sequence in chronological order, and then the adjacent two items in the sequence are judged to be greater than the relationship, that is, the first two items are judged to be greater than the next two items. Is the cycle value greater than If the judgment is true, it is counted as an increment. If not, the current increment segment is interrupted and re-counted. In this process, the variable Record the length of consecutive true segments. In the example, if the resource concentration sequence of a task is 42.3, 48.9, and 52.1, then it is true in two adjacent comparisons, and finally we get If the sequence is 45.2, 44.6, 50.3, then interrupt once and only keep the single segment increment record, and get ,Finally, a key-value pair structure is constructed and outputted based on the task number and its incremental length, and the incremental segment length value of the resource concentration is obtained; S202: Based on the incremental segment length value of the resource concentration, according to the growth segment length of each task, a set resource density threshold is obtained, the resource usage concentration sequence of the task in three cycles is called, a greater than relationship judgment is performed on each value, the number of items that meet the threshold condition is counted and compared with the growth segment length, the task identifiers that meet the three-cycle increment and are above the threshold are obtained, and a resource trend-compliant task number set is generated; According to the length of the increasing segment of each task in the resource concentration increment length value, after obtaining the resource density threshold, the resource concentration values ​​of the tasks in the three cycles are compared and judged item by item. The resource density threshold is set to The threshold is set based on the average resource consumption of the task. Concentration deviation coefficient allowed by the system , using the formula ,in is the average resource concentration of all tasks in the cycle, is a fixed setting value, which is generally set to 6. , , then , to determine whether each task exceeds the threshold in three cycles, and to execute more than If three conditions are met consecutively and the corresponding value of the incrementing segment length value of the task in the resource set is equal to 2, it means that the task meets the two conditions of "continuous growth and high density". The task number that meets the conditions is recorded. In the example, if the resource values ​​of taskX are 52.3, 54.6, and 56.2, and the incrementing segment length is 2, it is recorded as a qualified task. Finally, a numbering structure is formed, and the resource trend meets the task number set; S203: Call the task number set that meets the resource trend, filter the original task list according to the task identification information, filter out task items that do not meet the continuous growth and high resource density conditions, obtain a task set that meets the access tendency judgment requirements, and establish a priority task list.

[0028] The resource trend matches the task identification information in the task number set, and the task number filtering operation is performed on the original task list. Only the task entries with matching task numbers are retained. The task number field in the original task table is compared with the number set. If there is a match, it is retained. If not, it is eliminated. The task attribute field information is further extracted based on the retained items. The task ID, scheduling cycle index, CPU and memory resource requirement value fields are structured and constructed, and organized into a schedulable task structure example. Among them, if taskZ exists in the number set and its fields are CPU core request 4 cores, memory request 8GB, and scheduling cycle 1-3, it is directly included in the structure output result, completing the task access judgment process, and finally establishing a priority task list.

