Multi-task coordination parallel service orchestration method based on dynamic integration

Through the dynamically integrated multi-task coordinated parallel service orchestration method, the problem of insufficient interconnection and interoperability in the software system is solved, multi-task parallel execution and resource optimization are realized, and task execution efficiency and system performance are improved.

CN118963933APending Publication Date: 2024-11-15INST OF WAR STUDIES ACAD OF MILITARY SCI OF THE CHINESE PEOPLES LIBERATION ARMY
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410935844.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing technology lacks software interconnection, interoperability and interoperability capabilities in software systems, resulting in weak software integration capabilities, high development costs, product quality risks and maintenance costs, and the interface matching between intelligent models is difficult to ensure, resulting in tasks. Inefficient execution and waste of resources.

Method used

Multi-task coordinated parallel service orchestration method based on dynamic integration is adopted to achieve multi-task parallel execution by setting up a cloud service platform, generating task execution processes, establishing task resource pools and adjusting pools, and setting task dependencies and priorities.

Benefits of technology

It improves the efficiency of multi-task execution, optimizes the utilization of equipment computing resources, reduces repetitive labor and resource waste, and improves the overall performance and reliability of the software system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963933A_ABST
    Figure CN118963933A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-task coordination parallel service orchestration method based on dynamic integration, relates to the technical field of task scheduling, and effectively improves the utilization efficiency of computing resources of equipment. The method comprises the steps of generating a corresponding task execution process according to a task data demand and a task data interaction protocol, setting a dependency relationship between task steps in the task execution process, establishing a task resource pool and a task adjustment pool, inputting the task execution process into task resources to generate a plurality of task request elements, and executing the task execution process according to the task request elements. The method comprises the following steps of: acquiring task request elements, marking required calculation resource quantity or storage resource quantity for each task request element, inputting the task request elements of different task execution processes into a task resource pool, and setting priorities for each task request element by the task resource pool; and executing a plurality of task execution flows in parallel according to the priority of each task request element and the required computing resource or storage resource.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of task scheduling, and in particular to a multi-task coordinated parallel service arrangement method based on dynamic integration. Background Art

[0002] In large-scale software systems, the systematic R&D requirements are relatively complex, and the interconnection, intercommunication and interoperability between software products are required to be high. Data format standards and information exchange standards are not unified, and their generalization and serialization design levels are low, resulting in insufficient software interconnection, intercommunication and interoperability; software in various fields has different forms and styles and is difficult to integrate quickly, which cannot meet the requirements of on-demand access and rapid deployment, resulting in weak software integration capabilities; software products are developed according to individual projects, lack of serialization design concepts, diversified user needs, high development investment costs, and poor product maturity and stability, resulting in increased quality risks and maintenance costs for military products. Military software products have many defects, poor reliability, difficult maintenance, slow response to demand changes, and difficult to control software development progress, which have become bottlenecks in the development of large-scale software systems.

[0003] Existing technologies usually require many related intelligent models to be combined and integrated as sub-modules. Currently, this process is usually completed at the code level, and the integration results are usually put into use as an independent product.

[0004] The problem with the existing technology is that with the continuous development of technology, the matching of interfaces between intelligent models as submodules is difficult to ensure, resulting in certain difficulties in the process of code writing. The variety of tasks and task objectives has led to an exponential increase in the number of combinations of intelligent models, resulting in problems such as duplication of work and waste of resources.

[0005] Therefore, how to improve the task execution efficiency when executing multiple tasks while improving the utilization efficiency of device computing resources is a difficulty in the existing technology. For this purpose, a multi-task coordinated parallel service orchestration method based on dynamic integration is provided. Summary of the invention

[0006] In order to solve the above technical problems, the purpose of the present invention is to provide a multi-task coordinated parallel service orchestration method based on dynamic integration.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] The multi-task coordinated parallel service orchestration method based on dynamic integration includes the following steps:

[0009] S1. Setting up a cloud service platform, wherein the cloud service platform is provided with a client and a task management terminal, and then the client generates a task requirement and sends it to the task management terminal;

[0010] S2, generate task execution process according to task requirements;

[0011] S3, setting dependencies between various task steps in the task execution process;

[0012] S4. Establish a task resource pool and a task adjustment pool, input the task execution process into the task resources to generate a number of task request elements, and then mark the required computing resources or storage resources for each task request element;

[0013] S5. Install the task adjustment pool on the client, and input the task request elements of different task execution processes into the task resource pool. After the task resource pool sets the priority for each task request element, the client executes multiple task execution processes in parallel according to the priority of each task request element and the required amount of computing resources or storage resources.

