Cloud computing system based on distributed multi-task request

Abstract

The invention relates to the field of multi-task request cloud computing system design, in particular to a distributed multi-task request-based cloud computing system, which is provided with a task statistics module for storing historical task data executed by a user side and task data newly created by the user side; the task analysis module is used for calling the data stored by the task statistics module, analyzing jump information among the tasks to obtain jump sensitive characteristics, and analyzing jump response characterization parameters according to the jump sensitive characteristics to identify the jump response timeliness degree of the user side; and the resource allocation module is respectively connected with the task statistics module and the task analysis module, and is used for adaptively carrying out resource allocation on the user side according to the timeliness degree of the jump response, so that the timely response of the jump between the tasks and the normal execution of each task process are ensured on the premise of reasonably allocating the computing resources.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing system design for multi-task requests, and in particular to a cloud computing system based on distributed multi-task requests. Background Art

[0002] Due to the combined effects of factors such as the growth of computing demand, the uneven distribution of computing resources, and the need for load balancing, distributed cloud computing is adopted and appropriate resource allocation methods are selected to improve system performance and efficiency by comprehensively considering task characteristics, system status, and resource limitations. Therefore, the allocation of cloud computing resources is crucial to ensuring efficient task execution and system performance optimization.

[0003] Chinese patent publication number: CN117370032A, discloses a method for optimizing resource allocation of a cloud computing server, including: collecting usage data of various resources of a virtual machine, obtaining an expected value of the demand for various resources of the virtual machine according to the average value of all usage data of various resources during the working hours of the virtual machine, correcting the expected value according to the first demand of the usage data of various resources, and obtaining the corrected expected value of the demand for various resources; estimating the resource utilization rate of the server according to the corrected expected value of the demand for various resources of all enabled virtual machines in the server and the total amount of various resources of the server, and allocating virtual machine resources according to the estimated resource utilization rate of the server and the corrected expected value of the demand for various resources of the virtual machine.

[0004] However, there are still the following problems in the prior art:

[0005] When jumping between tasks, since the user side executes multiple tasks or processes in parallel, there may be insufficient computing resources on the user side, which may easily lead to abnormal situations such as incomplete storage of the current task content, too long response time required to jump from the current task to the next task, and too slow execution of the next task. Summary of the invention

[0006] To this end, the present invention provides a cloud computing system based on distributed multi-task requests to overcome the problem in the prior art that when jumping between tasks occurs, since the user end is in a mode of executing multiple tasks or processes in parallel, there may be a situation where the user end's current computing resources are insufficient, which may easily lead to abnormal situations such as incomplete storage of the current task content, too long response time required for the current task to jump to the next task, and too slow execution of the next task.

[0007] To achieve the above object, the present invention provides a cloud computing system based on distributed multi-task request, which includes:

[0008] The task statistics module is used to store the historical task data executed by the user end and the task data newly created by the user end;

[0009] A task analysis module, which is connected to the task statistics module, is used to call the data stored in the task statistics module, analyze the jump information between the tasks, determine other tasks that have a jump relationship with the newly created task of the user end, so as to obtain jump sensitivity features, and analyze jump response characterization parameters according to the jump sensitivity features to identify the timeliness of the jump response of the user end, wherein the jump sensitivity features include the response time required for the task jump and the CPU load of the user end;

[0010] The resource allocation module is connected to the task statistics module and the task analysis module respectively, and is used to allocate resources to the user terminal according to the timeliness of the jump response, including:

[0011] Improve the number of service nodes allocated to the user end, determine the newly created task of the user end, analyze the task execution degree characterization parameter based on the user end response characteristics of the task jump operation in the historical task data stored in the task statistics module, determine whether the current resources meet the allocation standard according to the task execution degree characterization parameter, and adjust the number of service nodes allocated to the user end based on the task execution degree characterization parameter;

[0012] Or, adjusting the number of service nodes allocated to the user terminal according to the jump response characterization parameter;

[0013] The user-side response characteristics include the preservation completeness of the task content and the execution time of the jump task.