[0029] See also Figure 4 The steps for obtaining the node scheduling adaptation segment list are as follows: S301: Based on the prioritized task list, according to the assigned node corresponding to each task, the number of CPU instruction cycles completed and the memory usage percentage of the node in the last three scheduling cycles are collected, and the time period corresponding to the maximum value is extracted from the CPU usage sequence in each cycle and recorded as the periodic CPU peak period, thereby generating a periodic peak time index set; Get the assigned node corresponding to each task in the priority task list, extract the mapping information between tasks and nodes one by one, locate the task number and node identification field in the task scheduling relationship table, retrieve the resource statistics records of each node in the last three scheduling cycles, extract the number of CPU instruction cycles completed under the cycle dimension and the memory usage percentage sequence of the corresponding cycle, aggregate the CPU execution records into a continuous utilization sequence by second or millisecond time slices, call the CPU utilization sequence in each cycle, extract the peak utilization point and record its corresponding absolute timestamp to form three time values, and then organize them into a set of time series values. If the maximum CPU utilization recorded by a node in three cycles occurs at the 10th second, 24th second and 38th second respectively, then its timestamp is , , , this group of time points constitutes the period peak time index set; S302: Based on the period peak time index set, calculate the interval on the time axis according to two adjacent CPU peak time periods and sequentially construct a time interval sequence, call the node memory usage percentage sequence in the corresponding period on the adjacent interval sequence, and calculate to obtain the scheduling period stability scale value; The formula for calculating the scheduling period stability scale value is specifically: ; in, represents the scheduling period stability scale value, Representative The normalized value of the CPU peak time point, Representative The normalized value of the CPU peak time point, Representative The memory usage percentage of the node in the scheduling cycle, Represents the normalized time interval divided by the mean of the corresponding memory indicator. Represents the total number of scheduling cycles; The scheduling cycle stability scale value is used to quantify the degree of fluctuation in the scheduling load of the computing network fusion node over multiple scheduling cycles. It is essentially a variable standard deviation indicator, reflecting the stability of the CPU load peak time point and the smoothness of resource scheduling linked to the node memory usage. The smaller this value is, the smaller the time interval variation of the CPU peak within the scheduling cycle, and the more stable the system scheduling. Conversely, it indicates that the system has sudden load peaks and the scheduling interval is unstable. This formula is essentially the square root of the variance of a weighted normalized time interval series, conforming to the standard statistical modeling model for stability indicators. The "+1" is added to avoid division-by-zero anomalies caused by zero memory and to enhance the amplification effect in low-memory, high-volatility scenarios.

[0030] Based on the three CPU peak timestamps in the cycle peak time index, each time value is first normalized and the time value is transformed using the maximum and minimum normalization method. The specific processing method is: assuming that the peak time of the node in the three cycles is the 10th second, the 24th second, and the 38th second respectively, then 、 、 , the normalized formula is: ; in, 、 , then: ; ; ; Next, we construct the time difference sequence and obtain: ; ; Then get the memory usage percentage of the node monitored in these three cycles and set it as: 、 , expressed as a proportional value 、 , calculate the intermediate result of the sum of each time difference and memory ratio: ; ; Then calculate the average : ; Substituting into the formula: ; Finally, the scheduling cycle stability scale value of the node in the current three cycles is This value is used to determine whether the stability threshold conditions set in subsequent steps are met. The entire process completes the complete processing flow from raw time data and memory usage data to normalization, ratio conversion, mean extraction, and standard deviation calculation, forming the basis for determining scheduling fluctuations.

[0031] S303: According to the scheduling cycle stability scale value, determine whether it is lower than the set stability threshold range, obtain the cycle index segment that meets the scheduling conditions, mark it as a schedulable window, extract the cycle time period range consistent with the task mapping relationship, and establish a node scheduling adaptation segment list.

[0032] Make interval judgment based on the scheduling cycle stability scale value and call the set stability threshold The threshold setting is determined jointly based on the node hardware attributes and the task type. The specific setting logic is: when the CPU type of the task carried by the node is a general task, the system's tolerance for fluctuation stability is set to 0.15; if the task type is a graph computing task, it has stricter requirements on the scheduling rhythm, and the fluctuation threshold is set to 0.004. In the example, when the node , its task is graph computing, so Does not meet the stability requirements, so it is not marked as a schedulable window. If the node , then it satisfies The judgment criteria mark the period as a schedulable window, and further combine the task scheduling information to map the corresponding time period to the scheduling period index assigned to the task, complete the extraction of the spatiotemporal correspondence between the task nodes, and finally establish a list of node scheduling adaptation segments.

