Multi-task scheduling method, system and device based on trusted computing and storage medium

Abstract

The invention relates to the technical field of task scheduling, in particular to a multi-task scheduling method, system and device based on trusted computing and a storage medium, and the method comprises the following steps: obtaining trusted authentication node resources and task logs, packaging to generate a resource trajectory set, extracting task scheduling information to construct an access intention trajectory, and comparing by utilizing behavior trajectories to obtain an access intention trajectory set; according to the method, the credible authentication identifier of the computing resource and the resource behavior track information are bound and integrated, and the access intention track is established according to the access path structure of the task, so that the access intention track of the computing resource can be obtained; and the authority configuration is dynamically adjusted based on the access strategy level of the calling interface, so that the resource utilization rate and behavior controllability in the scheduling process are improved, the interference risk caused by unexpected behaviors to resource contention is reduced, and orderly execution of the scheduling behaviors in a trusted computing environment is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of task scheduling, in particular to a multi-task scheduling method, system and device based on trusted computing and a storage medium. BACKGROUND

[0002] The technical field of task scheduling involves the ordered allocation and execution of multiple tasks on processors, computing nodes or virtual resources in a system environment with limited computing resources. This field includes task priority determination, resource allocation strategies, scheduling algorithm design and optimization, task dependency management, and other aspects, and is widely applicable to operating system scheduling, parallel computing, distributed systems, cloud computing platforms, and other environments. The technology aims to improve resource utilization, shorten task completion time, reduce task conflicts and waiting time, and thus improve the overall efficiency of the system. In industrial applications, task scheduling technology is also commonly used in manufacturing production scheduling, real-time control systems, and various high-performance computing scenarios.

[0003] Among them, the multi-task scheduling method is a method of allocating tasks to one or more computing resources for coordinated execution in the presence of multiple tasks to be executed. Its purpose is to solve the resource competition, timing conflict and performance bottleneck problems that may occur during task execution. By constructing a reasonable scheduling model and strategy, the task execution order and resource usage are optimized and coordinated, improving the overall processing efficiency of the task, ensuring the timely completion of critical tasks, and meeting the specific quality of service requirements. The method is commonly used in high-concurrency scenarios such as data processing, intelligent control, and cloud computing platform task distribution.

[0004] Traditional scheduling methods lack a trusted evaluation of node resource behavior and a structured expression of task access intentions, which can easily cause behavior conflicts or resource contention in scenarios with dense resource calls or complex interface call sequences. For example, when multiple tasks frequently access the same type of interface without path intersection, the lack of behavior trajectory comparison can lead to incorrect determination of their dependency relationships, resulting in scheduling instruction mismatches, interface blocking or access failures. For tasks with abnormal behavior during scheduling, there is a lack of judgment criteria based on path consistency and call frequency differences, making it difficult to identify potential risk tasks and perform permission shrinkage control in a timely manner, resulting in problems such as decreased stability and disordered execution sequence of task scheduling in high-concurrency environments. SUMMARY

[0005] To solve the technical problems existing in the prior art, the embodiments of the present application provide a multi-task scheduling method based on trusted computing, comprising the following steps: In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme: a multi-task scheduling method based on trusted computing, comprising the following steps: S1: Obtain the computing resource node identified by the trusted authentication unit, generate the behavior trajectory segment arranged by time for each type of resource, bind and integrate the behavior trajectory segment and the trusted identification value of the node, and obtain a trusted resource trajectory encapsulation set; S2: Obtain the to-be-scheduled task identification of the submitted resource binding application, filter high-frequency calling items according to the calling interface priority set by the scheduling control unit, combine the access logic of the task into a structured path description unit, and obtain a task access intention trajectory segment; S3: Based on the task access intention trajectory segment and the trajectory segment of the corresponding resource node in the trusted resource trajectory encapsulation set, perform difference comparison on the calling interval time and the cycle time in the behavior block, perform path comparison according to the interface order and the resource behavior order, count the consistency rate and assign a trustworthiness label, and obtain a behavior path offset comparison result; S4: Call the trustworthiness label field in the behavior path offset comparison result, set the access interface in the current scheduling period to a minimum permission interface set for the task with an abnormal behavior label, and set a maximum calling frequency limit value, to obtain a scheduling task permission configuration group; S5: Call the scheduling path structure of the permission adjustment task in the scheduling task permission configuration group, verify the path continuity, pack the task scheduling position number, control parameter set and permission configuration content included in the continuous path segment, build a path binding label, and write it into the scheduling queue metadata, to obtain a trusted path binding label set.

[0006] As a further scheme of the application, the trusted resource trajectory encapsulation set includes resource node identification, encapsulated calling behavior sequence and trusted authentication state value, the task access intention trajectory segment includes task identification path, scheduling calling intention structure and interface frequency analysis result, the behavior path offset comparison result includes path offset label group, interface calling abnormal segment set and sequence consistency label, the scheduling task permission configuration group includes interface access level label, frequency limit parameter set and permission adjustment record number, and the trusted path binding label set includes path segment node mapping value, control parameter set and metadata writing index.

[0007] As a further scheme of the application, the specific steps of S1 are as follows: S101: Obtain the computing resource node identified by the trusted authentication unit, call the scheduling task execution log in the node, extract the recorded task unique identification number, calling interface number sequence, calling frequency value and task completion state code, perform corresponding aggregation on the calling frequency value and the task completion state code, and classify and merge according to the task number, to generate a task behavior basic data set; S102: based on the calling interface number sequence and the calling frequency value in the task behavior basis data set, the timestamp field and the resource node identifier field corresponding to the calling data item, a calling index arranged in time sequence is constructed, and a calling period segment set corresponding to the task number is extracted, the time segment and the resource identifier field are combined, a resource behavior trajectory paragraph is constructed, and a resource node trajectory arrangement sequence is obtained; S103: according to the time segment set and the resource identifier field in the resource node trajectory arrangement sequence, the trusted authentication unit identifier value attached to each type of resource is bound one by one, the bound data structure is fused with the trajectory arrangement sequence, a complete structured trajectory group arranged in time and distinguished by node is constructed, and a trusted resource trajectory encapsulation set is obtained.