[0014] Furthermore, the cloud service platform is provided with a client and a task management terminal;

[0015] The client is used to generate task requirements and execute the task execution process from the task management end;

[0016] The task needs to include a data exchange protocol and current client computer resources, wherein the data exchange protocol includes the data type and format requirements that the client can receive, the parameters required for the task, and the relationship and requirements between the parameters;

[0017] The task management end is used to establish a task execution process and a task request pool according to task requirements, and a task resource pool based on current client computer resources, and then adjust the steps of multiple task execution processes of the same client, and send the adjusted task execution process to the client.

[0018] Furthermore, the process of generating a task execution flow according to task requirements includes:

[0019] The task management terminal establishes a task data input step according to the data exchange protocol, wherein the task data input step is provided with a data content check node and a data format check node;

[0020] At the same time, corresponding task steps are established according to the corresponding relationship and requirements between various parameters, and the task steps include operations such as data format conversion, data cleaning, and data analysis;

[0021] Then, the task management end uses the task data requirement as the starting point of the task execution process, connects each task step in sequence to obtain the task execution process, and sets a task data output step at the end of the task execution process, wherein the data step is provided with a data format check node that is the same as the task data output.

[0022] Furthermore, the dependency relationship includes:

[0023] Activation dependency: For two interrelated task steps s1 and s2 in the task execution process, when there is an activation dependency between task steps s1 and s2, when task step s1 is executed, task step s2 must be executed, which can be expressed as s1∈ActCond(s2);

[0024] Alternative dependency: When task steps s1, s2, ..., s n and task steps m When there is an alternative dependency between the task steps s1, s3, ..., s n When any task step is executed, task step s m Execute immediately, which can be expressed as s n ∈AltCond(s m ), where n and m are natural numbers greater than 0, and n is not equal to m;

[0025] Abort dependency: When there is an abort dependency between task step s1 and task step s2, it means that when task step s1 is aborted, task step s2 is also aborted synchronously, which can be expressed as s1∈AbtCond(s2);

[0026] Recovery dependency: When there is a recovery dependency between task step s1 and task step s2, it means that when task step s1 is reset, task step s2 is reset synchronously, which can be expressed as s1∈CpsCond(s2);

[0027] Cancellation dependency: When there is a cancellation dependency between task step s1 and task step s2, it means that when task step s1 is canceled, task step s2 is canceled synchronously, which can be expressed as s1∈CnlCond(s2);

[0028] Then, the task management end sets corresponding dependencies between the task steps according to the relationships between the parameters represented in the data exchange protocol.

[0029] Further, the task management terminal establishes and sets a task resource pool and a task adjustment pool according to the current client computer resources of the client in the task requirements, and installs the task adjustment pool in the corresponding client;

[0030] The task resource pool includes the current computing resources and storage resources of the corresponding client, and then the task management end generates corresponding task request elements according to each task step in the task execution process.

[0031] Furthermore, the process of marking the amount of computing resources and storage resources for each task request element includes:

[0032] The task request elements of different task execution processes are input into the task resource pool, and then the task resource pool operates according to the parameters and parameters required by the task request elements to obtain the computing resources and storage resources required by the corresponding task request elements, and the computing resources and storage resources are marked on the corresponding task request elements;

[0033] When the computing resource amounts and storage resource amounts of all task request elements are marked, the task resource pool sends all task request elements to the task adjustment pool in the client.

[0034] Furthermore, the process of setting the priority of each task request element includes:

[0035] The task adjustment pool is provided with a priority adjustment mechanism and a resource usage queue. The priority adjustment mechanism includes: absolute preemptive, queue preemptive, sequential preemptive and idle preemptive;

[0036] Then, the task adjustment pool sets the priority of each task request element in each task adjustment pool according to the dependency relationship between the task steps corresponding to each task request element in the task execution process, that is, when the dependency relationship between two task steps is an activation dependency, the priority of the latter task step is set to absolute preemptive;

[0037] When the dependency between two task steps is an alternative dependency, the priority of the latter task step is set to idle preemptive;

[0038] When the dependency between two task steps is a recovery dependency, the priority of the latter task step is set to queue preemptive;

[0039] When the dependency between two task steps is abort dependency or cancel dependency, the priority of the latter task step is set to sequential preemptive.