[0033] See also Figure 5 The steps for obtaining the multi-task matching node mapping table are as follows: S401: Based on the node scheduling adaptation segment list, extract the executable state interval of each node in the task request cycle, collect the current remaining number of CPU cores and unallocated memory capacity of each node, record the node resource status information, and establish a node adjustable resource parameter set; Based on the node scheduling adaptation segment list, read the scheduling cycle range and node number corresponding to each task in turn, extract the node corresponding to the task from the scheduling node mapping table, and retrieve the status record within the task request cycle based on the node number, locate the corresponding cycle index field, and extract the executable identification field in the node cycle status data. The interval marked as "1" is used as the valid adjustable segment of the node. On this basis, further extract the remaining resource status of the node within the time period. The specific fields are the current number of available CPU cores and unallocated memory capacity, and extract the field values ​​respectively. 、 In this example, if the number of remaining CPU cores of a node N1 is 6 and the remaining memory capacity is 18GB during the task request cycle, then the parameter information is associated with the task mapping table to construct the resource supply set that can be adjusted for the task in this cycle, and the task number, node number, 、 The four contents are structured and combined into items, and all task node combinations are processed in sequence to finally establish a node adjustable resource parameter set; S402: Based on the node adjustable resource parameter set, the requested number of CPU cores and the requested memory capacity of each task are obtained, and the ratios of these values ​​with the remaining number of CPU cores and the remaining memory capacity of the node are calculated. The node task response delay value and the duration of the occupancy period are integrated for joint adjustment. The resource adaptation strength value between the task and the node is calculated and compared with the set resource adaptation threshold. Task-node pairs that exceed the threshold are screened to obtain a resource adaptation matching result. The formula for calculating the resource adaptation strength value between the task and the node is as follows: ; in, Indicates the The resource adaptation strength value of each task, Indicates the The number of CPU cores requested by each task, Indicates the The number of CPU cores currently available for allocation to each node. Indicates the The memory capacity requested by each task, Indicates the The current remaining allocatable memory capacity of each node, Indicates the The number of cycles that a task occupies, Indicates the The normalized response delay value of a task under the assigned node; The resource adaptation strength value is used to quantify the degree of fit between each task and the target node resources, measuring whether the task is suitable for scheduling and execution. It comprehensively considers the matching degree of CPU core count, the compactness of memory capacity adaptation, and the coupling effect between task lifecycle and node response delay. The lower this value, the higher the adaptability of the task and node resources, and the more suitable it is for the current scheduling match; when it is above the threshold, it means that the resource matching burden is too heavy or the scheduling delay is unreasonable. Item 1 : Measures the CPU core matching strength and directly reflects whether it can be loaded; Item 2 : represents the coupling error between memory usage efficiency and task lifecycle, and amplifies the impact of mismatch by weighting node response delay. This formula integrates resource load ratio, lifecycle compactness, and execution delay sensitivity, and can more realistically reflect the adaptability of task scheduling in heterogeneous resource environments. It has the advantage of balancing engineering practicality and algorithm accuracy. According to the resource comparison relationship between tasks and nodes in the node adjustable resource parameter set, the resource requirement parameters of each task are called to read the CPU core request number and memory capacity request value respectively. 、 , and extract the number of cycles it occupies And the normalized value of the task's response delay on the node , substitute these parameters into the resource adaptation strength formula: ; In this example, if task T5 requests a 4-core CPU and 8GB of memory, takes 3 cycles, node N1 currently has 6-core CPUs and 18GB of memory, and the normalized response latency is 0.36, the calculation is as follows: ; ; ; The median difference is , the final calculation result is: ; System-set resource adaptation threshold , the formula shows that the adaptation value of task T5 is 0.7834, which is lower than the set threshold. It is determined that the task and the node are matched effectively. The task number, node number and Record them together and finally output the resource adaptation and matching results; S403: Calling the resource adaptation matching result, summarizing and arranging the numbers, parameters and matching status between tasks and nodes according to the task node pair identifier, eliminating matching combinations that do not meet the conditions, and establishing a multi-task matching node mapping table; Call all task and node combination identifiers in the resource adaptation matching results, perform double primary key matching operations on task numbers and node numbers, and filter and eliminate all resource adaptation values Exceeding the set threshold The combination of tasks that meet the pairing conditions is retained, and the node number and resource configuration parameters are mapped and aggregated according to the task number. The matching node number, the number of CPU cores used, the memory capacity value, and the resource adaptation strength value field corresponding to each task are recorded and summarized into a task mapping information structure. In the example, if task T5 matches node N1, the entry T5→N1{CPU:4,MEM:8, :0.7834}, and thus complete the pairing and sorting of all tasks, and finally establish a multi-task matching node mapping table.