[0008] As a further scheme of the application, the specific steps of S2 are: S201: obtain the task identifier of the to-be-scheduled task for which the resource binding application is submitted, extract the scheduling target resource code, estimated execution time, calling sequence and calling frequency from the submitted scheduling application data block, map the interface number and the calling frequency value one by one to construct a calling frequency structure table, and generate a calling frequency configuration data set; S202: based on the calling frequency value field in the calling frequency configuration data set, according to the interface priority order table preset by the scheduling control unit, screen the interface number items whose position ranking in the priority sequence is not lower than the threshold level, and calculate the average interval value between the calling timestamps of the screened interface items, extract the time-dense segment in the task behavior trajectory, and obtain the preferred interface frequency structure unit; S203: based on the interface number sequence and the average calling interval value extracted in the preferred interface frequency structure unit, a task access path node sequence is constructed, the resource target code and the calling sequence number of each node are mapped to the path structure index, the path sequence is aggregated into a continuous access block according to the time density, and is integrated into a structured access order description data frame to obtain a task access intention trajectory segment.

[0009] As a further scheme of the application, the specific steps of S3 are: S301: call the task access intention trajectory segment and the trusted resource trajectory encapsulation set, extract the calling frequency value and the calling order index value of the two behavior paths according to the interface calling number sequence field in the corresponding mapping scheduling resource node and the interface sequence field in the task path, establish a behavior block index structure according to the calling number field of the two types of data, and obtain an interface behavior matching structure set; S302: Based on the call interval time field and the period time field in the interface behavior matching structure set, the time difference value of the behavior block data is calculated, the threshold comparison is performed on each difference value record, the interface call record set with the difference value greater than the behavior consistency threshold is screened, the behavior block position index is marked, and the abnormal behavior block identification information is obtained; S303: According to the abnormal behavior block identification information and the call sequence index value in the interface behavior matching structure set, the behavior path sequence mapping table between the task and the resource is constructed, the two-side sequence matching rate value is calculated and normalized to convert into a credibility score level, the credibility state marking field is generated by combining the abnormal behavior block quantity and the score level, the structured behavior comparison label frame is established, and the behavior path offset comparison result is obtained.

[0010] As a further scheme of the application, the specific steps of S4 are: S401: The credibility marking field in the behavior path offset comparison result is called, the access control policy level table is retrieved, the credibility marking value is mapped with the matching rule field of the permission level in the level table to perform position index mapping, it is judged whether the task exists the permission downgrade condition corresponding to the abnormal behavior marking, and the task identification set meeting the downgrade strategy is screened to obtain the task permission judgment mapping set; S402: According to the task identification set field in the task permission judgment mapping set, the target resource node currently bound by each type of task in the scheduling queue is retrieved, the interface access level field in the permission configuration item is called, the minimum permission interface set is set for the task with the permission downgrade demand, and the interface frequency upper limit threshold field in the interface frequency management unit is called to set the maximum call frequency of the task, and the interface access restriction parameter group is generated; S403: Based on the task identification, the target resource node and the frequency control threshold parameter associated in the interface access restriction parameter group, the permission configuration record is constructed, the task permission level field, the interface call restriction field and the scheduling period index field are structured and packaged, and written into the scheduling permission mapping area, the task scheduling behavior is bound with the permission identification, and the scheduling task permission configuration group is obtained.

[0011] As a further scheme of the application, the specific steps of S5 are: S501: The task scheduling path structure identified as the permission adjustment state in the scheduling task permission configuration group is called, the call interface sequence and the target resource node number of the current task are extracted, the interface sequence position index one-to-one mapping is performed in combination with the behavior template path structure in the trusted resource trajectory encapsulation set, the Boolean judgment is performed according to whether the task path segment is continuous with the template path segment, and the path continuity verification label set is obtained. S502: Based on the path continuity verification label set in the continuous marking path segment, the corresponding path segment is extracted, the task scheduling position number field, the binding control parameter set field and the permission configuration parameter field are extracted, each type of field is aggregated into a unit block structure in the order of the path segment, the data is reorganized according to the task identification number, the path number and the permission control element are integrated into a combined structure set, and a scheduling path binding structure frame is generated; S503: According to each path segment combined structure in the scheduling path binding structure frame, a structured scheduling label field block including task identification, path number and permission configuration data is constructed, the field block is written into the corresponding task slot in the scheduling queue metadata allocation area, a scheduling path binding index structure is established for the task scheduling record, and a trusted path binding label set is obtained.

[0012] A multi-task scheduling system based on trusted computing, the system comprising: A resource trajectory identification module, which obtains a computing resource node identified by a trusted authentication unit, generates a behavior trajectory segment arranged in time for each type of resource, integrates the behavior trajectory segment with the trusted identification value of the node, and obtains a trusted resource trajectory encapsulation set; An access intention analysis module, which obtains a to-be-scheduled task identification submitted for resource binding application, filters high-frequency calling items according to the calling interface priority set by the scheduling control unit, combines the access logic of the task into a structured path description unit, and obtains a task access intention trajectory segment; A path offset comparison module, which performs difference comparison on the calling interval time and the cycle time in the behavior block based on the task access intention trajectory segment and the trajectory segment of the corresponding resource node in the trusted resource trajectory encapsulation set, performs path comparison according to the interface order and the resource behavior order, calculates the consistency rate and assigns a trustworthiness mark, and obtains a behavior path offset comparison result; A task permission analysis module, which calls the trustworthiness mark field in the behavior path offset comparison result, sets the access interface in the current scheduling period as a minimum permission interface set for the task with an abnormal behavior mark, and sets a maximum calling frequency limit value, and obtains a scheduling task permission configuration group; A path binding processing module, which calls the scheduling path structure of the permission adjustment task in the scheduling task permission configuration group, verifies the path continuity, packs the task scheduling position number, the control parameter set and the permission configuration content included in the continuous path segment, constructs a path binding label, and writes the path binding label into the scheduling queue metadata to obtain a trusted path binding label set.

[0013] A computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the multi-task scheduling system based on trusted computing as described above when executing the computer program.

[0014] A computer readable storage medium, having stored thereon a computer program, the computer program being executed by a processor to implement the steps of the multi-task scheduling method based on trusted computing as described above.

