CN111506413B - Intelligent task scheduling method and system based on business efficiency optimization - Google Patents
Intelligent task scheduling method and system based on business efficiency optimization Download PDFInfo
- Publication number
- CN111506413B CN111506413B CN202010623617.4A CN202010623617A CN111506413B CN 111506413 B CN111506413 B CN 111506413B CN 202010623617 A CN202010623617 A CN 202010623617A CN 111506413 B CN111506413 B CN 111506413B
- Authority
- CN
- China
- Prior art keywords
- task
- resource
- tasks
- execution
- weight coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
- G06F9/4887—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues involving deadlines, e.g. rate based, periodic
Abstract
The invention discloses an intelligent task scheduling method and system based on service efficiency optimization. The intelligent task scheduling method based on the service efficiency optimization comprises the following steps: dividing a task running queue into a plurality of different task running channels according to different service types established by a plurality of resources, obtaining the comprehensive calculation expected value of each resource task according to a user priority weight coefficient, a task waiting time weight coefficient and a task retry frequency weight coefficient, taking the comprehensive calculation expected value as the execution priority of the resource task, dynamically adjusting the execution priorities of the plurality of resource tasks in the different task running channels in real time in the execution process of the resource task, monitoring the current state of the resource task, and adjusting the current state of the resource task. The invention can meet the requirement of the execution efficiency of a high-level user and can improve the overall execution efficiency of all tasks.
Description
Technical Field
The invention relates to the technical field of task scheduling, in particular to an intelligent task scheduling method and system based on service efficiency optimization.
Background
In the operation and maintenance process of the cloud computing platform, a task scheduling problem with complex execution order dependence is often encountered.
The task scheduling method in the prior art generally comprises: the latter task needs to be executed when the former task or the former tasks that depend on the latter task need to be executed, and if there are many tasks or there is a complex execution sequence that depends on the tasks, the existing scheduling method can schedule the tasks, but does not consider that users of different levels have different requirements for the execution efficiency of the tasks, and does not consider the overall efficiency of task execution, resulting in longer average time for task completion, especially poor experience of users of high level, and thus needs to be improved urgently.
Disclosure of Invention
In view of the foregoing drawbacks of the prior art, an object of the present invention is to provide an intelligent task scheduling method and system based on service efficiency optimization, which are used to solve the problems that although the existing scheduling method can perform task scheduling, users of different levels have different requirements on the execution efficiency of tasks and the overall efficiency of task execution is not considered, resulting in long average time for task completion, and especially poor experience effect for users of high level.
In order to achieve the above and other related objects, the present invention provides an intelligent task scheduling method based on service efficiency optimization, including:
s0, defining the current state of the resource task, wherein the current state comprises a to-be-run state, a running state, a pause state, a running success state and a running failure state;
s1, arranging a plurality of resources according to the requirements of each user to obtain a plurality of resource tasks which are related to the requirements of the user and have a dependence execution sequence, and initializing the current state of each resource task as a to-be-run state;
s2, outputting all resource tasks with the dependent execution sequence to a task waiting queue;
s3, dividing a task running queue into a plurality of different task running channels according to different service types created by a plurality of resources; the plurality of different task operation channels comprise one or more of a newly-built task operation channel, a backup task operation channel, an updated task operation channel and a deleted task operation channel;
s4, calculating a user priority weight coefficient, a task waiting time weight coefficient and a task retry number weight coefficient;
s5, obtaining the comprehensive calculation expected value of each resource task according to the user priority weight coefficient, the task waiting time weight coefficient and the task retry time weight coefficient, and taking the comprehensive calculation expected value as the execution priority of the resource task;
s6, according to different service types of resource creation, a plurality of resource tasks are transmitted to different task operation channels;
s7, executing the resource tasks according to the descending order of the execution priority, and dynamically adjusting the execution priority of a plurality of resource tasks in different task operation channels in real time in the execution process of the resource tasks;
s8, removing the resource tasks which are successfully executed and the resource tasks which are failed to be executed from the task running channel;
s9, monitoring the current state of the resource task, adjusting the current state of the resource task, and if the current state of the resource task is adjusted to be a to-be-run state, outputting the resource task with the current state to be the to-be-run state to a task waiting queue;
s10, waiting until all the resource tasks in different task running channels are successfully executed, and ending the resource task scheduling;
the step of calculating the task waiting time weight coefficient in step S4 includes:
s411, obtaining queue time for executing the resource taskPQueuing time of said resource taskPIs the time period from the resource task entering the task waiting queue to the time before the resource task is executed;
s412, calculating the average queuing time of all resource tasksP avg ；
S413, queuing time according to the resource taskPAnd average queuing time of all resource tasksP avg To obtain a task latency weight coefficientα，
；
S414, the task waiting time weight coefficientαIncluding values between 0 and 1.
In an embodiment of the present invention, the step of calculating the user priority weighting factor in step S4 includes:
s41, obtaining the user level according to the monthly consumption amount or the annual consumption amount of the user;
s42, obtaining a user priority weighting coefficient according to the user levelR；
S43, the user priority weighting coefficientRIncluding values between 0 and 1.
In an embodiment of the present invention, the step of calculating the task retry number weighting factor in step S4 includes:
s4111, obtaining retry times of executing the resource taskNNumber of retries of said resource taskNRetry execution times after the execution of the resource task fails;
s4112, calculating average retry times of all resource tasksN avg ；
S4113, retrying times according to the resource taskNAnd average number of retries for all resource tasksN avg To obtain the task retry number weight coefficientβ，
；
S4114, weighting coefficient of the number of task retriesβIncluding values between 0 and 1.