[0034] See also Figure 6 The steps for obtaining the task scheduling execution information are as follows: S501: Based on the multi-task matching node mapping table and the established task and node pairing information, extract the transmission link of each task-dependent node, collect the minimum round-trip delay and maximum round-trip delay of the link in the past three monitoring cycles, and construct a round-trip delay ratio sequence in each cycle to obtain a round-trip delay variation ratio sequence set; Based on the established task and node pairing information in the multi-task matching node mapping table, the transmission link number of each task-dependent node is extracted, the communication path identifier between each task node is read according to the node topology table, and the link performance data recorded by the network performance monitoring module is called to obtain the minimum round-trip delay of the target link in the past three scheduling monitoring cycles. and maximum round trip delay , where the period index , perform ratio calculation operation on each cycle to build the round-trip delay ratio In this example, if the minimum and maximum round-trip delays of a link in three cycles are: cycle 1: 5ms and 9ms, cycle 2: 6ms and 12ms, cycle 3: 7ms and 14ms, then the ratio sequence is 、 、 , store the ratio sequence of the link into a structured data table, perform the same processing on all links, and finally construct the round-trip delay variation ratio sequence set corresponding to the task link; S502: Calling the round-trip delay variation ratio sequence set, selecting corresponding positions of the current cycle and the previous cycle, calculating the difference between the corresponding ratios of the two cycles, calling a set round-trip delay variation threshold, comparing the difference, marking links greater than the threshold as unstable paths, and generating link fluctuation status marking information; Call the ratio sequence of each link in the round-trip delay variation ratio sequence set, compare the ratio item of the current monitoring cycle (set as the third cycle) with the previous cycle (the second cycle), that is, perform difference calculation on each link , and set a threshold for the difference value range For comparison and judgment, the threshold is set according to the link type. If the link is a trunk channel between nodes, the tolerance is set to 0.3. If it is an edge node or an inter-domain path, the threshold is set to 0.15. In the example, if the ratio of link number L7 in two cycles is 2.0 and 1.6 respectively, the difference is 0.4, and its type is a trunk channel, then it satisfies ,The link is marked as unstable, and a field structure is established as a link number and stability flag pair. All links greater than the threshold are marked as "unstable", and the rest are marked as "stable". The complete status record is output to generate the link fluctuation status mark value; S503: calling the link fluctuation state mark information, removing the corresponding link-task pair from the multi-task matching node mapping table, extracting the remaining matching task corresponding links as data transmission paths and completing task data forwarding, and establishing task scheduling execution information; Based on the link numbers marked as "stable" in the link fluctuation status tag value, a filtering operation is performed on all task pairs in the multi-task matching node mapping table. The mapping table is screened one-to-one based on the task corresponding link identifier, and all combinations where the links on which the tasks depend are marked as "unstable" are eliminated. Only the task node mapping items corresponding to the stable path are retained. The link numbers in the retained combinations are then used as the data transmission paths in the task scheduling phase. During the execution transmission phase, the corresponding task number, link number, path communication average delay, and task transmission data volume fields are recorded. In this example, if the matching path for task T3 is L5, L5 is determined to be a stable path, the transmission data volume is 12MB, and the average round-trip delay is 8.5ms, the task scheduling execution information is recorded as T3→L5{12MB,8.5ms}. Finally, the mapping information collation in the task scheduling phase is completed, and the task scheduling execution information is established.