[0015] Compared with the prior art, the application has the advantages and positive effects that: In the application, by binding and integrating the trusted authentication identifier of the computing resource with the resource behavior trajectory information, and establishing the access intention trajectory according to the access path structure of the task, the abnormal deviation in the calling behavior can be accurately identified by comparing the sequence difference of the behavior trajectory and the access intention in the calling frequency, calling sequence and time interval, and the permission configuration is dynamically adjusted based on the access policy level of the calling interface, the fine matching of the resource access and the permission control in the task scheduling is realized by verifying the task path continuity and constructing the path binding label combined with the scheduling position and the control parameter, the resource utilization and the behavior controllability in the scheduling process are improved, the interference risk caused by the resource contention of the unexpected behavior is reduced, the real-time response capability and the control strength of the scheduling system to the abnormal behavior are enhanced, the ordered execution of the scheduling behavior in the trusted computing environment is ensured, and the task scheduling safety and the execution stability in the multi-task environment are improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The step flowchart of the present application is shown in the figure. Figure 2 The S1 refinement diagram of the present application is shown in the figure. Figure 3 The S2 refinement diagram of the present application is shown in the figure. Figure 4 The S3 refinement diagram of the present application is shown in the figure. Figure 5 The S4 refinement diagram of the present application is shown in the figure. Figure 6 The S5 refinement diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0018] The technical solutions in the present application will be described below with reference to the drawings.

[0019] In the embodiments of the present application, the words such as "exemplary", "for example", etc. are used to represent an example, illustration, or description. Any embodiment or design scheme described as "exemplary" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "exemplary" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0020] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized. "Of", "corresponding" and "corresponding" can be used interchangeably at times, and it should be pointed out that the meanings expressed are consistent when the distinction is not emphasized.

[0021] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1, and the meanings expressed are consistent when the distinction is not emphasized.

[0022] In order to make the technical problems, technical schemes and advantages to be solved by the present application more clear, the following will be described in detail in conjunction with the drawings and specific embodiments.

[0023] Please refer to Figure 1 The embodiments of the present application provide a multi-task scheduling method based on trusted computing, comprising the following steps: S1: Obtain the computing resource nodes identified by the trusted authentication unit, extract the task binding log recorded in each node, combine the task identification, call number sequence, call frequency and task state confirmation code to perform encapsulation, construct a resource usage order table according to the time stamp and resource ID, generate a time-arranged behavior trajectory segment for each type of resource, bind and integrate the behavior trajectory segment with the trusted identification value of the node, and obtain a trusted resource trajectory encapsulation set; S2: Obtain the task identification of the to-be-scheduled task submitted for resource binding application, extract the scheduling target resource code, estimated execution time, call sequence and call frequency from the submitted scheduling application data block, filter high-frequency call items according to the call interface priority set by the scheduling control unit, and record the average call interval time, combine the task access logic into a structured path description unit, and obtain a task access intention trajectory segment; S3: Based on the task access intention trajectory segment and the corresponding resource node trajectory segment in the trusted resource trajectory encapsulation set, the behavior block sequence in the calling frequency and calling sequence is extracted, the calling interval time and periodic time in the behavior block are mapped and difference comparison is performed, the behavior block with a deviation range exceeding the behavior consistency threshold set by the scheduling unit is recorded as an abnormal item, path comparison is performed according to the interface sequence and resource behavior sequence, the order consistency rate is counted and a trustworthiness label is given, and a behavior path deviation comparison result is obtained; The behavior consistency threshold refers to the calling interval deviation interval range allowed by the system calling behavior between different tasks, which is used to judge whether the task behavior mode is within the trusted range, and is usually composed of the calling interval standard deviation and the set tolerance coefficient statistically calculated from the historical behavior data, that is, the behavior consistency threshold is the product of the calling interval standard deviation and the tolerance coefficient, the calling interval standard deviation is the fluctuation degree of the different interface calling time interval, and the tolerance coefficient is generally the value corresponding to the 95% confidence interval of the normal distribution; The trustworthiness label is a classification label of the similarity degree of the task behavior trajectory in the scheduling process to the historical trajectory of the resource, which is used to adjust the access permission policy, and is an enumeration class identification value, which is used for scheduling system permission decision; S4: The trustworthiness label field in the calling behavior path deviation comparison result is matched and judged with the access control policy level table, the tasks with abnormal behavior labels are set as the minimum permission interface set in the current scheduling period, and the maximum calling frequency limit value is set, the corresponding permission adjustment record is established, and a scheduling task permission configuration group is obtained; S5: The scheduling path structure of the permission adjustment task in the calling scheduling task permission configuration group is matched according to the current calling sequence of the task and the behavior template path in the trusted resource node, the path continuity is verified, the task scheduling position number, control parameter set and permission configuration content included in the continuous path segment are packaged, the path binding label is constructed, and written into the scheduling queue metadata, and a trusted path binding label set is obtained.

[0024] The trusted resource trajectory encapsulation set includes resource node identification, encapsulated calling behavior sequence and trusted authentication state value, the task access intention trajectory segment includes task identification path, scheduling calling intention structure and interface frequency analysis result, the behavior path deviation comparison result includes path deviation label group, interface calling abnormal segment set and order consistency label, the scheduling task permission configuration group includes interface access level label, frequency limit parameter set and permission adjustment record number, and the trusted path binding label set includes path segment node mapping value, control parameter set and metadata writing index.

[0025] Please refer to Figure 2 , the specific steps of S1 are: S101: Obtain the computing resource node identified by the trusted authentication unit, call the scheduling task execution log in the node, extract the recorded task unique identification number, call interface number sequence, call frequency value and task completion status code, correspondingly aggregate the call frequency value and task completion status code, and classify and merge according to the task number to generate a task behavior basic data set; A server resource with node number "Node-A7B3" is set, and a scheduling task execution log file stored in the local log system of "Node-A7B3" is accessed and called, and the storage path of the log file is " / var / log / scheduler / execution.log". For all task execution entries recorded in the log file within the past day, each entry is parsed. Four key fields in each log record are extracted: task unique identification number, such as "Task-001"; call interface number sequence, which is a time-ordered list recording all called interface numbers during task execution, such as "[API-01, API-02, API-01, API-03]"; call frequency value, which records the average number of interface calls per unit time (second) of the task, such as "15 times / second"; and task completion status code, which is an integer representing the final execution result of the task, such as "200" for success and "503" for service unavailable. The extracted data is processed, and multiple call frequency values generated by each task during its life cycle are associated with the corresponding task completion status code one by one to form a set containing multiple (call frequency value, task completion status code) data pairs. For example, for task "Task-001", its records in different execution stages may be (15 times / second, 200), (18 times / second, 200) and (12 times / second, 503). These data pairs are aggregated with the call interface number sequence of the task, and finally classified and merged according to the task unique identification number. For example, the final merged record of task "Task-001" is: {task number: "Task-001", interface sequence: "[API-01, API-02, API-01, API-03]", frequency and status set: [(15, 200), (18, 200), (12, 503)]}. Repeat this process for all different task identifiers in the log file to finally generate a structured task behavior basic data set.