[0014] Furthermore, the task parsing module is used to analyze the jump information between tasks to obtain jump sensitive features including:

[0015] To determine the jump instructions issued by each task;

[0016] to confirm other tasks having a jump relationship with the task according to the jump instruction of the task;

[0017] It is used to extract the average response time and average CPU load required for task jump when the task jumps to each task with a jump relationship.

[0018] Furthermore, the task parsing module is used to parse jump response characterization parameters according to jump sensitivity features, including:

[0019]

[0020] In formula (1), M represents the jump response characterization parameter, T represents the average response time required for task jump, T0 represents the response time threshold required for task jump, P represents the average CPU load on the user side after executing the jump, P0 represents the CPU load threshold on the user side, α represents the response time weight coefficient required for jump, and β represents the CPU load weight coefficient.

[0021] Furthermore, the task analysis module is used to identify the timeliness of the jump response of the user end, including:

[0022] If the jump response characterization parameter is greater than or equal to the jump response characterization parameter threshold, the jump response of the user terminal is determined to be of a weak timeliness category;

[0023] If the jump response characterization parameter is less than the jump response characterization parameter threshold, the jump response of the user terminal is determined to be of a strong timeliness category.

[0024] Furthermore, the resource allocation module is used to allocate resources to the user terminal according to the timeliness of the jump response, including:

[0025] If the jump response of the user end is of the weak timeliness category, the number of service nodes allocated to the user end is increased, the newly created task of the user end is determined, and the response characteristics of the user end for the jump operation of the task in the historical task data stored in the task statistics module are analyzed, and the task execution degree characterization parameter is determined according to the task execution degree characterization parameter. Whether the current resource meets the allocation standard, the number of service nodes allocated to the user end is adjusted based on the task execution degree characterization parameter;

[0026] If the jump response of the user terminal is of a strong timeliness category, the number of service nodes allocated to the user terminal is adjusted according to the jump response characterization parameter.

[0027] Furthermore, the resource allocation module is used to analyze the parameters representing the degree of task execution, including:

[0028] Used to determine the newly created tasks on the user side;

[0029] To call the task stack information before the task jump is completed and the task stack information after the task jump is completed in the task statistics module, and calculate the overlap of the task stack information;

[0030] To confirm the overlap as the preservation completeness of the task content;

[0031] The preservation completeness of the task content is used as the first feature;

[0032] The ratio of the execution time threshold to the execution time of the task is calculated as the second feature;

[0033] The sum of the first feature and the second feature is used to determine the task execution degree characterization parameter.

[0034] Furthermore, the resource allocation module is used to determine whether the current resources meet the allocation criteria, including:

[0035] If the task execution degree characterization parameter is greater than or equal to the task execution degree characterization parameter threshold, it is determined that the current resource meets the allocation criteria;

[0036] If the task execution degree characterization parameter is less than the task execution degree characterization parameter threshold, it is determined that the current resources do not meet the allocation criteria.

[0037] Furthermore, the resource allocation module is used to determine whether to adjust the number of service nodes allocated to the user end, including:

[0038] If the current resources do not meet the allocation criteria, the number of service nodes allocated to the user end is adjusted based on the task execution degree characterization parameter.

[0039] Furthermore, the resource allocation module is used to adjust the number of service nodes allocated to the user terminal based on the task execution degree characterization parameter.

[0040] The number of service node allocations is increased, and the increase in the number of allocations is negatively correlated with the parameter characterizing the degree of task execution.

[0041] Furthermore, the resource allocation module is used to adjust the number of service nodes allocated to the user terminal according to the jump response characterization parameter, including:

[0042] The number of service node allocations is increased, and the increase in the number of allocations is positively correlated with the jump response characterization parameter.