[0035] See also Figure 7 A multi-task processing system in a computing-network fusion environment is provided. The multi-task processing system in a computing-network fusion environment is used to execute the multi-task processing method in the computing-network fusion environment. The system includes: The cycle load analysis module obtains the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture over three consecutive resource allocation cycles, calculates the resource usage concentration of tasks in each cycle, and generates a compact set of cycle load records; The task priority evaluation module calculates the length of the monotonically increasing segment of the sequence based on the periodic load compact record set and the resource usage concentration sequence of the task, and compares it with the set resource density threshold. If the sequence increases for three consecutive cycles and exceeds the threshold, the task is listed as having a high access tendency and a priority task list is obtained; The node scheduling analysis module extracts the CPU peak period in each cycle based on the prioritized task list, calculates the interval between two adjacent peaks of the node, calculates the degree of sequence fluctuation, determines whether the fluctuation is lower than the set stability threshold, and generates a node scheduling adaptation segment list; The task matching evaluation module calculates the resource adaptation ratio based on the node scheduling adaptation segment list and the task request parameters. If the resource adaptation ratio is greater than the set resource adaptation threshold, it is marked as a successful match and a multi-task matching node mapping table is obtained; The scheduling execution module constructs a round-trip delay variation ratio sequence based on the multi-task matching node mapping table and the transmission link of the node, and calculates the difference with the corresponding position of the previous cycle sequence. If the difference is greater than the round-trip delay variation threshold, the path is eliminated and the remaining links are used to implement task data transmission processing to obtain task scheduling execution information.

[0036] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0037] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0038] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0039] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0040] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0041] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0042] 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, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0043] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0044] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0045] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A multi-task processing method in a computing-network fusion environment, characterized in that: The following steps are involved: S1: Obtain the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture over three consecutive resource allocation cycles, calculate the resource usage concentration of the tasks in each cycle, and generate a compact record set of cycle loads; S2: Based on the periodic load compact record set, the length of the monotonically increasing segment of the sequence is calculated according to the resource usage concentration sequence of the task, and compared with the set resource density threshold. If it is increased for three consecutive cycles and exceeds the threshold, the task is listed as having a high access tendency and a priority task list is obtained; S3: Based on the prioritized task list, extract the CPU peak period in each cycle, calculate the interval between two adjacent peaks of the node, calculate the degree of sequence fluctuation, determine whether the fluctuation is lower than the set stability threshold, and generate a node scheduling adaptation segment list; S4: Based on the node scheduling adaptation segment list and the task request parameters, the resource adaptation ratio is calculated. If the resource adaptation ratio is greater than the set resource adaptation threshold, it is marked as a successful match, and a multi-task matching node mapping table is obtained.

2. The multi-task processing method in a computing-network fusion environment according to claim 1, characterized in that: The periodic load compact record set includes task period resource scheduling continuity parameters, unit period peak resource occupancy factor and period average resource load balancing index; the priority task list specifically includes task access demand trend label, period priority identification status value and density trend stability flag item; the node scheduling adaptation segment list includes period fluctuation mean factor, period standard deviation stability mark and adjustable segment boundary index; the multi-task matching node mapping table specifically refers to the node resource remaining capacity index, task resource demand mapping relationship and matching task identification comparison number.

3. The multi-task processing method in a computing-network fusion environment according to claim 2, characterized in that: The steps for obtaining the periodic load compact record set are specifically as follows: S101: Obtain the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture within three consecutive resource allocation cycles. Based on the task trigger time, end time, and resource usage segment identifier in the task scheduling log, extract the CPU usage sequence and memory occupancy sequence corresponding to the task within each cycle unit time period to establish a basic dataset of periodic resource usage corresponding to the task. S102: Based on the periodic resource usage basic data set, extract the maximum value of the CPU usage sequence of each task in a unit time period as the CPU occupancy peak value, call the length of the time period in which the memory occupancy sequence is continuously not less than a set threshold, calculate the resource usage interval concentration of the tasks in the period, and generate a resource concentration expression value list; S103: Based on the resource concentration expression value list, according to the resource interval concentration degree values ​​of the task in three consecutive cycles, a weighted average operation is performed on the numerical sequence of the same task in each cycle, and combined with the task existence identification and execution segment coverage in the cycle, the periodic load convergence intensity of the task is obtained, and a compact record set of the periodic load is established.