[0026] S102: Based on the call interface number sequence and call frequency value in the task behavior basic data set, the timestamp field and resource node identification field corresponding to the data item, construct a call index arranged in chronological order, and extract the call cycle segment set corresponding to the task number, combine the time segment and resource identification field, construct the resource behavior track paragraph, and obtain the resource node track arrangement sequence; From the original execution log, two additional fields are extracted for each interface call behavior: the timestamp field corresponding to the data item, accurate to milliseconds, and the resource node identification field executing the call. Based on the interface call records supplemented with timestamps, a global, strictly ordered by time, call index is constructed. This index structure takes the timestamp as the primary key and records the interface call events occurring at a specific time. Next, for each independent task number, all call records belonging to the task are extracted from the time-ordered call index. By analyzing the intervals between the timestamps of these records, the call period segments are divided. A "call interruption time threshold" is set, and when the time interval between two consecutive interface calls exceeds this threshold, it is considered that a call period segment ends and the next segment begins. The setting of this threshold is based on statistical analysis of the interface call time intervals of more than ten thousand completed tasks in the past month, taking the 95th percentile value as the standard. For example, through analysis, it is found that 95% of the call intervals are less than 500 milliseconds, so the "call interruption time threshold" is set to 500 milliseconds. If the task "Task-001" on the resource node "Node-A7B3" has a call timestamp sequence of [...10:00:01.100, 10:00:01.300, 10:00:02.500, 10:00:02.600...], since the time difference between.300 and.500 is 1200 milliseconds, which is greater than 500 milliseconds, it is divided between.300 and.500 to form two independent call period segments. Each time segment (including the start time and end time) thus divided is combined with the resource identification field "Node-A7B3" to construct a resource behavior trajectory paragraph. Finally, all resource behavior trajectory paragraphs generated by a task on different resource nodes are arranged and combined in time order to obtain the complete resource node trajectory arrangement sequence of the task.

[0027] S103: According to the time segment set in the resource node trajectory arrangement sequence and the resource identification field, the identification value of the trusted authentication unit attached to each type of resource is bound one by one. The bound data structure is fused with the trajectory arrangement sequence to construct a complete structured trajectory group aggregated by time and distinguished by node, and a trusted resource trajectory encapsulation set is obtained. The pre-established "trusted resource authentication library" is accessed, and the library stores the trusted authentication unit identification value attached to each computing resource node and various resources deployed therein. The identification value is a unique encrypted signature generated by the trusted platform module (TPM) when the resource is registered in the trusted pool. From the track arrangement sequence of the resource node, the time segment and its corresponding resource identification field are extracted one by one. According to the resource identification field, the "trusted resource authentication library" is queried to find and extract the trusted authentication unit identification value corresponding to the resource identification. The trusted authentication unit identification value obtained by the query is bound with the original time segment and resource identification field to form a new data structure. For example, the original track paragraph is {time segment: [10:00:01.100, 10:00:01.300], resource identification: "Node-A7B3"}, the trusted authentication identification value of "Node-A7B3" obtained by the query is "Cert-XYZ-789", and the bound data structure is {time segment: [10:00:01.100, 10:00:01.300], resource identification: "Node-A7B3", trusted identification: "Cert-XYZ-789"}. All the new data structures after the binding processing are fused with the original track arrangement sequence. The specific process of the fusion operation is to add a "trusted identification" field in each track paragraph of the original sequence, and fill in the bound value. In this way, a complete structured track group aggregated by time and distinguished by node is constructed. The track group not only records the behavior time sequence of the resource, but also attaches the trusted proof of the execution environment to each behavior segment, and finally obtains the trusted resource track encapsulation set.

[0028] Please refer to Figure 3 The specific steps of S2 are: S201: Obtain the identification of the to-be-scheduled task submitted by the resource binding application, extract the scheduling target resource code, estimated execution time, call sequence and call frequency from the submitted scheduling application data block, map the interface number and call frequency value one by one to construct the call frequency structure table, and generate the call frequency configuration data set; Obtain its unique to-be-scheduled task identifier, such as "Task-New-002". Parse the scheduling application data block submitted together with this task, which is a structured data packet, usually in JSON or XML format. Extract four types of information from the data block: scheduling target resource code, that is, the unique identifier of the computing resource node on which the task expects to run, such as "Node-B8C4"; estimated execution time, that is, the running time estimated by the task developer according to the task complexity and data volume, such as "3600 seconds"; call sequence, that is, the list of interface numbers called by the task plan in sequence, such as "[API-DataRead, API-DataProcess, API-DataWrite]"; and call frequency, that is, the estimated total number of calls for each interface, such as "{API-DataRead: 5000, API-DataProcess: 10000, API-DataWrite: 5000}". According to the extracted call sequence and call frequency information, map the interface number to the corresponding estimated call frequency value one by one to construct a call frequency structure table. This table clearly describes the access intensity of the task to each interface. For example, for task "Task-New-002", its call frequency structure table is: [{interface number: "API-DataRead", call frequency: 5000}, {interface number: "API-DataProcess", call frequency: 10000}, {interface number: "API-DataWrite", call frequency: 5000}]. Integrate this call frequency structure table with other extracted information (task identifier, target resource, estimated time) to generate the call frequency configuration dataset of the to-be-scheduled task.

[0029] S202: Based on the call frequency configuration dataset, according to the interface priority ranking table preset by the scheduling control unit, filter the interface number items whose position ranking in the priority sequence is not lower than the threshold level, and calculate the average interval value between the call timestamps of the filtered interface items to extract the time-dense segment in the task behavior trajectory, obtaining the preferred interface frequency structure unit; An interface priority ranking table is accessed, which ranks all available interfaces in the system according to their importance and the degree of consumption of system resources. For example, the priority of a core data processing interface is higher than that of a regular log recording interface. The priority setting benchmark is statistical analysis of the impact of various types of task running on system load in history. Interface calls that cause the average CPU load to exceed 80% are classified as high priority, 50% to 80% as medium priority, and less than 50% as low priority. Next, each interface item in the call frequency configuration data set is traversed, and its interface number is compared with the interface priority ranking table to filter out interface number items whose position ranking in the priority sequence is not lower than the preset "priority threshold level". This threshold level is set to "medium priority", that is, all medium and high priority interfaces are filtered out. For example, if "API-DataProcess" is high priority and "API-DataRead" and "API-DataWrite" are medium priority, then these three interfaces are all filtered out. For all filtered interface items, the average interval value between the call timestamps of these high priority interfaces during task execution is calculated. This calculation process is to calculate the estimated timestamp of each call according to the call sequence and estimated execution time in the simulation execution environment, and then calculate the average interval. Through this average interval value, the time-dense segment with the most intensive call activity in the task behavior track is identified and extracted. Finally, the filtered interfaces and their related call frequency data are combined into preferred interface frequency structure units.