In an embodiment of the present invention, the step of obtaining the total calculated expected value of each resource task according to the user priority weighting factor, the task waiting time weighting factor and the task retry number weighting factor in step S5 includes:
according to a comprehensive calculation expected value formula:V=R×（α+β) Wherein, in the step (A),
，
then, the expected value is calculated comprehensively
Wherein, in the step (A),Rrepresents a user priority weight coefficient and,αweight system for representing task latencyThe number of the first and second groups is,βa weight coefficient indicating the number of task retries,Pindicating the queuing time of the resource task,P avg representing the average queuing time of all resource tasks,Nindicating the number of retries of the resource task,N avg indicating the average number of retries for all resource tasks.
In an embodiment of the present invention, the step of monitoring the current status of the resource task and adjusting the current status of the resource task in step S9 includes:
s91, monitoring the current state of the resource task through a human-computer interaction interface;
and S92, adjusting the current state of the resource task through the human-computer interaction interface.
In an embodiment of the present invention, the step of adjusting the current state of the resource task through the human-machine interface in step S92 includes:
defining an expiration completion time for the resource task;
adjusting the current state of the resource task from a to-be-run state to a pause state;
adjusting the current state of the resource task from a pause state to a to-be-run state;
adjusting the current state of the resource task from a running failure state to a to-be-run state;
and adjusting the current state of the resource task from the running success state to the to-be-run state.
In an embodiment of the present invention, the sequentially executing the resource tasks in step S7 is to execute the resource tasks in different task execution channels in parallel and sequentially.
The invention also provides an intelligent task scheduling system based on the optimization of the service efficiency, which comprises the following components:
the resource task arranging equipment is used for arranging a plurality of resources according to the requirements of each user to obtain a plurality of resource tasks which are related to the requirements of the user and have a dependence execution sequence, and initializing the current state of each resource task as a to-be-run state;
the task delivery unit is used for outputting all resource tasks with the execution-dependent sequence to the task waiting queue;
the task delivery strategy control unit is used for dividing a task running queue into a plurality of different task running channels according to different service types established by a plurality of resources; the plurality of different task operation channels comprise one or more of a newly-built task operation channel, a backup task operation channel, an updated task operation channel and a deleted task operation channel;
the task delivery strategy control unit is used for calculating a user priority weight coefficient, a task waiting time weight coefficient and a task retry frequency weight coefficient;
the task delivery strategy control unit is used for obtaining the comprehensive calculation expected value of each resource task according to the user priority weight coefficient, the task waiting time weight coefficient and the task retry frequency weight coefficient, and taking the comprehensive calculation expected value as the execution priority of the resource tasks;
the task delivery strategy control unit is used for transmitting a plurality of resource tasks to different task operation channels according to different service types established by the resources;
the task execution equipment is used for executing the resource tasks according to the descending order of the execution priority;
the task execution strategy control unit is used for dynamically adjusting the execution priority of a plurality of resource tasks in different task operation channels in real time in the execution process of the resource tasks;
the task execution equipment is used for removing the resource tasks which are successfully executed and the resource tasks which are failed to be executed from the task operation channel;
the human-computer interaction interface is used for monitoring the current state of the resource task and adjusting the current state of the resource task;
the task execution device is used for outputting the resource task of which the current state is the to-be-operated state to a task waiting queue if the current state of the resource task is adjusted to the to-be-operated state;
and the task execution equipment is used for finishing the resource task scheduling after waiting until all the resource tasks in the different task operation channels are successfully executed.
The invention also provides electronic equipment which comprises a processor and a memory, wherein the memory stores program instructions, and the processor runs the program instructions to realize the intelligent task scheduling method based on the service efficiency optimization.
The present invention also provides a computer-readable storage medium, which stores computer instructions for causing the computer to execute the above-mentioned intelligent task scheduling method based on business efficiency optimization.
As described above, the intelligent task scheduling method and system based on the optimization of the service efficiency according to the present invention have the following beneficial effects:
the intelligent task scheduling method based on the optimization of the service efficiency considers that users of different levels have different requirements on the execution efficiency of the tasks and the overall efficiency of task execution, the average time consumption of task completion is short, and the experience effect of the users is good in the task scheduling process.
The intelligent task scheduling method based on the business efficiency optimization dynamically adjusts the execution priority of a plurality of resource tasks in different task operation channels in real time in the execution process of the resource tasks, can adopt a DRTP algorithm to define the task value according to the user priority, the task waiting time and the task retry times of the resource tasks, dynamically obtains the value density by integrating the task value and the execution time, and ensures that the resource tasks with the highest value density are executed preferentially.
The intelligent task scheduling method based on the optimization of the service efficiency comprehensively considers the user priority, the task waiting time and the task retry times, and ensures the resource task with high value density to be executed preferentially.
The intelligent task scheduling method based on the optimization of the service efficiency transmits a plurality of resource tasks to different task operation channels according to different service types created by the resources, and the tasks in the different task operation channels can be executed in parallel without influencing each other and occupying resources.
The intelligent task scheduling method based on the optimization of the service efficiency can monitor the current state of the resource task in real time, and effectively ensures the normal execution of the resource task.