4. The multi-task processing method in a computing-network fusion environment according to claim 3 is characterized in that: The steps for obtaining the priority task list are as follows: S201: Based on the periodic load compact record set, extract the resource usage concentration value of each task in three consecutive scheduling cycles, arrange them into a period sequence in chronological order, perform a greater than relationship judgment on adjacent items in the period sequence, and count the number of consecutive segments that meet the relationship, obtain the length of the resource usage growth trend of the task in the corresponding period segment, and generate the resource concentration incremental segment length value; S202: Based on the incremental segment length value of the resource concentration, according to the growth segment length of each task, a set resource density threshold is obtained, the resource usage concentration sequence of the task in three cycles is called, a greater than relationship judgment is performed on each value, the number of items that meet the threshold condition is counted and compared with the growth segment length, the task identifiers that meet the three-cycle increment and are above the threshold are obtained, and a resource trend-compliant task number set is generated; S203: Call the task number set that meets the resource trend, filter the original task list according to the task identification information, filter out task items that do not meet the continuous growth and high resource density conditions, obtain a task set that meets the access tendency judgment requirements, and establish a priority task list.

5. The multi-task processing method in a computing-network fusion environment according to claim 4, characterized in that: The steps for obtaining the node scheduling adaptation segment list are specifically as follows: S301: Based on the prioritized task list, according to the assigned node corresponding to each task, the number of CPU instruction cycles completed and the memory usage percentage of the node in the last three scheduling cycles are collected, and the time period corresponding to the maximum value is extracted from the CPU usage sequence in each cycle and recorded as the periodic CPU peak period, thereby generating a periodic peak time index set; S302: Based on the period peak time index set, calculate the interval on the time axis according to two adjacent CPU peak time periods and sequentially construct a time interval sequence, call the node memory usage percentage sequence in the corresponding period on the adjacent interval sequence, and calculate to obtain the scheduling period stability scale value; S303: According to the scheduling cycle stability scale value, determine whether it is lower than the set stability threshold range, obtain the cycle index segment that meets the scheduling conditions, mark it as a schedulable window, extract the cycle time period range consistent with the task mapping relationship, and establish a node scheduling adaptation segment list.

6. The multi-task processing method in a computing-network fusion environment according to claim 5, characterized in that: The formula for calculating the scheduling period stability scale value is specifically: ； in, represents the scheduling period stability scale value, Representative The normalized value of the CPU peak time point, Representative The normalized value of the CPU peak time point, Representative The memory usage percentage of the node in the scheduling cycle, Represents the normalized time interval divided by the mean of the corresponding memory indicator. Represents the total number of scheduling cycles.

7. The multi-task processing method in a computing-network fusion environment according to claim 6, characterized in that: The steps for obtaining the multi-task matching node mapping table are specifically as follows: S401: Based on the node scheduling adaptation segment list, extract the executable state interval of each node in the task request cycle, collect the current remaining number of CPU cores and unallocated memory capacity of each node, record the node resource status information, and establish a node adjustable resource parameter set; S402: Based on the node adjustable resource parameter set, the requested number of CPU cores and the requested memory capacity of each task are obtained, and the ratios of these values ​​with the remaining number of CPU cores and the remaining memory capacity of the node are calculated. The node task response delay value and the duration of the occupancy period are integrated for joint adjustment. The resource adaptation strength value between the task and the node is calculated and compared with the set resource adaptation threshold. Task-node pairs that exceed the threshold are screened to obtain a resource adaptation matching result. The formula for calculating the resource adaptation strength value between the task and the node is as follows: ； in, Indicates the The resource adaptation strength value of each task, Indicates the The number of CPU cores requested by each task, Indicates the The number of CPU cores currently available for allocation to each node. Indicates the The memory capacity requested by each task, Indicates the The current remaining allocatable memory capacity of each node, Indicates the The number of cycles that a task occupies, Indicates the The normalized response delay value of a task under the assigned node; S403: calling the resource adaptation matching result, summarizing and arranging the numbers, parameters and matching status between tasks and nodes according to the task node pair identifier, eliminating matching combinations that do not meet the conditions, and establishing a multi-task matching node mapping table.