[0040] Furthermore, the process of the client executing the task execution flow includes:

[0041] When multiple task execution processes are executed in parallel on the client, each time the client executes a new task step in any task execution process, the corresponding task request element in the task adjustment pool is activated and arranged in the resource usage queue according to its priority;

[0042] When the current idle computing resources or storage resources of the client meet the task request element at the head of the resource usage queue, the client executes the corresponding task steps and repeats the above operations until the entire task execution process is completed.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention generates a corresponding task execution process according to task data requirements and a task data interaction protocol, sets dependencies between various task steps in the task execution process, inputs the task execution process into task resources to generate a plurality of task request elements, and then labels each task request element with required computing resources or storage resources, installs a task adjustment pool on a client, and inputs task request elements of different task execution processes into a task resource pool. After the task resource pool sets priorities for each task request element, multiple task execution processes are executed in parallel according to the priorities of each task request element and the required computing resources or storage resources, thereby improving the task execution efficiency when multiple tasks are executed and improving the utilization efficiency of device computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention.

[0046] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0047] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0048] Embodiment 1,

[0049] like Figure 1 As shown, the multi-task coordinated parallel service orchestration method based on dynamic integration includes the following steps:

[0050] S1. Setting up a cloud service platform, wherein the cloud service platform is provided with a client and a task management terminal, and then the client generates a task requirement and sends it to the task management terminal;

[0051] S2. Generate the corresponding task execution process according to the task data requirements and task data interaction protocol;

[0052] S3, setting dependencies between various task steps in the task execution process;

[0053] S4. Establish a task resource pool and a task adjustment pool, input the task execution process into the task resources to generate a number of task request elements, and then mark the required computing resources or storage resources for each task request element;

[0054] S5. Install the task adjustment pool on the client, and input the task request elements of different task execution processes into the task resource pool. After the task resource pool sets the priority for each task request element, the client executes multiple task execution processes in parallel according to the priority of each task request element and the required computing resources or storage resources.

[0055] Embodiment 2,

[0056] This embodiment is a further limitation of Embodiment 1, and Step S1 is implemented by the following process:

[0057] Setting up a cloud service platform, wherein the cloud service platform is provided with a client and a task management terminal;

[0058] The client is used to generate task requirements and execute the task execution process from the task management end;

[0059] The task needs to include a data exchange protocol and current client computer resources, wherein the data exchange protocol includes the data type and format requirements that the client can receive, the parameters required for the task, and the relationship and requirements between the parameters;

[0060] For example, if the value of parameter A is X, the value of parameter B must be Y. Before the client and the task management end transmit the task execution process, the data exchange protocol is used to verify whether there is a conflict between the parameters in the task execution process transmission;

[0061] The task management terminal is used to establish a task execution process and a task request pool according to task requirements, and a task resource pool based on current client computer resources, and then adjust the steps of multiple task execution processes of the same client, and send the adjusted task execution process to the client;

[0062] The client then generates task requirements and sends them to the task management end.

[0063] Embodiment 3,

[0064] This embodiment is a further limitation of Embodiment 1, and Step S2 is implemented by the following process:

[0065] After receiving the task requirements, the task management end establishes the task execution process according to the data exchange protocol in the task requirements;

[0066] The process of establishing the task execution flow includes:

[0067] The task management terminal establishes a task data input step according to the data exchange protocol, wherein the task data input step is provided with a data content check node and a data format check node;

[0068] It should be noted that the process of setting the data content check node and the data format check node includes: the task management end sets a number of feature words according to the parameters required for the task in the data exchange protocol, and the feature words are in a one-to-one correspondence with the parameters required for the task mentioned in the data exchange protocol, and then all the feature words are fitted to establish the data content check node, and at the same time, a corresponding data format check program and a data format conversion program are established according to the data format required in the data exchange protocol;

[0069] At the same time, corresponding task steps are established according to the corresponding relationship and requirements between various parameters, and the task steps include operations such as data format conversion, data cleaning, data analysis, data storage and data call;

[0070] Then, the task management end uses the task data requirement as the starting point of the task execution process, connects each task step in sequence to obtain the task execution process, and sets a task data output step at the end of the task execution process, wherein the data step is provided with a data format check node that is the same as the task data output.

[0071] Embodiment 4,

[0072] This embodiment is a further limitation of Embodiment 1, and step S3 is implemented by the following process:

[0073] A dependency mechanism is set between each task step in the task execution process, and the dependency mechanism includes:

[0074] Activation dependency: For two interrelated task steps s1 and s2 in the task execution process, when there is an activation dependency between task steps s1 and s2, when task step s1 is executed, task step s2 must be executed, which can be expressed as s1∈ActCond(s2);

[0075] Alternative dependency: When task steps s1, s2, ..., s n and task steps m When there is an alternative dependency between the task steps s1, s3, ..., s n When any task step is executed, task step s m Execute immediately, which can be expressed as s n ∈AltCond(s m ), where n and m are natural numbers greater than 0, and n is not equal to m;