The intelligent task scheduling method based on the optimization of the service efficiency is provided with a human-computer interaction interface, and the current state of the resource task can be adjusted in time according to the requirement.
Drawings
Fig. 1 is a flowchart of a method for scheduling an intelligent task based on service efficiency optimization according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a sample task execution dependency sequence for arranging multiple resources according to an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present application.
Fig. 3 is a schematic diagram illustrating a principle of a resource task and a task waiting queue having an execution order dependency according to an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present application.
Fig. 4 is a flowchart illustrating a user priority weighting factor calculating process in step S4 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present disclosure.
Fig. 5 is a flowchart illustrating the task waiting time weighting factor calculation in step S4 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present application.
Fig. 6 is a flowchart illustrating the task retry number weighting factor calculation in step S4 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present application.
Fig. 7 is a flowchart of a working process of a DRTP algorithm of an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present application.
Fig. 8 is a flowchart of a step S9 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present application.
Fig. 9 is a schematic block diagram of a structure of an intelligent task scheduling system based on service efficiency optimization according to an embodiment of the present application.
Fig. 10 is a schematic block diagram of a structure of an electronic device according to an embodiment of the present application.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the drawings only show the components related to the present invention rather than the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Referring to fig. 1, fig. 1 is a flowchart illustrating an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present disclosure. The invention provides an intelligent task scheduling method based on service efficiency optimization, which ensures that a resource task with the highest value density is executed preferentially, and comprises the following steps:
as shown in fig. 1, in step S0, current states of the resource task are defined, where the current states include a to-be-run state, an in-run state, a pause state, a successful-run state, and a failed-run state. And the to-be-run state indicates that the resource task is in a queuing state and is waiting to be executed. The in-flight state indicates that a resource task is executing. The pause state indicates that the resource task is paused but still remains in the current task execution path. The run-successful status indicates that the resource task was successfully executed. The operation failure state indicates that the resource task fails to execute and needs to be executed again.
As shown in fig. 1, in step S1, a plurality of resources are arranged according to the user requirements, so as to obtain a plurality of resource tasks related to the user requirements and having an execution order, and a current state of each resource task is initialized to a to-be-executed state. Specifically, in step S1, a plurality of resources are organized according to the needs of each user to obtain a plurality of organized resources with dependency relationships. And arranging the plurality of resources with the dependency relationship to obtain a plurality of resource tasks with dependency execution sequences related to the user requirements.
Referring to fig. 2, fig. 2 is a schematic diagram of a task execution dependency sequence for arranging multiple resources according to an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present application. In step S1, a plurality of resources may be orchestrated, but not limited to, through a cloud orchestration interface. A plurality of such resources include, but are not limited to, Virtual Private Cloud (VPC), subnet, Network Address Translation (NAT), gateway, eip (elastic ip), rds (relational Database service), Virtual Machine (VM), Cloud disk, and the like. In step S1, the resource task may be orchestrated, but not limited to, through a cloud orchestration interface. The types of the resource tasks include, but are not limited to, a new resource task, a backup resource task, an update resource task, and a delete resource task. Specifically, the resource task may be, but is not limited to, a new project atomic task, an updated project atomic task, a deleted project atomic task, a new VPC atomic task, an updated VPC atomic task, a deleted VPC atomic task, a new subnet atomic task, an updated subnet atomic task, a deleted subnet atomic task, a new EIP atomic task, an updated EIP atomic task, a deleted EIP atomic task, a new cloud disk atomic task, an updated cloud disk atomic task, a deleted cloud disk atomic task, a new security group atomic task, an updated security group atomic task, a deleted security group atomic task, a new key atomic task, a deleted key atomic task, a new security group rule atomic task, a deleted security group rule atomic task, a new NAT gateway atomic task, an updated NAT gateway atomic task, a deleted NAT gateway, a new LB atomic task, a new key atomic task, a deleted key atomic task, a new security group rule atomic task, a, The method comprises the steps of updating LB atomic task, deleting LB atomic task, newly building RDS atomic task, updating RDS atomic task, deleting RDS atomic task, newly building port atomic task, deleting port atomic task, newly building virtual machine atomic task, updating virtual machine atomic task, deleting virtual machine atomic task, binding port and virtual machine atomic task, unbinding port and virtual machine atomic task, binding security group and port atomic task, unbinding security group and port atomic task, binding port and NAT gateway atomic task, unbinding port and NAT gateway atomic task, binding EIP and port atomic task, unbinding EIP and port atomic task, binding virtual machine and cloud disk atomic task, and unbinding virtual machine and cloud disk atomic task. For example, if the resource is a VPC, the resource tasks include a new VPC atomic task, an update VPC atomic task, and a delete VPC atomic task. For example, if the resource is an EIP, the resource tasks include a new EIP atomic task, an update EIP atomic task, and a delete EIP atomic task. The plurality of resource tasks having a dependent execution order include: for example, as shown in fig. 2, a cloud disk binding atomic task needs to be executed first, and after the cloud disk binding atomic task is completed, a new cloud disk atomic task and a new virtual machine atomic task are executed. The plurality of resource tasks having a dependent execution order include: as another example, as shown in fig. 2, a new virtual machine atomic task and a new RDS atomic task are executed, the new virtual machine atomic task and the new RDS atomic task are all completed, and then a new sub-network atomic task is executed. Specifically, the resource tasks with execution order dependency can be obtained by arranging the resources through a cloud arrangement interface of an industry standard.