[0043] Compared with the prior art, the present invention sets up a task statistics module for storing historical task data executed by the user end and newly created task data of the user end; a task analysis module for calling the data stored in the task statistics module, analyzing the jump information between tasks to obtain jump sensitivity features, and analyzing jump response characterization parameters based on the jump sensitivity features to identify the timeliness of the jump response of the user end; a resource allocation module, which is respectively connected to the task statistics module and the task analysis module, for adaptively allocating resources to the user end according to the timeliness of the jump response, and then, under the premise of reasonable allocation of computing resources, ensuring the timely response of jumps between tasks and the normal execution of various task processes.

[0044] In particular, the present invention parses jump response characterization parameters based on jump sensitive features. During the task jump process, the current existing resources of the user end need to meet the different execution processes of each task. If the existing resources are insufficient, it may affect the process and execution efficiency of tasks that are not allocated resources. For example, the response time required for the process from the end of the current task to the start of the execution process of the next task to be jumped is too long, and the CPU load of the user end is too high after the jump instruction is issued. Therefore, the present application calculates the jump response characterization parameters through the response time required for the task jump and the CPU load to characterize the response degree of the user end after executing the jump instruction, and provides data support for the subsequent identification of the timeliness of the jump response of the user end, so as to adaptively allocate resources to the user end, and then, under the premise of reasonable allocation of computing resources, ensure the timely response of the jump between tasks and the normal execution of various task processes.

[0045] In particular, the present invention pre-increases the number of service nodes allocated to the user side to meet the normal execution and jump response of several existing tasks in the task statistics module, and further finely analyzes the response status and resource demand level of the user side after creating a new task on the premise of ensuring the stability of the current user side's execution of several tasks.

[0046] In particular, the present invention identifies the timeliness of the jump response of the user end. For example, the jump response of the user end is a weak timeliness category, which refers to the situation that the user end responds slowly to the jump operation of any task and the subsequent jump task execution process is abnormal, reducing the task execution efficiency. Therefore, the present application determines the correlation between the newly created task of the user end and other tasks that may potentially perform jump operations through the historical task data stored in the task statistics module, estimates the response characteristics presented by the user end when the jump operation potentially occurs, and analyzes the task execution degree characterization parameter to characterize the completeness of the preservation of the current task content and the execution degree of the next task to be jumped, and then determines whether the current resources of the user end meet the allocation standards and whether they can meet the realization of different execution processes of the current task and the next task to be jumped, and adjusts the number of service nodes of the user end by calculating the task execution degree characterization parameter based on the preservation completeness of the current task content and the execution time of the jump task, and then, under the premise of reasonable allocation of computing resources, avoids slow task execution and system overload, and ensures the timely response of the jump between tasks and the normal execution of various task processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A functional module diagram of a cloud computing system based on distributed multi-task requests according to an embodiment of the invention;

[0048] Figure 2 A logic decision diagram for identifying the timeliness of a jump response of a user terminal in an embodiment of the invention;

[0049] Figure 3 A logical decision diagram for determining whether the current resources meet the allocation criteria for the embodiment of the invention

[0050] Figure 4 A logical decision diagram for determining whether to adjust the number of service nodes allocated to a user terminal according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0053] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0054] See also Figures 1 to 4 As shown, Figure 1 FIG. 1 is a functional module diagram of a cloud computing system based on distributed multi-task requests according to an embodiment of the present invention. Figure 2 A logical decision diagram for identifying the timeliness of a jump response of a user terminal in an embodiment of the invention, Figure 3 A logical decision diagram for determining whether the current resources meet the allocation criteria for the embodiment of the invention, Figure 4 The logical decision diagram for determining whether to adjust the number of service nodes allocated to the user end according to the embodiment of the invention is as follows. The cloud computing system based on distributed multi-task request according to the embodiment of the invention includes:

[0055] The task statistics module is used to store the historical task data executed by the user end and the task data newly created by the user end;

[0056] A task analysis module, which is connected to the task statistics module, is used to call the data stored in the task statistics module, analyze the jump information between the tasks, determine other tasks that have a jump relationship with the newly created task of the user end, so as to obtain jump sensitivity features, and analyze jump response characterization parameters according to the jump sensitivity features to identify the timeliness of the jump response of the user end, wherein the jump sensitivity features include the response time required for the task jump and the CPU load of the user end;