[0076] Abort dependency: When there is an abort dependency between task step s1 and task step s2, it means that when task step s1 is aborted, task step s2 is also aborted synchronously, which can be expressed as s1∈AbtCond(s2);

[0077] Recovery dependency: When there is a recovery dependency between task step s1 and task step s2, it means that when task step s1 is reset, task step s2 is reset synchronously, which can be expressed as s1∈CpsCond(s2);

[0078] Cancellation dependency: When there is a cancellation dependency between task step s1 and task step s2, it means that when task step s1 is canceled, task step s2 is canceled synchronously, which can be expressed as s1∈CnlCond(s2);

[0079] Then, the task management end sets corresponding dependencies between the task steps according to the relationships between the parameters represented in the data exchange protocol.

[0080] Embodiment 5,

[0081] This embodiment is a further limitation of Embodiment 1, and step S4 is implemented by the following process:

[0082] The task management terminal establishes and sets a task resource pool and a task adjustment pool according to the current client computer resources of the client in the task requirements, and installs the task adjustment pool in the corresponding client;

[0083] The task resource pool includes the current computing resources and storage resources of the corresponding client, and then the task management end generates the corresponding task request element according to each task step in the task execution process;

[0084] The task request elements of different task execution processes are input into the task resource pool, and then the task resource pool operates according to the parameters and parameters required by the task request elements to obtain the computing resources and storage resources required by the corresponding task request elements, and the computing resources and storage resources are marked on the corresponding task request elements;

[0085] When the computing resource amounts and storage resource amounts of all task request elements are marked, the task resource pool sends all task request elements to the task adjustment pool in the client.

[0086] Embodiment 6,

[0087] This embodiment is a further limitation of Embodiment 1, and step S5 is implemented by the following process:

[0088] The task adjustment pool is provided with a priority adjustment mechanism and a resource usage queue, and the priority adjustment mechanism includes:

[0089] Absolute preemptive: When the task execution process executes to an absolutely preemptive task request element, the task step corresponding to the absolutely preemptive task request element can directly skip the queue and directly occupy the required computing resources or storage resources without waiting for the completion of other currently executed task steps, and enter the head of the resource usage queue;

[0090] Queued preemptive: If there is no absolutely preemptive task request element in the currently executed task step, it has the highest priority and enters the head of the resource usage queue;

[0091] Sequential preemptive: queue up and execute in the resource usage queue according to priority;

[0092] Idle preemptive: The task request element with the lowest priority can be executed only when there are enough idle computing resources or storage resources to execute the corresponding task steps;

[0093] It should be noted that when task request elements with absolute preemptive or queued preemptive modes exist at the same time, they are arranged in the resource usage queue according to the order of execution time;

[0094] Then, the task adjustment pool sets the priority of each task request element in each task adjustment pool according to the dependency relationship between the task steps corresponding to each task request element in the task execution process, that is, when the dependency relationship between two task steps is an activation dependency, the priority of the latter task step is set to absolute preemptive;

[0095] When the dependency between two task steps is an alternative dependency, the priority of the latter task step is set to idle preemptive;

[0096] When the dependency between two task steps is a recovery dependency, the priority of the latter task step is set to queue preemptive;

[0097] When the dependency between two task steps is abort dependency or cancel dependency, the priority of the latter task step is set to sequential preemptive;

[0098] Furthermore, when multiple task execution processes are executed in parallel on the client, each time the client executes a new task step in any task execution process, the corresponding task request element in the task adjustment pool is activated and arranged in the resource usage queue according to its priority;

[0099] When the client's current idle computing resources or storage resources meet the task request element at the head of the resource usage queue, the client executes the corresponding task steps;

[0100] Repeat the above steps until all task execution processes are completed.

[0101] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A multi-task coordinated parallel service orchestration method based on dynamic integration, characterized in that: The following steps are involved: S1. Setting up a cloud service platform, wherein the cloud service platform is provided with a client and a task management terminal, and then the client generates a task requirement and sends it to the task management terminal; S2, generate task execution process according to task requirements; S3, setting dependencies between various task steps in the task execution process; S4. Establish a task resource pool and a task adjustment pool, input the task execution process into the task resources to generate a number of task request elements, and then mark the required computing resources or storage resources for each task request element; S5. Install the task adjustment pool on the client, and input the task request elements of different task execution processes into the task resource pool. After the task resource pool sets the priority for each task request element, the client executes multiple task execution processes in parallel according to the priority of each task request element and the required amount of computing resources or storage resources.

2. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 1, characterized in that: The cloud service platform is provided with a client and a task management terminal; The client is used to generate task requirements and execute the task execution process from the task management end; The task needs to include a data exchange protocol and current client computer resources, wherein the data exchange protocol includes the data type and format requirements that the client can receive, the parameters required for the task, and the relationship and requirements between the parameters; The task management end is used to establish a task execution process and a task request pool according to task requirements, and a task resource pool based on current client computer resources, and then adjust the steps of multiple task execution processes of the same client, and send the adjusted task execution process to the client.

3. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 2, characterized in that: The process of generating a task execution flow based on task requirements includes: The task management terminal establishes a task data input step according to the data exchange protocol, wherein the task data input step is provided with a data content check node and a data format check node; At the same time, corresponding task steps are established according to the corresponding relationship and requirements between various parameters, and the task steps include data format conversion, data cleaning, data analysis, data storage and data calling; Then, the task management end uses the task data requirement as the starting point of the task execution process, connects each task step in sequence to obtain the task execution process, and sets a task data output step at the end of the task execution process, wherein the data step is provided with a data format check node that is the same as the task data output.

4. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 3, characterized in that: The dependencies include: Activation dependency: For two interrelated task steps s1 and s2 in the task execution process, when there is an activation dependency between task steps s1 and s2, when task step s1 is executed, task step s2 must be executed, which can be expressed as s1∈ActCond(s2); Alternative dependency: When task steps s1, s2, ..., s n and task steps m When there is an alternative dependency between the task steps s1, s3, ..., s n When any task step is executed, task step s m Execute immediately, which can be expressed as s n ∈AltCond(s m ), where n and m are natural numbers greater than 0, and n is not equal to m; Abort dependency: When there is an abort dependency between task step s1 and task step s2, it means that when task step s1 is aborted, task step s2 is also aborted synchronously, which can be expressed as s1∈AbtCond(s2); Recovery dependency: When there is a recovery dependency between task step s1 and task step s2, it means that when task step s1 is reset, task step s2 is reset synchronously, which can be expressed as s1∈CpsCond(s2); Cancellation dependency: When there is a cancellation dependency between task step s1 and task step s2, it means that when task step s1 is canceled, task step s2 is canceled synchronously, which can be expressed as s1∈CnlCond(s2); Then, the task management end sets corresponding dependencies between the various task steps according to the relationships between the various parameters represented in the data exchange protocol.

5. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 4, characterized in that: The task management terminal establishes and sets a task resource pool and a task adjustment pool according to the current client computer resources of the client in the task requirements, and installs the task adjustment pool in the corresponding client; The task resource pool includes the current computing resources and storage resources of the corresponding client, and then the task management end generates corresponding task request elements according to each task step in the task execution process.

6. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 5, characterized in that: The process of marking the amount of computing resources and storage resources for each task request element includes: The task request elements of different task execution processes are input into the task resource pool, and then the task resource pool operates according to the parameters and parameters required by the task request elements to obtain the amount of computing resources and storage resources required by the corresponding task request elements, and the amount of computing resources and storage resources are marked on the corresponding task request elements; When the computing resource amounts and storage resource amounts of all task request elements are marked, the task resource pool sends all task request elements to the task adjustment pool in the client.

7. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 5, characterized in that: The process of prioritizing the various task request elements involves: The task adjustment pool is provided with a priority adjustment mechanism and a resource usage queue. The priority adjustment mechanism includes: absolute preemptive, queue preemptive, sequential preemptive and idle preemptive; Then, the task adjustment pool sets the priority of each task request element in each task adjustment pool according to the dependency relationship between the task steps corresponding to each task request element in the task execution process, that is, when the dependency relationship between two task steps is an activation dependency, the priority of the latter task step is set to absolute preemptive; When the dependency between two task steps is an alternative dependency, the priority of the latter task step is set to idle preemptive; When the dependency between two task steps is a recovery dependency, the priority of the latter task step is set to queue preemptive; When the dependency between two task steps is abort dependency or cancel dependency, the priority of the latter task step is set to sequential preemptive.

8. The method for arranging multi-task coordinated parallel services based on dynamic integration according to claim 6, characterized in that: The process of the client executing the task execution flow includes: When multiple task execution processes are executed in parallel on the client, each time the client executes a new task step in any task execution process, the corresponding task request element in the task adjustment pool is activated and arranged in the resource usage queue according to its priority; When the current idle computing resources or storage resources of the client meet the task request element at the head of the resource usage queue, the client executes the corresponding task steps and repeats the above operations until the entire task execution process is completed.

Citation Information

Cited By

  • Full-scene multi-factor authentication equipment scheduling method and system under Linux

    CN119892415A