[0030] S203: Based on the extracted interface number sequence and average call interval value in the preferred interface frequency structure unit, a task access path node sequence is constructed, the resource target code of each node is mapped to the call sequence number for path structure indexing, the path sequence is aggregated according to time density into continuous access blocks, integrated into structured access order description data frames, and a task access intention track segment is obtained; A task access path node sequence is constructed, and each node in the sequence represents an interface call event. The resource target code (obtained from the original application) of each node is mapped to the path structure index with the node number in the call sequence (i.e. its position index in the sequence). For example, for the task "Task-New-002", the first node of its access path node sequence can be represented as {resource target: "Node-B8C4", sequence number: 1, interface: "API-DataRead"}. Next, according to the average call interval value of the interface, the consecutive nodes in the path sequence with similar time intervals are aggregated into "continuous access blocks". The aggregation criterion is that if the estimated call time intervals of consecutive nodes are all less than 1.2 times the average call interval value, these nodes are divided into a continuous access block. This 1.2 times coefficient is an empirical value based on the analysis of historical normal task execution data. Through the statistical analysis of the call interval fluctuation of more than five thousand task samples, it is found that 90% of the normal fluctuation range is within 20%, so this coefficient is set. All the continuous access blocks formed by aggregation are integrated, and each access block represents the centralized access behavior of the task to the resource within a time period. Finally, these continuous access blocks are integrated into a structured access order description data frame, which is the task access intention trajectory segment, which describes how the task plan accesses the target resource in stages and intensity.

[0031] Referring to Figure 4 The specific steps of S3 are as follows: S301: Call the task access intention trajectory segment and the trusted resource trajectory encapsulation set, extract the call frequency value and the call order index value of the two behavior paths according to the corresponding mapping of the interface call number sequence field in the resource node and the interface sequence field in the task path, establish the behavior block index structure according to the call number field for the two types of data, and obtain the interface behavior matching structure set; From the task access intention trajectory segment, the interface call number sequence field is extracted. From the trusted resource trajectory encapsulation set, the interface call number sequence field in the history of the corresponding resource node is extracted. The interface sequence of the task and the historical interface sequence of the resource are traversed, and a temporary behavior block index structure is established according to the same interface call number field appearing in the two sequences. Specifically, taking the interface in the task sequence as the benchmark, the positions of all occurrences of the interface in the resource historical sequence are found, and these position information is recorded. For example, the first interface of the task sequence is “API-DataRead”, and “API-DataRead” is also called at the 3rd, 15th and 28th positions in the resource historical sequence. The index is established to associate the 1st position of the task with the 3rd, 15th and 28th positions of the resource history. While establishing the index, the call frequency value of each interface is extracted from the task access intention trajectory segment, and the average call frequency value of the interface in the history is extracted from the trusted resource trajectory encapsulation set. At the same time, the call order index value of both sides, i.e. the position number of the interface in the respective sequence, is also extracted. Fill these extracted call frequency values and call order index values according to the established behavior block index structure, match and organize the two types of data. Finally, an interface behavior matching structure set is obtained, wherein each record contains the paired comparison of the frequency and order information of the task intention and the historical behavior of the resource at a specific interface.

[0032] S302: Based on the call interval time field and the period time field in the interface behavior matching structure set, the time difference value of the behavior block data is calculated, the threshold value comparison is performed on each difference value record, the interface call record set with a difference value greater than the behavior consistency threshold value is screened, the behavior block position index is marked, and the abnormal behavior block identification information is obtained; The calling interval time is from the task access intention track segment, indicating the calling rhythm of the task plan; the cycle time field is from the trusted resource track encapsulation set, indicating the average calling rhythm in the resource history. For each matching record in the interface behavior matching structure set, calculate the time difference value, that is, subtract the cycle time of the resource history from the calling interval time of the task, and take the absolute value. Then, compare each difference record calculated with a preset "behavior consistency threshold". The setting of this threshold is through the analysis of a large number of (for example, more than ten thousand) known normal task historical execution data on the target resource, calculating the difference between the planned calling interval and the actual execution average interval, and taking the 85th percentile of all difference values as the threshold. For example, according to statistical calculation, 85% of the normal behavior difference is within 30 milliseconds, and the "behavior consistency threshold" is set to 30 milliseconds. When performing threshold comparison, a set of interface calling records with a time difference value greater than 30 milliseconds is screened out. For each screened record, that is, the inconsistent record, mark the behavior block position index in the interface behavior matching structure set. This set of marked information, that is, the abnormal behavior block identification information, clearly indicates where the task access intention deviates significantly from the historical behavior pattern of the resource at specific calling points.

[0033] S303: According to the abnormal behavior block identification information and the calling sequence index value in the interface behavior matching structure set, construct the behavior path sequence mapping table between the task and the resource, calculate the sequence matching rate value on both sides and normalize it to convert it into a credibility score level, combine the abnormal behavior block quantity and the score level to generate a trust state marking field, establish a structured behavior comparison tag frame, and obtain the behavior path deviation comparison result; The mapping table associates the calling sequence index of the task with the calling sequence index in the historical behavior of the resource, and calculates the matching rate value of the sequence on both sides. The calculation method is as follows: first, count the number of matching items in the behavior path sequence mapping table whose sequence indexes are completely consistent or within the allowed small deviation (for example, the deviation is less than or equal to 2), and mark it as "total number of matching items". Then, divide the "total number of matching items" by the total length of the task calling sequence to obtain the original sequence matching rate. Normalize the calculated sequence matching rate value to convert it into a credibility score level of 0 to 100 points. The normalization conversion rule is: multiply the matching rate by 100. For example, if the matching rate is 0.95, the credibility score is 95. Combine the number of abnormal behavior blocks to modify the credibility score. The modification rule is: for each abnormal behavior block, deduct a fixed score, for example, 5 points, from the credibility score. The setting of this deduction score is based on the risk assessment model to quantify the security risks that different numbers of abnormal behaviors may cause. For example, the initial score is 95 points, and there are 2 abnormal behavior blocks, so the final score is 95-2 5=85 points. According to the final credibility score level, a credibility status marker field is generated, such as "high credibility", "moderate credibility", "low credibility". The score interval is defined as: 80 points and above is "high credibility", 60 to 79 points is "moderate credibility", and less than 60 points is "low credibility". Finally, the credibility score, credibility status marker and detailed path mapping relationship are integrated to establish a structured behavior comparison label frame, which is the behavior path deviation comparison result.