[0057] The resource allocation module is connected to the task statistics module and the task analysis module respectively, and is used to allocate resources to the user terminal according to the timeliness of the jump response, including:

[0058] Improve the number of service nodes allocated to the user end, determine the newly created task of the user end, analyze the task execution degree characterization parameter based on the user end response characteristics of the task jump operation in the historical task data stored in the task statistics module, determine whether the current resources meet the allocation standard according to the task execution degree characterization parameter, and adjust the number of service nodes allocated to the user end based on the task execution degree characterization parameter;

[0059] Or, adjusting the number of service nodes allocated to the user terminal according to the jump response characterization parameter;

[0060] The user-side response characteristics include the preservation completeness of the task content and the execution time of the jump task.

[0061] Specifically, there is no limitation on the specific structure of the task statistics module, which may be a database for storing information, and the historical task data may be stored in the database, which will not be elaborated here.

[0062] Specifically, there is no limitation on the task statistics module, task analysis module, resource allocation module and the specific structure of the module, which itself or each unit therein can be composed of logical components or a combination of logical components, and the logical components include a field programmable processor, a computer or a microprocessor in a computer.

[0063] Specifically, the task parsing module is used to analyze the jump information between tasks to obtain jump sensitive features including:

[0064] To determine the jump instructions issued by each task;

[0065] to confirm other tasks having a jump relationship with the task according to the jump instruction of the task;

[0066] It is used to extract the average response time and average CPU load required for task jump when the task jumps to each task with a jump relationship.

[0067] It is understandable that if a jump occurs between two tasks, there is a jump relationship between the two tasks.

[0068] Specifically, the task parsing module is used to parse jump response characterization parameters according to jump sensitivity features, including:

[0069]

[0070] In formula (1), M represents the jump response characterization parameter, T represents the average response time required for task jump, T0 represents the response time threshold required for task jump, P represents the average CPU load on the user side after executing the jump, P0 represents the CPU load threshold on the user side, α represents the response time weight coefficient required for jump, and β represents the CPU load weight coefficient.

[0071] The response time required for task jump is the response time from the end of the current task to the start of the execution of the jumped task.

[0072] In this embodiment, the process from the end of the current task process to the start of the execution of the next task to be jumped is regarded as the jump response process. The response time threshold T0 required for task jump and the user-side CPU load threshold P0 are obtained in advance. The historical task data executed by the user side are obtained, and the jump instruction data issued between several tasks are recorded. After the jump instruction is executed, the response time data required for task jump and the CPU load data of the user side are combined, and the average response time ΔT required for jumps between several tasks and the average CPU load ΔP of the user side are solved. It is set that T0=r1×ΔT, P0=r2×ΔP, r1 is the first precision coefficient, r2 is the second precision coefficient, 1.05<r1<1.15, 0.95<r1<1.05, α is 0.55, and β is 0.45.

[0073] The present invention parses jump response characterization parameters based on jump sensitive features. During the task jump process, the current existing resources of the user end need to meet the different execution processes of each task. If the existing resources are insufficient, it may affect the process and execution efficiency of tasks that are not allocated resources. For example, the response time required for the process from the end of the current task to the start of the execution process of the next task to be jumped is too long, and the CPU load of the user end is too high after the jump instruction is issued. Therefore, the present application calculates the jump response characterization parameters through the response time required for the task jump and the CPU load to characterize the response degree of the user end after executing the jump instruction, and provides data support for the subsequent identification of the timeliness of the jump response of the user end, so as to adaptively allocate resources to the user end, and then, under the premise of reasonable allocation of computing resources, ensure the timely response of the jump between tasks and the normal execution of various task processes.