As shown in fig. 1, step S2 outputs all resource tasks having a dependent execution order to the task waiting queue 2. Specifically, the resource tasks of each user have a dependent execution order, and the resource tasks of different users do not have a dependent execution order.
Referring to fig. 3, fig. 3 is a schematic diagram illustrating a principle of a task waiting queue and resource tasks with execution order dependency according to an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present application. And if the user a needs to newly build a communicable virtual machine, obtaining a newly-built virtual machine port atomic task, a virtual machine port atomic task updating and a virtual machine atomic task newly-built which are related to the newly-built virtual machine and have a dependency on an execution sequence, and if the user b needs to newly build a cloud disk, obtaining a cloud disk binding atomic task and a cloud disk atomic task newly-built which are related to the newly-built cloud disk and have a dependency on the execution sequence. The dependent execution sequence of the tasks belonging to the user a is the resource task execution dependent sequence 12 obtained according to the requirement of the user a, which is to execute the newly-built virtual machine port atomic task, then execute the updated virtual machine port atomic task, and finally execute the newly-built virtual machine atomic task. The dependent execution sequence of the tasks belonging to the user b is the resource task execution dependent sequence 13 obtained according to the requirement of the user b, namely, the bound cloud disk atomic task is executed firstly, and then the newly-built cloud disk atomic task is executed. The tasks belonging to the user a and the tasks belonging to the user b in the task waiting queue 2 do not depend on the execution order.
As shown in fig. 1, step S3, according to the difference of the service types created by the multiple resources, divides the task execution queue 7 into multiple different task execution channels; the plurality of different task execution channels include, but are not limited to, one or more of a new task execution channel 71, a backup task execution channel 72, an update task execution channel 73, and a delete task execution channel 74. Specifically, as shown in fig. 2, for example, the newly created VPC atomic task, the newly created subnet atomic task, the newly created EIP atomic task, the newly created cloud disk atomic task, and the like may be transmitted to a newly created task operation channel 71, the backup VPC atomic task, the backup subnet atomic task, the backup EIP atomic task, the backup cloud disk atomic task, and the like may be transmitted to a backup task operation channel 72, the updated VPC atomic task, the updated subnet atomic task, the updated EIP atomic task, the updated cloud disk atomic task may be transmitted to an updated task operation channel 73, the deleted VPC atomic task, the deleted subnet atomic task, the deleted EIP atomic task, and the deleted cloud disk atomic task may be transmitted to a deleted task operation channel 74.
As shown in fig. 1, step S4 calculates a user priority weighting factor, a task waiting time weighting factor, and a task retry number weighting factor. Specifically, when entering different task running channels, user priority, task waiting time and task retry times are comprehensively considered.
Referring to fig. 4, fig. 4 is a flowchart illustrating a user priority weighting factor calculating process in step S4 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present application. The step of calculating the user priority weight coefficient in step S4 includes: and step S41, obtaining the user level according to the monthly consumption amount or the annual consumption amount of the user. Step S42, according to the user level, obtaining a user priority weighting coefficientR. Step S43, the user priority weighting factorRIncluding values between 0 and 1. Specifically, the user level includes, but is not limited to, a first level user, a second level user, a third level user, a fourth level user, and the like, the first level user may correspond to a VVIP user, the second level user may correspond to a VIP user, the third level user may correspond to a normal user, and the fourth level user may correspond to a low quality user. The higher the user's monthly or yearly spending amount, the higher the user level and user priority weighting factor. For example, the monthly consumption amount of the low-quality user is lower than 50% of the average amount, the user priority weighting coefficient is 0.2, the monthly consumption amount of the common user is higher than 50% of the average amount, the user priority weighting coefficient is 0.5, and the user priority of the VIP user is 0.2The priority weighting coefficient is 0.8, and the user priority weighting coefficient of the VVIP user is 1.
Referring to fig. 5, fig. 5 is a flowchart illustrating the task waiting time weight coefficient calculation in step S4 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present application. The step of calculating the task waiting time weight coefficient in step S4 includes: step S411, obtaining the queue time for executing the resource taskPQueuing time of said resource taskPIs the period of time from when the resource task enters the task wait queue 2 until the resource task is executed. Step S412, calculating the average queuing time of all resource tasksP avg . Step S413, according to the queuing time of the resource taskPAnd average queuing time of all resource tasksP avg To obtain a task latency weight coefficientα，
. Step S414, the task waiting time weight coefficientαIncluding values between 0 and 1. In particular, the queuing time of the resource taskPThe larger the task is, the longer the waiting time of the task is, the task waiting time weight coefficientαThe larger.
Referring to fig. 6, fig. 6 is a flowchart illustrating the task retry number weighting factor calculation in step S4 of the intelligent task scheduling method based on service efficiency optimization in fig. 1 according to an embodiment of the present application. The step of calculating a task retry number weighting factor in step S4 includes: s4111, obtaining retry times of executing the resource taskNNumber of retries of said resource taskNAnd the number of times of retry execution after the execution of the resource task fails. S4112, calculating average retry times of all resource tasksN avg . S4113, retrying times according to the resource taskNAnd average number of retries for all resource tasksN avg To obtain the task retry number weight coefficientβ，
. S4114, weighting coefficient of the number of task retriesβIncluding values between 0 and 1. Specifically, the number of retries of the resource taskNThe larger the number of times of retry execution after the execution failure of the resource task is, the more the weight coefficient of the number of times of retry execution of the task isβThe larger.