[0034] Please refer to Figure 5 The specific steps of S4 are: S401: Call the credibility marker field in the behavior path deviation comparison result, retrieve the access control policy level table, perform position index mapping of the credibility marker value and the matching rule field of the permission level in the level table, judge whether the task has the permission downgrade condition corresponding to the abnormal behavior marker, and filter the task identifier set that meets the downgrade strategy to obtain the task permission judgment mapping set; The pre-configured access control policy level table is retrieved, which defines the matching rules between different credibility marker values and specific permission levels. For example, the policy table stipulates that "high credibility" corresponds to "full access permission", "moderate credibility" corresponds to "restricted access permission", and "low credibility" corresponds to "access denied". The credibility marker value obtained from the behavior path deviation comparison result is mapped with the rule field in the level table. The specific operation is to use the credibility marker value as an index to find its corresponding permission level in the policy table. For example, if the marker value is "moderate credibility", it is mapped to "restricted access permission". At the same time of mapping, it is judged whether the task has the permission downgrade condition corresponding to the abnormal behavior marker. The basis for judgment is that if the permission level obtained by mapping is lower than the permission level originally applied by the task (for example, the task applies for "full access permission", but the mapping result is "restricted access permission"), it is determined that the task meets the permission downgrade condition. Repeat this process for all tasks being judged for permission, and filter out the unique identifiers of all tasks that meet the downgrade strategy to form a task identifier set. This set finally constitutes the task permission judgment mapping set, which contains a list of all tasks that need to be adjusted for permission.

[0035] S402: According to the task identifier set field in the task permission judgment mapping set, retrieve the target resource node currently bound by each type of task in the scheduling queue, call the interface access level field in the permission configuration item, set the minimum permission interface set for the task that needs to be downgraded, and call the interface frequency upper threshold field in the interface frequency management unit to set the maximum call frequency of the task, to generate an interface access restriction parameter group; For each task identifier in the set, the current scheduling queue is retrieved to find the target resource node that each task is currently scheduled to bind or has bound to. Then, the permission configuration item associated with the target resource node is accessed, particularly the interface access level field, which lists the interfaces allowed to access under different permission levels. For each task that requires permission downgrade (i.e., in the task identifier set), its permission configuration is set to the "minimum permission interface set". The definition of the "minimum permission interface set" is that, under the mapped "restricted access permission" level, only the access right to the interfaces necessary for core functions and with the lowest security risk assessment is retained. For example, the access right to system configuration modification type interfaces is deprived, and only data reading interfaces are retained. At the same time, the interface frequency management unit is called to obtain the interface frequency upper threshold field preset for the "restricted access permission" level. This threshold is determined based on the stability test data of the system under different loads, and is intended to prevent abuse of resources. For example, the maximum calling frequency of any interface by the task in the restricted state is set to not more than "10 times per second". The "minimum permission interface set" and the "maximum calling frequency" are combined to generate a specific interface access restriction parameter set for each task that requires downgrade.

[0036] S403: Based on the task identifier, target resource node, and frequency control threshold parameter in the interface access restriction parameter set, a permission configuration record is constructed, the task permission level field, interface calling restriction field, and scheduling period index field are structured and packaged, and written into the scheduling permission mapping area to bind the task scheduling behavior with the permission identifier, obtaining a scheduling task permission configuration group; For each task that requires permission adjustment, a new permission configuration record is constructed. In this record, the permission level field of the task is updated to the downgraded level (e.g., from "full access" to "restricted access"), the interface calling restriction field is set to the "minimum permission interface set", and the scheduling period index field is associated with the new maximum calling frequency. These fields are structured and packaged to form a unified data packet. Subsequently, the data packet is written into the scheduling permission mapping area, a core component of the scheduling system. The writing operation will overwrite or update the original permission configuration of the task, thereby realizing the immediate permission identifier binding of the task scheduling behavior. This process ensures that when the scheduler processes the task next time, it will perform operations according to the latest and limited permission. The set of all newly generated permission configuration records finally constitutes the scheduling task permission configuration group, which serves as the final basis for system permission control.

[0037] Please refer to Figure 6 , the specific steps of S5 are as follows: S501: Call the task scheduling path structure identified as the permission adjustment state in the scheduling task permission configuration group, extract the current task's call interface sequence and target resource node number, combine the resource node's behavior template path structure in the trusted resource trajectory encapsulation set, perform interface sequence position index one-to-one mapping, perform Boolean judgment according to whether the task path segment is continuous with the template path segment, and obtain the path continuity verification tag set; The interface sequence of the current task plan call and the locked target resource node number are extracted. At the same time, from the trusted resource trajectory encapsulation set, find the behavior template path structure corresponding to the target resource node number, which represents the historical behavior mode verified as normal and reliable on the resource node. Next, the position index one-to-one mapping of the interface sequence is performed. The specific process is to compare the call interface sequence of the task with the interface sequence in the behavior template path structure of the resource, starting from the first position. According to whether a certain continuous segment of the task path completely coincides with the corresponding segment of the template path, a Boolean judgment is performed. The definition of complete coincidence is that in the continuous N (N is a preset minimum matching length of path segment, for example, 5) interface calls, the interface number and relative order of the two sequences are completely consistent. If consistent, the judgment result of the path segment is "true" (True), otherwise "false" (False). This judgment is performed on the entire call path segment of the task, and finally a path continuity verification tag set composed of Boolean values is obtained, which identifies which part of the task path is consistent with the historical trusted behavior.