[0074] Specifically, the task analysis module is used to identify the timeliness of the jump response of the user end, including:

[0075] If the jump response characterization parameter is greater than or equal to the jump response characterization parameter threshold, the jump response of the user terminal is determined to be of a weak timeliness category;

[0076] If the jump response characterization parameter is less than the jump response characterization parameter threshold, the jump response of the user terminal is determined to be of a strong timeliness category.

[0077] The jump response characterization parameter M0 is selected in the interval [1.21,1.36].

[0078] Specifically, the resource allocation module is used to allocate resources to the user terminal according to the timeliness of the jump response, including:

[0079] If the jump response of the user end is of the weak timeliness category, the number of service nodes allocated to the user end is increased, the newly created task of the user end is determined, and the response characteristics of the user end for the jump operation of the task in the historical task data stored in the task statistics module are analyzed, and the task execution degree characterization parameter is determined according to the task execution degree characterization parameter. Whether the current resource meets the allocation standard, the number of service nodes allocated to the user end is adjusted based on the task execution degree characterization parameter;

[0080] If the jump response of the user terminal is of a strong timeliness category, the number of service nodes allocated to the user terminal is adjusted according to the jump response characterization parameter.

[0081] The present invention increases the number of service nodes allocated to the user end in advance to meet the normal execution and jump response of several existing tasks in the task statistics module, and further finely analyzes the response status and resource demand degree of the user end after the new task is created, while ensuring the stability of the execution of several tasks by the current user end.

[0082] Specifically, the resource allocation module is used to analyze the task execution degree characterization parameters, including:

[0083] Used to determine the newly created tasks on the user side;

[0084] To call the task stack information before the task jump is completed and the task stack information after the task jump is completed in the task statistics module, and calculate the overlap of the task stack information;

[0085] To confirm the overlap as the preservation completeness of the task content;

[0086] The preservation completeness of the task content is used as the first feature;

[0087] The ratio of the execution time threshold to the execution time of the task is calculated as the second feature;

[0088] The sum of the first feature and the second feature is used to determine the task execution degree characterization parameter.

[0089] It can be understood that there are several records of task completion jump in the historical task data, so the overlap is the average overlap of the task stack information before and after the task completion jump, and the execution time is the average execution time of the several tasks;

[0090] In this embodiment, the task execution program characterization parameter is determined according to the following formula:

[0091]

[0092] Where C represents the parameter characterizing the degree of task execution, W represents the preservation completeness of the current task content, X0 represents the execution time threshold, and X represents the execution time of the jump task.

[0093] The execution time threshold X0 is obtained from the pre-measurement point. The execution time data of several tasks after jumping are obtained, and the average execution time ΔX is solved. It is set that X0=z×ΔX, where z is the offset coefficient, 1.05<r1<1.15.

[0094] The execution time is the time taken from the start of the jump task to the completion of the execution after the jump is completed.

[0095] Specifically, the information covered by the task itself includes stack information, memory mapping, and task status. The stack information saves information such as local variables, function parameters, and return addresses. The stack information content before the current task jumps stored in the task statistics module is compared with the stack information content after the current task jumps. The overlap between the two versions of the stack information content is identified to determine the completeness of the current task content.

[0096] Specifically, during the task jump process, saving the current task content is a key process to ensure that the task can maintain its state and the integrity of the task content when it is resumed later, and ensure that the tasks can be smoothly connected and continued.

[0097] Specifically, the resource allocation module is used to determine whether the current resources meet the allocation criteria, including:

[0098] If the task execution degree characterization parameter is greater than or equal to the task execution degree characterization parameter threshold, it is determined that the current resources meet the allocation criteria.

[0099] If the task execution degree characterization parameter is less than the task execution degree characterization parameter threshold, it is determined that the current resources do not meet the allocation criteria.

[0100] The task execution program characterization parameter C0 is selected in the interval [1.83,1.94].

[0101] Specifically, the resource allocation module is used to determine whether to adjust the number of service nodes allocated to the user terminal, including:

[0102] If the current resources do not meet the allocation criteria, the number of service nodes allocated to the user end is adjusted based on the task execution degree characterization parameter.