As shown in fig. 1, step S5 is to obtain a total calculated expected value of each resource task according to the user priority weighting factor, the task waiting time weighting factor and the task retry number weighting factor, and use the total calculated expected value as the execution priority of the resource task. Specifically, according to the formula of comprehensively calculating the expected value:V=R×（α+β) Wherein, in the step (A),
，
then, the expected value is calculated comprehensively
Wherein, in the step (A),Rrepresents a user priority weight coefficient and,αa weight coefficient representing the task latency is represented,βa weight coefficient indicating the number of task retries,Pindicating the queuing time of the resource task,P avg representing the average queuing time of all resource tasks,Nindicating the number of retries of the resource task,N avg indicating the average number of retries for all resource tasks. The user priority weighting factorRTask latency weighting factorαTask retry number weighting factorβThe larger the value, the more the expected value is calculatedVThe larger the size, the better the task with the highest value density can be executed.
As shown in fig. 1, step S6 is to deliver a plurality of resource tasks to different task execution channels according to different service types created by the resource. Specifically, according to different service types created by resources, the task running queue 7 may be divided into a newly-created task running channel 71, a backup task running channel 72, an update task running channel 73, a delete task running channel 74, and the like, and by placing resource tasks into different task running channels, tasks in different task running channels may be executed in parallel without affecting each other and occupying resources.
As shown in fig. 1, in step S7, the resource tasks are executed in the descending order of the execution priority, and in the execution process of the resource tasks, the execution priorities of the resource tasks in different task execution channels are dynamically adjusted in real time. A DRTP (Dynamic Real-time Transaction Scheduling) algorithm can be adopted to dynamically adjust the execution priority of a plurality of resource tasks in different task running channels in Real time, the task value is defined according to the user priority, the task waiting time and the task retry times of the resource tasks, the value density is obtained by integrating the task value and the execution time dynamically, and the resource task with the highest value density is guaranteed to be executed preferentially.
Referring to fig. 7, fig. 7 is a flowchart illustrating a working flow of a DRTP algorithm of an intelligent task scheduling method based on service efficiency optimization according to an embodiment of the present application. The DRTP algorithm comprises the following steps:
as shown in FIG. 7, step S01 is executed to traverse the next task in the task execution queue 7T i . In particular, the next task to traverseT i As a current taskT i Current taskT i Is set as a taskT i Integrated calculated expected value ofV i ，V i The expected value is calculated for the combination defined in step S5 of fig. 1.
As shown in FIG. 7, step S02 is to determine the current task according to the task-dependent execution orderT i Whether all the prepositive tasks are executed and finished or not, if the current task is finishedT i If all the tasks are completed, the operation of step S03 is executed; if the current task isT i The execution of all the preceding tasks is not completed, the operation of step S04 is performed.
As shown in FIG. 7, step S03 is executed to determine the current taskT i If the task is in a suspended state, if the current task is in the suspended stateT i If not, the operation of step S05 is executed; if the current task isT i In the suspend state, the operation of step S06 is performed.
As shown in FIG. 7, step S04, Current taskT i Is set to 0, the operation of step S05 is performed.
As shown in FIG. 7, step S05 is executed to determine the current taskT i Whether or not there is the highest static priority in the task run queue 7, if the current taskT i If the task has the highest static priority in the task running queue 7, the operation of step S07 is executed, and if the current task has the highest static priorityT i The task run queue 7 does not have the highest static priority, the operation of step S01 is performed.
As shown in FIG. 7, step S06, Current taskT i Is set to 0, the operation of step S05 is performed.
As shown in FIG. 7, step S07 is executed to get the current taskT i Set to currently executing taskT c ，
As shown in FIG. 7, step S08, the currently executing task is executedT c 。
As shown in FIG. 7, step S09 is executed to traverse the next task in the task execution queue 7T j 。
As shown in FIG. 7, step S010 judges the taskT j Whether all the prepositive tasks are executed and finished or not, if the tasks are executed and finishedT j If all the prepositive tasks are executed, executing the operation of the step S011; if taskT j If all the tasks are not completed, the operation of step S012 is executed.
As shown in FIG. 7, step S011, judge taskT j Whether it is the currently executing taskT c If the task isT j Not currently performing the taskT c If yes, executing operation of step S013; if taskT j For the currently performed taskT c Then, the operation of step S014 is performed.
As shown in fig. 7, step S012 and taskT j Is set to 0, the operation of step S017 is performed.
As shown in FIG. 7, step S013 and the judgment taskT j Whether the task is in a suspended state or not, if the task is in a suspended stateT j If the task is in the suspended state, the operation of step S016 is executed, and if the task is in the suspended stateT j If not, the operation of step S015 is executed.
As shown in FIG. 7, step S014, taskT j Is set to 0, the operation of step S017 is performed.