[0038] S502: Based on the continuous marked path segment in the path continuity verification tag set, extract the task scheduling position number field, the binding control parameter set field and the permission configuration parameter field in the corresponding path segment, aggregate each type of field into a unit block structure according to the path segment order, reorganize the data according to the task identification number, integrate the path number and the permission control element into a combined structure set, and generate a scheduling path binding structure frame; All the continuous marked path segments marked as "True" are filtered out. For each of these path segments, three key field sets are extracted from the original scheduling information: the numbering field of each scheduling position within the path segment, the control parameter set field (e.g. timeout time, retry times) bound to these scheduling positions, and the set permission configuration parameter field (e.g. restricted interface list and frequency upper limit). Each type of extracted field is strictly aggregated according to its original order within the path segment to form a unit block structure. Then, all the unit block structures belonging to the same task are reorganized according to the unique identification number of the task. In the reorganization process, the numbering of the path segment and the permission control elements (i.e. control parameters and permission parameters) contained therein are integrated into a close combination structure set. This structure set clearly describes the detailed execution rules and permission restrictions that the task should follow on a trusted path segment. A scheduling path binding structure frame is generated for each task, which firmly binds the trusted execution path of the task with the specific control and permission policy.

[0039] S503: According to each path segment combination structure in the scheduling path binding structure frame, a structured scheduling label field block including the task identification, path numbering, and permission configuration data is constructed, the field block is written into the corresponding task slot in the metadata allocation area of the scheduling queue, a scheduling path binding index structure is established for the task scheduling record, and a trusted path binding label set is obtained; A structured scheduling label field block is constructed for each structure, which is a compact data unit that explicitly contains the task identification, path numbering, and detailed permission configuration data. After construction, the field block is written into the metadata allocation area of the scheduling queue. The specific operation is to locate the task slot corresponding to the task identification in the scheduling queue, and write the field block into the reserved metadata field. This write operation is a direct update to the task record in the scheduling queue, ensuring that the scheduler can read the latest and most accurate scheduling and permission information before executing the task. At the same time of writing, a scheduling path binding index structure is established for the scheduling record of the task. This index structure allows the system to quickly query all the trusted paths bound to the task identification and their corresponding permission configurations without re-parsing the entire scheduling plan. By repeating this process for all verified tasks in the queue, a trusted path binding label set is finally obtained, which is the final scheduling instruction set with fine-grained trusted and secure policies applied to actual task execution.

[0040] The multi-task scheduling system based on trusted computing comprises: A resource track identification module obtains the computing resource nodes identified by the trusted authentication unit, generates a behavior track segment arranged by time for each type of resource, binds and integrates the behavior track segment and the trusted identification value of the node, and obtains a trusted resource track encapsulation set; An access intention analysis module obtains the to-be-scheduled task identification of the submitted resource binding application, filters high-frequency calling items according to the calling interface priority set by the scheduling control unit, combines the access logic of the task into a structured path description unit, and obtains a task access intention track segment. A path offset comparison module performs difference comparison on the calling interval time and the cycle time in the behavior block based on the task access intention track segment and the track segment of the corresponding resource node in the trusted resource track encapsulation set, performs path comparison according to the interface order and the resource behavior order, counts the consistency rate and gives a trustworthiness label, and obtains a behavior path offset comparison result. A task permission analysis module calls the trustworthiness label field in the behavior path offset comparison result, sets the access interface in the current scheduling period as a minimum permission interface set for the task with an abnormal behavior label, and sets a maximum calling frequency limit value, and obtains a scheduling task permission configuration group. A path binding processing module calls the scheduling path structure of the permission adjustment task in the scheduling task permission configuration group, verifies the path continuity, packs the task scheduling position number, control parameter set and permission configuration content included in the continuous path segment, constructs a path binding label, and writes the path binding label into the scheduling queue metadata, and obtains a trusted path binding label set.

[0041] A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned multi-task scheduling system based on trusted computing when executing the computer program.

[0042] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the above-mentioned multi-task scheduling method based on trusted computing.

[0043] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A multi-task scheduling method based on trusted computing, characterized in that, The method comprises the following steps: S1: obtaining a computing resource node identified by a trusted authentication unit, generating a time-arranged behavior track segment for each type of resource, binding and integrating the behavior track segment and the trusted identification value of the node to obtain a trusted resource track encapsulation set; S2: obtaining a to-be-scheduled task identification submitted by a resource binding application, screening high-frequency calling items according to a calling interface priority set by a scheduling control unit, combining the access logic of the task into a structured path description unit to obtain a task access intention track segment; S3: performing difference comparison on the calling interval time and the periodic time in the behavior block based on the task access intention track segment and the track segment of the corresponding resource node in the trusted resource track encapsulation set, performing path comparison according to the interface sequence and the resource behavior sequence, counting the consistency rate and assigning a trustworthiness label to obtain a behavior path offset comparison result; S4: calling the trustworthiness label field in the behavior path offset comparison result, setting the access interface in the current scheduling period as a minimum permission interface set for the task with an abnormal behavior label, and setting a maximum calling frequency limit value to obtain a scheduling task permission configuration group; S5: calling the scheduling path structure of the permission adjustment task in the scheduling task permission configuration group, verifying the path continuity, packing the task scheduling position number, control parameter set and permission configuration content included in the continuous path segment, constructing a path binding label, and writing the path binding label into the scheduling queue metadata to obtain a trusted path binding label set.

2. The multi-task scheduling method based on trusted computing according to claim 1, characterized in that, The trusted resource track encapsulation set comprises a resource node identification, an encapsulated calling behavior sequence and a trusted authentication state value, the task access intention track segment comprises a task identification path, a scheduling calling intention structure and an interface frequency analysis result, the behavior path offset comparison result comprises a path offset label group, an interface calling abnormal segment set and a sequence consistency label, the scheduling task permission configuration group comprises an interface access level label, a frequency limit parameter set and a permission adjustment record number, and the trusted path binding label set comprises a path segment node mapping value, a control parameter set and a metadata writing index. 3.The multi-task scheduling method based on trusted computing according to claim 1, wherein, The specific steps of S1 are as follows: S101: obtaining a computing resource node identified by a trusted authentication unit, calling the scheduling task execution log in the node, extracting the recorded task unique identification number, calling interface number sequence, calling frequency value and task completion state code, corresponding aggregation of the calling frequency value and the task completion state code, and classification and merging according to the task number to generate a task behavior basic data set; S102: based on the calling interface number sequence and the calling frequency value in the task behavior basic data set, calling the timestamp field and the resource node identification field corresponding to the data item, constructing a calling index arranged in time sequence, and extracting a calling period segment set corresponding to the task number, combining the time segment and the resource identification field, constructing a resource behavior track paragraph, and obtaining a resource node track arrangement sequence; S103: According to the time segment set in the resource node trajectory arrangement sequence and the resource identification field, the identification value of the trusted authentication unit attached to each type of resource is bound one by one, the bound data structure is fused with the trajectory arrangement sequence, a complete structured trajectory group is constructed according to time aggregation and node differentiation, and a trusted resource trajectory encapsulation set is obtained.