8. The multi-task processing method in a computing-network fusion environment according to claim 7, characterized in that: The method further comprises the following steps: S5: Based on the multi-task matching node mapping table, a round-trip delay variation ratio sequence is constructed according to the transmission links of the nodes, and the difference is calculated with the corresponding position of the previous cycle sequence. If the difference is greater than the round-trip delay variation threshold, the path is eliminated and the task data transmission processing is carried out using the remaining links to obtain task scheduling execution information; The task scheduling execution information includes a link stability status mark, a path connectivity level classification item, and a node path screening index set.

9. The multi-task processing method in a computing-network fusion environment according to claim 8, characterized in that: The steps for obtaining the task scheduling execution information are specifically as follows: S501: Based on the multi-task matching node mapping table and the established task and node pairing information, extract the transmission link of each task-dependent node, collect the minimum round-trip delay and maximum round-trip delay of the link in the past three monitoring cycles, and construct a round-trip delay ratio sequence in each cycle to obtain a round-trip delay variation ratio sequence set; S502: Calling the round-trip delay variation ratio sequence set, selecting corresponding positions of the current cycle and the previous cycle, calculating the difference between the corresponding ratios of the two cycles, calling a set round-trip delay variation threshold, comparing the difference, marking links greater than the threshold as unstable paths, and generating link fluctuation status marking information; S503: Call the link fluctuation status mark information, remove the corresponding link-task pair from the multi-task matching node mapping table, extract the remaining matching task corresponding links as data transmission paths and complete task data forwarding, and establish task scheduling execution information.

10. A multi-task processing system in a computing-network fusion environment, characterized in that: The system is used to implement the multi-task processing method in a computing-network fusion environment according to any one of claims 1 to 9, and the system includes: The cycle load analysis module obtains the CPU usage, memory occupancy, and task scheduling logs of registered tasks in the computing-network convergence architecture over three consecutive resource allocation cycles, calculates the resource usage concentration of tasks in each cycle, and generates a compact set of cycle load records; The task priority evaluation module calculates the length of the monotonically increasing segment of the sequence based on the periodic load compact record set and the resource usage concentration sequence of the task, and compares it with the set resource density threshold. If the sequence increases for three consecutive cycles and exceeds the threshold, the task is listed as having a high access tendency and a priority task list is obtained; The node scheduling analysis module extracts the CPU peak period in each cycle based on the prioritized task list, calculates the interval between two adjacent peaks of the node, calculates the degree of sequence fluctuation, determines whether the fluctuation is lower than the set stability threshold, and generates a node scheduling adaptation segment list; The task matching evaluation module calculates the resource adaptation ratio based on the node scheduling adaptation segment list and the task request parameters. If the resource adaptation ratio is greater than the set resource adaptation threshold, it is marked as a successful match and a multi-task matching node mapping table is obtained; The scheduling execution module constructs a round-trip delay variation ratio sequence based on the multi-task matching node mapping table and the transmission link of the node, and calculates the difference with the corresponding position of the previous cycle sequence. If the difference is greater than the round-trip delay variation threshold, the path is eliminated and the remaining links are used to implement task data transmission processing to obtain task scheduling execution information.

Images