As shown in FIG. 7, step S015, the calculation taskT j Continues to perform the operation of step S017. Wherein the computing taskT j The step of dynamic prioritization of (2) comprises:
a. calculating an instant value ofIV j ，
Wherein the content of the first and second substances,IV j (t) Representing tasksT j ImplementtTime per unit timeTHas an immediate value ofIV j (t)；V j Representing tasksT j The expected value of the overall calculation of (c),V j calculating an expected value for the combination defined in step S5 of fig. 1;Ta collection of tasks is represented that are performed,T=｛T 1, T 2,..., T n ｝，C j representing tasksT j The theoretical execution time of (2) is, in the present embodiment,C j the average value of the execution time of the tasks of the same type is estimated by using historical records,pindicating adjustment tasksT j The dynamic cost density affects the weighting factor of the priority, which, in this embodiment, is, for example,pthe value was chosen to be 1.0.
b. Computing tasksT j The remaining value density produced per unit time within its remaining execution time, the taskT j The remaining value of (A) is recorded asRVD j (t) Task ofT j The residual value density of (a) is defined as:
，
wherein the content of the first and second substances,V j representing tasksT j The expected value of the overall calculation of (c),IV j (t) Representing tasksT j ImplementtThe instantaneous value of the time per unit of time,C j representing tasksT j The theoretical execution time of (a) is,pindicating adjustment tasksT j The dynamic value density affects the coefficients of the weights for the priorities.
c. Computing tasksT j Intensity of execution of, taskT j The execution strength of (A) is defined as the ratio of the execution time required for completing the task to the task free time, and is recorded as
Task ofT j The execution intensity calculation formula of (1) is:
wherein, in the step (A),C j representing tasksT j The theoretical execution time of (a) is,trepresenting tasksT j The time that has been executed has been elapsed,d j representing tasksT j The absolute time-out period of (a) is,τ j indicating the current time of day, in this embodiment,d j given by operation and maintenance personnel through the human-computer interaction interface 9, the default value is the slave taskT j The generation time was counted 24 hours later. To ensure the task is satisfiedT j Completed according to the deadline and the taskT j The degree of urgency to be fulfilled as quickly as possible is called a taskT j Urgency of execution, notedTask ofT j Urgency of executionComprises the following steps:
，
wherein the content of the first and second substances,qindicating adjustment tasksT j The temporal urgency affects the coefficients of the weights to the priority,
representing tasksT j The intensity of execution of (a) is,C j representing tasksT j The theoretical execution time of (a) is,trepresenting tasksT j The time that has been executed has been elapsed,d j representing tasksT j The absolute time-out period of (a) is,τ j indicating the current time by selecting different parametersqThe influence weight of the dynamic value density and the urgency of the task on the task priority can be changed, and the adaptability to different application environments is improved, in the embodiment, for example,qthe value was chosen to be 1.1.
d. Computing tasksT j Dynamic priority ofDyPri(T j ) Dynamic priorityDyPri(T j ) The calculation formula is as follows:
，
wherein the content of the first and second substances,RVD j (t) Representing tasksT j The remaining value density of (a) of (b), j (t) Representing tasksT j The urgency of the execution,C j representing tasksT j The theoretical execution time of (a) is,trepresenting tasksT j The time that has been executed has been elapsed,V j representing tasksT j The expected value of the overall calculation of (c),pindicating adjustment tasksT j The dynamic value density affects the coefficients of the weights for the priorities,qindicating adjustment tasksT j The temporal urgency affects the coefficients of the weights to the priority,d j representing tasksT j The absolute time-out period of (a) is,τ j indicating the current time of day.
As shown in fig. 7, step S016 and taskT j Is set to 0, the operation of step S017 is performed.
As shown in FIG. 7, step S017, the task is judged according to the formula (1)T j Whether the dynamic priority of (1) is greater than or equal to the currently executing taskT c The formula (1) is:T j ＞β×T c wherein, in the step (A),βrepresents a bump avoidance coefficient; in a real-time system based on dynamic priority, a phenomenon that two or more tasks alternately rise in priority to cause repeated preemption among the tasks occurs, which is called a thrashing phenomenon. The reason for introducing the bump avoidance coefficient in equation (1) is to avoid the system from generating the bump phenomenon, and in this embodiment,βthe value of (c) may be chosen to be 1.1. If the formula (1) is satisfied, the operation of step S018 is performed, and if the formula (1) is not satisfied, the operation of step S019 is performed.
As shown in fig. 7, in step S018, it is determined whether the first time comparison formula is satisfied as formula (2), where formula (2) is:d j －τ j ＜C j －t j +C c －t c wherein, in the step (A),d j representing user definable tasksT j The time-off for completion of the process,τ j which indicates the current time of day,d j －τ j indicating that the rest is available to perform the taskT j The total time of the operation of the motor,C j representing tasksT j The theoretical execution time of the completion is,t j representing tasksT j The time that has been currently executed is,C j －t j representing tasksT j But also how long it is necessary to perform,C c indicating a currently executing taskT c The theoretical execution time of the completion is,t c indicating a currently executing taskT c The time that has been currently executed is,C c －t c indicating a currently executing taskT c How long it takes to execute, if equation (2) holds, then the task is statedT j Time-critical, taskT j From the current moment to the deadline, it can be used to execute the taskT j Is less than the taskT j And the currently executing taskT c If the sum of the time required for completing the two tasks is less than the preset time, executing the step S020, and if the formula (2) is not satisfied, indicating that the tasks are completedT j Time is not urgent and task is not urgentT j From the current moment to the deadline, it can be used to execute the taskT j Is greater than or equal to the taskT j And the currently executing taskT c The sum of the time required for both tasks to complete, the operation of step S022 is performed.
As shown in FIG. 7, step S019 continues to execute the currently executing taskT c Task ofT j The original state is maintained, and the operation of step S023 is continuously performed.