4. The multi-task scheduling method based on trusted computing according to claim 1, characterized in that, The specific steps of S2 are: S201: Obtain the to-be-scheduled task identification of the submitted resource binding application, extract the scheduling target resource code, estimated execution time, call sequence and call frequency from the submitted scheduling application data block, map the interface number and call frequency value one by one to construct a call frequency structure table, and generate a call frequency configuration data set; S202: Based on the call frequency value field in the call frequency configuration data set, according to the interface priority order table preset by the scheduling control unit, filter the interface number items whose position ranking in the priority sequence is not lower than the threshold level, and calculate the average interval value between the call time stamps of the filtered interface items, extract the time dense segment in the task behavior trajectory, and obtain the preferred interface frequency structure unit; S203: Based on the interface number sequence and the average call interval value extracted in the preferred interface frequency structure unit, construct a task access path node sequence, map the resource target code of each node and the call sequence number to the path structure index, aggregate the path sequence into continuous access blocks according to time density, integrate into a structured access order description data frame, and obtain a task access intention trajectory segment.

5. The multi-task scheduling method based on trusted computing according to claim 1, characterized in that, The specific steps of S3 are: S301: Call the task access intention trajectory segment and the trusted resource trajectory encapsulation set, extract the call frequency value and call order index value of the two behavior paths according to the corresponding mapping of the interface call number sequence field in the scheduling resource node and the interface sequence field in the task path, establish a behavior block index structure according to the call number field of the two types of data, and obtain an interface behavior matching structure set; S302: Based on the call interval time field and the period time field in the interface behavior matching structure set, calculate the time difference value of the behavior block data, perform threshold comparison on each difference record, filter the interface call record set whose difference value is greater than the behavior consistency threshold, mark the behavior block position index, and obtain the abnormal behavior block identification information; S303: According to the abnormal behavior block identification information and the call order index value in the interface behavior matching structure set, construct a behavior path order mapping table between the task and the resource, calculate the order matching rate value on both sides and normalize it into a trust score level, combine the abnormal behavior block quantity and the score level to generate a trusted state marker field, establish a structured behavior comparison label frame, and obtain a behavior path offset comparison result.

6. The multi-task scheduling method based on trusted computing according to claim 1, characterized in that, The specific steps of S4 are: S401: Call the credibility mark field in the behavior path offset comparison result, retrieve the access control policy level table, perform position index mapping on the credibility mark value and the matching rule field of the permission level in the level table, judge whether the task exists the permission downgrade condition corresponding to the abnormal behavior mark, and filter the task identification set that meets the downgrade strategy to obtain the task permission judgment mapping set; S402: According to the task identification set field in the task permission judgment mapping set, retrieve the target resource node currently bound by each type of task in the scheduling queue, call the interface access level field in the permission configuration item, set the minimum permission interface set for the task with permission downgrade demand, and call the interface frequency upper limit threshold field in the interface frequency management unit to set the maximum calling frequency of the task, and generate the interface access restriction parameter group; S403: Based on the task identification, target resource node and frequency control threshold parameter associated in the interface access restriction parameter group, construct the permission configuration record, structure package coding of the task permission level field, interface calling restriction field and scheduling period index field, and write into the scheduling permission mapping area, bind the permission identification to the task scheduling behavior, and obtain the scheduling task permission configuration group.

7. The multi-task scheduling method based on trusted computing according to claim 1, characterized in that, The specific steps of S5 are: S501: Call the task scheduling path structure identified as the permission adjustment state in the scheduling task permission configuration group, extract the calling interface sequence and target resource node number of the current task, combine the behavior template path structure of the resource node in the trusted resource trajectory encapsulation set, perform one-to-one mapping of the interface sequence position index, and perform Boolean judgment according to whether the task path segment is continuous with the template path segment to obtain the path continuity verification tag set; S502: Based on the continuous mark path segment in the path continuity verification tag set, extract the task scheduling position number field, binding control parameter set field and permission configuration parameter field in the corresponding path segment, aggregate each type of field into a unit block structure according to the path segment order, reorganize the data according to the task identification number, integrate the path number and permission control element into a combined structure set, and generate a scheduling path binding structure frame; S503: According to each path segment combined structure in the scheduling path binding structure frame, construct a structured scheduling tag field block including task identification, path number and permission configuration data, write the field block into the corresponding task slot of the scheduling queue metadata allocation area, establish a scheduling path binding index structure for the task scheduling record, and obtain a trusted path binding tag set.

8. A multi-task scheduling system based on trusted computing, characterized in that, The system is used to implement the multi-task scheduling method based on trusted computing according to any one of claims 1-7, and the system comprises: A resource trajectory identification module obtains a computing resource node identified by a trusted authentication unit, generates a behavior trajectory segment arranged by time for each type of resource, integrates the behavior trajectory segment and the trusted identification value of the node, and obtains a trusted resource trajectory encapsulation set; The access intention analysis module obtains the to-be-scheduled task identifier of the submitted resource binding application, screens high-frequency calling items according to the calling interface priority set by the scheduling control unit, combines the access logic of the task into a structured path description unit, and obtains a task access intention track segment; The path offset comparison module performs difference comparison on the calling interval time and the cycle time in the behavior block based on the track segment of the corresponding resource node in the task access intention track segment and the trusted resource track encapsulation set, performs path comparison according to the interface sequence and the resource behavior sequence, counts the consistency rate and gives a credibility mark, and obtains a behavior path offset comparison result; The task permission analysis module calls the credibility mark field in the behavior path offset comparison result, sets the access interface in the current scheduling period as a minimum permission interface set for the task with an abnormal behavior mark, and sets a maximum calling frequency limit value, to obtain a scheduling task permission configuration group; The path binding processing module calls the scheduling path structure of the permission adjustment task in the scheduling task permission configuration group, verifies the path continuity, packs the task scheduling position number, the control parameter set and the permission configuration content included in the continuous path segment, constructs a path binding label, writes the path binding label into the scheduling queue metadata, and obtains a trusted path binding label set. 9.A computer device, comprising a memory and a processor, and characterized in that, The memory stores a computer program, and the processor executes the computer program to realize the multi-task scheduling system based on trusted computing in claim 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the multi-task scheduling method based on trusted computing in any one of claims 1-7.

Images