[0103] Specifically, the resource allocation module is used to adjust the number of service nodes allocated to the user terminal based on the task execution degree characterization parameter.

[0104] The number of service node allocations is increased, and the increase in the number of allocations is negatively correlated with the parameter characterizing the degree of task execution.

[0105] In this embodiment, optionally,

[0106] The task execution degree characterization parameter C is compared with the first task execution degree characterization parameter comparison threshold C1 and the second task execution degree characterization parameter comparison threshold C2,

[0107] If C>C2, the allocation quantity increase amount is determined to be the first allocation quantity increase amount a1, and a1=[0.27a0] is set;

[0108] If C1≤C≤C2, the allocation quantity increase amount is determined to be the second allocation quantity increase amount a2, and a2=[0.33a0] is set;

[0109] If C<C1, the allocation quantity increase amount is determined to be the third allocation quantity increase amount a3, and a3=[0.41a0] is set;

[0110] Wherein, a0 represents the number of initial service node allocations, C1=0.7C0, C2=0.9C0.

[0111] Regarding the initial number of service nodes to be allocated, those skilled in the art may determine it based on the performance indicators of the user end, hardware resources, cloud service quotas, and task requirements, which will not be elaborated herein.

[0112] The present invention identifies the timeliness of the jump response of the user end. For example, the jump response of the user end is of the weak timeliness category, which refers to the situation that the user end responds slowly to the jump operation of any task and the subsequent jump task execution process is abnormal, which reduces the task execution efficiency. Therefore, the present application determines the correlation between the newly created task of the user end and other tasks that may potentially perform jump operations through the historical task data stored in the task statistics module, estimates the response characteristics presented by the user end when the jump operation potentially occurs, and analyzes the task execution degree characterization parameter to characterize the completeness of the preservation of the current task content and the execution degree of the next task to be jumped, and then determines whether the current resources of the user end meet the allocation standards and whether they can meet the realization of different execution processes of the current task and the next task to be jumped, and adjusts the number of service nodes of the user end by calculating the task execution degree characterization parameter based on the preservation completeness of the current task content and the execution time of the jump task, and then, under the premise of reasonable allocation of computing resources, avoids slow task execution and system overload, and ensures the timely response of the jump between tasks and the normal execution of various task processes.

[0113] Specifically, the resource allocation module is used to adjust the number of service nodes allocated to the user terminal according to the jump response characterization parameter, including:

[0114] The number of service node allocations is increased, and the increase in the number of allocations is positively correlated with the jump response characterization parameter.

[0115] In this embodiment, optionally,

[0116] The jump response characterization parameter M is compared with the first jump response characterization parameter comparison threshold M1 and the second jump response characterization parameter comparison threshold M2,

[0117] If M>M2, the allocation quantity increase amount is determined to be the first allocation quantity increase amount a1, and a1=[0.43a0] is set;

[0118] If M1≤M≤M2, the allocation quantity increase amount is determined to be the second allocation quantity increase amount a2, and a2=[0.36a0] is set;

[0119] If M<M1, the allocation quantity increase amount is determined to be the third allocation quantity increase amount a3, and a3=[0.28a0] is set;

[0120] Wherein, a0 represents the number of initial service node allocations, M1=0.75M0, M2=0.85M0.

[0121] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A cloud computing system based on distributed multi-task requests, characterized in that: include: The task statistics module is used to store the historical task data executed by the user end and the task data newly created by the user end; A task analysis module, which is connected to the task statistics module, is used to call the data stored in the task statistics module, analyze the jump information between the tasks, determine other tasks that have a jump relationship with the newly created task of the user end, so as to obtain jump sensitivity features, and analyze jump response characterization parameters according to the jump sensitivity features to identify the timeliness of the jump response of the user end, wherein the jump sensitivity features include the response time required for the task jump and the CPU load of the user end; A resource allocation module, which is connected to the task statistics module and the task analysis module respectively, is used to allocate resources to the user terminal according to the timeliness of the jump response. include, Improve the number of service nodes allocated to the user end, determine the newly created task of the user end, analyze the task execution degree characterization parameter based on the user end response characteristics of the task jump operation in the historical task data stored in the task statistics module, determine whether the current resources meet the allocation standard according to the task execution degree characterization parameter, and adjust the number of service nodes allocated to the user end based on the task execution degree characterization parameter; Or, adjusting the number of service nodes allocated to the user terminal according to the jump response characterization parameter; The user-side response characteristics include the preservation completeness of the task content and the execution time of the jump task.

2. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The task parsing module is used to analyze the jump information between tasks to obtain jump sensitive features including: Used to determine the jump instructions issued by each task; to confirm other tasks having a jump relationship with the task according to the jump instruction of the task; It is used to extract the average response time and average CPU load required for task jump when the task jumps to each task with a jump relationship.

3. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The task parsing module is used to parse jump response characterization parameters according to jump sensitivity features, including: In formula (1), M represents the jump response characterization parameter, T represents the average response time required for task jump, T0 represents the response time threshold required for task jump, P represents the average CPU load on the user side after executing the jump, P0 represents the CPU load threshold on the user side, α represents the response time weight coefficient required for jump, and β represents the CPU load weight coefficient.

4. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The task analysis module is used to identify the timeliness of the jump response of the user end, including: If the jump response characterization parameter is greater than or equal to the jump response characterization parameter threshold, the jump response of the user terminal is determined to be of a weak timeliness category; If the jump response characterization parameter is less than the jump response characterization parameter threshold, the jump response of the user terminal is determined to be of a strong timeliness category.

5. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The resource allocation module is used to allocate resources to the user terminal according to the timeliness of the jump response, including: If the jump response of the user end is of the weak timeliness category, the number of service nodes allocated to the user end is increased, the newly created task of the user end is determined, and the response characteristics of the user end for the jump operation of the task in the historical task data stored in the task statistics module are analyzed, and the task execution degree characterization parameter is determined according to the task execution degree characterization parameter. Whether the current resource meets the allocation standard, the number of service nodes allocated to the user end is adjusted based on the task execution degree characterization parameter; If the jump response of the user terminal is of a strong timeliness category, the number of service nodes allocated to the user terminal is adjusted according to the jump response characterization parameter.

6. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The resource allocation module is used to analyze the task execution degree characterization parameters, including: Used to determine the newly created tasks on the user side; To call the task stack information before the task jump is completed and the task stack information after the task jump is completed in the task statistics module, and calculate the overlap of the task stack information; To confirm the overlap as the preservation completeness of the task content; The preservation completeness of the task content is used as the first feature; The ratio of the execution time threshold to the execution time of the task is calculated as the second feature; The sum of the first feature and the second feature is used to determine the task execution degree characterization parameter.

7. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The resource allocation module is used to determine whether the current resources meet the allocation criteria, including: If the task execution degree characterization parameter is greater than or equal to the task execution degree characterization parameter threshold, it is determined that the current resource meets the allocation criteria; If the task execution degree characterization parameter is less than the task execution degree characterization parameter threshold, it is determined that the current resources do not meet the allocation criteria.

8. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The resource allocation module is used to determine whether to adjust the number of service nodes allocated to the user end, including: If the current resources do not meet the allocation criteria, the number of service nodes allocated to the user end is adjusted based on the task execution degree characterization parameter.

9. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The resource allocation module is used to adjust the number of service nodes allocated to the user terminal based on the task execution degree characterization parameter, The number of service node allocations is increased, and the increase in the number of allocations is negatively correlated with the parameter characterizing the degree of task execution.

10. The cloud computing system based on distributed multi-task request according to claim 1, characterized in that: The resource allocation module is used to adjust the number of service nodes allocated to the user terminal according to the jump response characterization parameter, including: The number of service node allocations is increased, and the increase in the number of allocations is positively correlated with the jump response characterization parameter.

Images