As shown in fig. 7, in step S020, it is determined whether the second time comparison formula is formula (3), where formula (3) is:d c －τ j ＜C c －t c +C j －t j wherein, in the step (A),d c representing a user definable current executionLine taskT c The time-off for completion of the process,τ j which indicates the current time of day,d c －τ j indicating that the rest is available to perform the taskT c The total time of the operation of the motor,t j representing tasksT j The time that has been currently executed is,C j －t j representing tasksT j But also how long it is necessary to perform,C c indicating a currently executing taskT c The theoretical execution time of the completion is,t c indicating a currently executing taskT c The time that has been currently executed is,C c －t c indicating a currently executing taskT c How long it takes to execute, if equation (3) holds, it indicates the task currently being executedT c Time critical, currently executing taskT c From the current moment to the deadline, it can be used to execute the taskT c Is less than the taskT j And the currently executing taskT c And (4) if the sum of the time required for completing the two tasks is less than the preset time, executing the step S021, and if the formula (3) is not satisfied, indicating that the task is currently executedT c Is not time-critical, the task is currently executedT c From the current moment to the deadline, it can be used to execute the taskT c Is greater than or equal to the taskT j And the currently executing taskT c And if the sum of the time required for completing the two tasks is less than the preset time, executing the operation of step S019.
As shown in FIG. 7, step S021, current taskT j And the currently executing taskT c All the time of (2) is urgent, whether formula (3) is established or not is judged, and formula (3) is:
wherein the content of the first and second substances,C j representing tasksT j The theoretical execution time of the completion is,t j representing tasksT j The time that has been currently executed is,pindicating adjustment tasksT j The dynamic value density affects the coefficients of the weights for the priorities,V j representing tasksT j The expected value of the overall calculation of (c),V c representing tasksT c The expected value of the overall calculation of (c),C c indicating a currently executing taskT c The theoretical execution time of the completion is,t c representing tasksT c The time that has been currently executed is,
representing tasksT c The real-time value that has been generated when the death occurs,
representing tasksT j Compensation tasksT c The expected value of the protein after the death is complete,d j representing tasksT j The absolute time-out period of (a) is,τ j which indicates the current time of day,βthe bump avoidance coefficient is expressed as a coefficient of bump avoidance,qindicating adjustment tasksT j The temporal urgency affects the coefficients of the weights to the priority,d c indicating a currently executing taskT c Absolute deadline of (c). In particular, if the task isT c Is tasked withT j After the preemption, the task will miss the deadline and die, so the task will dieT c Will enable the taskT c The generated immediate value is not effective. Therefore, if the task isT j Task of death and deathT c Task ofT j Need to make up for the taskT c Losing immediate value due to premature death. Formula (II)
To correct the currently performed taskT c Task of death lossT j The current execution task is judged and corrected through the formula (3)T c Task of death lossT j Whether the dynamic priority of (1) is greater than or equal to the currently executing taskT c If equation (3) is satisfied, the operation of step S022 is performed, and if equation (3) is not satisfied, the operation of step S019 is performed.
Step S022, as shown in FIG. 7, a taskT j Preempting a currently executing taskT c . In particular, the task to be currently executedT c The state of the task is changed into a to-be-run state, and the current execution task is setT c Equaling tasksT j Executing the currently executing taskT c 。
As shown in FIG. 7, step S023, determining the currently executed taskT c Whether the execution is finished or not, if the task is currently executedT c If the execution is finished, executing the operation of the step S024; if the task is currently executedT c If the execution is not completed, the operation of step S09 is performed.
As shown in fig. 7, in step S024, it is determined whether there is an unexecuted task in the task list, and if there is an unexecuted task in the task list, the operation in step S01 is performed; and if the task list has no unexecuted tasks, ending the execution of the tasks.
As shown in FIG. 1, step S8 is to remove the resource task that executed successfully and the resource task that executed unsuccessfully from the task execution path.
As shown in fig. 1, in step S9, the current state of the resource task is monitored, the current state of the resource task is adjusted, and if the current state of the resource task is adjusted to be the to-be-executed state, the resource task whose current state is the to-be-executed state is output to the task waiting queue 2.
Referring to fig. 8, fig. 8 is a flowchart illustrating a step S9 of the intelligent task scheduling method based on business efficiency optimization in fig. 1 according to an embodiment of the present application. The step of monitoring the current state of the resource task and adjusting the current state of the resource task in step S9 includes: and S91, monitoring the current state of the resource task through the human-computer interaction interface 9. And S92, adjusting the current state of the resource task through the human-computer interaction interface 9. The current state of the resource task comprises a to-be-run state, a running state, a pause state, a running success state and a running failure state.
As shown in fig. 8, in step S91 and step S92, the resource tasks may be, but are not limited to being, displayed in a list form. The step S92 of adjusting the current state of the resource task through the human-computer interaction interface 9 includes: adjusting the current state of the resource task from a to-be-run state to a pause state, adjusting the current state of the resource task from the pause state to the to-be-run state, adjusting the current state of the resource task from a running failure state to the to-be-run state, adjusting the current state of the resource task from a running success state to the to-be-run state, and defining the ending completion time of the resource task. For example, when a user requires that a resource task needs to be executed preferentially, if the current state of the resource task is a suspended state, the resource task can be executed preferentially through the human-computer interaction interface 9, and the current state of the resource task is adjusted from the suspended state to a to-be-executed state. Specifically, defining the deadline completion time of the resource task includes: according to the user requirements, operation and maintenance personnel set the ending completion time of the resource task; the system automatically defines the deadline completion time for the resource task, typically 24 hours after the task creation time.
As shown in fig. 1, step S10 is to wait until all resource tasks located in different task execution channels are successfully executed, and then finish the resource task scheduling.
Referring to fig. 9, fig. 9 is a schematic structural diagram of an intelligent task scheduling system based on service efficiency optimization according to an embodiment of the present application. According to the principle of the intelligent task scheduling method based on the business efficiency optimization, the invention also provides an intelligent task scheduling system based on the business efficiency optimization, and the intelligent task scheduling system based on the business efficiency optimization comprises but is not limited to a resource task arranging device 1, a task waiting queue 2, a task delivery strategy control unit 3, a task executing device 4, a task executing strategy control unit 5, a task execution monitoring device 6, a task running queue 7, a task delivery unit 8 and a human-computer interaction interface 9.
As shown in fig. 9, the resource task orchestration device 1 is configured to orchestrate a plurality of resources according to requirements of each user, so as to obtain a plurality of resource tasks related to the user requirements and having an execution order, and initialize a current state of each resource task as a to-be-run state, specifically, the resource task orchestration device 1 may be, but is not limited to, a cloud orchestration interface. The task delivery unit 8 is configured to output all resource tasks with execution-dependent sequences to the task waiting queue 2. The task delivery strategy control unit 3 is configured to divide the task running queue 7 into a plurality of different task running channels according to different service types created by the plurality of resources; the plurality of different task operation channels include one or more of a newly-built task operation channel 71, a backup task operation channel 72, an updated task operation channel 73, and a deleted task operation channel 74. Specifically, the task execution channels may further include a binding task execution channel and a unbinding task execution channel, and may be set according to specific needs or actual application environments. Tasks in different task operation channels can be executed in parallel, are not influenced mutually and do not occupy resources mutually. The task waiting queue 2 is used for storing resource tasks. The task execution monitoring device 6 is configured to obtain a current state of each resource task, where the current state includes a to-be-run state, a running state, a suspension state, a running success state, and a running failure state.
As shown in fig. 9, the task delivery policy control unit 3 is configured to calculate a user priority weighting factor, a task waiting time weighting factor, and a task retry number weighting factor. The task delivery strategy control unit 3 is further configured to obtain a comprehensive calculation expected value of each resource task according to the user priority weight coefficient, the task waiting time weight coefficient, and the task retry number weight coefficient, and use the comprehensive calculation expected value as an execution priority of the resource task. The task delivery strategy control unit 3 is further configured to transmit the plurality of resource tasks to different task operation channels according to different service types created by the resources. The task running channel is used for storing a plurality of resource tasks according to different service types created by the resources.
As shown in fig. 9, the task execution device 4 is configured to execute the resource tasks in the descending order of the execution priority. The task execution policy control unit 5 is configured to dynamically adjust the execution priority of the resource tasks in different task operation channels in real time during the execution of the resource tasks. The task execution device 4 is further configured to remove the resource task that is successfully executed and the resource task that is failed to be executed from the task execution channel. The human-computer interaction interface 9 is used for monitoring the current state of the resource task and adjusting the current state of the resource task, and if the current state of the resource task is adjusted to be in a to-be-run state, the resource task in the current state to be in the to-be-run state is output to the task waiting queue 2 through the task execution device 4. The human-computer interaction interface 9 is a subsystem of the task execution monitoring device 6, and not only can output the current state of the resource task, but also can input a control signal into the resource task, for example, the control signal is the completion deadline of the resource task, and the human-computer interaction interface 9 may be, but not limited to, a browser. The task execution device 4 is further configured to wait until all resource tasks located in different task execution channels are successfully executed, and then finish resource task scheduling.
Referring to fig. 10, fig. 10 is a schematic structural block diagram of an electronic device according to an embodiment of the present disclosure. The invention further provides an electronic device, which comprises a processor 10 and a memory 11, wherein the memory 11 stores program instructions, and the processor 10 runs the program instructions to realize the intelligent task scheduling method based on the service efficiency optimization. The present invention also provides a computer-readable storage medium, which stores computer instructions for causing the computer to execute the above-mentioned intelligent task scheduling method based on business efficiency optimization. It should be noted that the Processor 10 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; or a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component; the Memory 11 may include a Random Access Memory (RAM), and may further include a Non-volatile Memory (Non-volatile Memory), such as at least one disk Memory. The Memory 11 may also be an internal Memory of Random Access Memory (RAM), and the processor 10 and the Memory 11 may be integrated into one or more independent circuits or hardware, such as: application Specific Integrated Circuit (ASIC). It should be noted that the computer program in the memory 11 may be implemented in the form of software functional units and may be stored in a computer readable storage medium when the computer program is sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, an electronic device, or a network device) to perform all or part of the steps of the method according to the embodiments of the present invention.
In summary, the intelligent task scheduling method based on the optimization of the service efficiency can meet the requirement of the execution efficiency of a high-level user in the task scheduling process, and can improve the overall execution efficiency of all tasks.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (9)
1. An intelligent task scheduling method based on service efficiency optimization is characterized in that the intelligent task scheduling method based on service efficiency optimization comprises the following steps:
s0, defining the current state of the resource task, wherein the current state comprises a to-be-run state, a running state, a pause state, a running success state and a running failure state;
s1, arranging a plurality of resources according to the requirements of each user to obtain a plurality of resource tasks which are related to the requirements of the user and have a dependence execution sequence, and initializing the current state of each resource task as a to-be-run state;
s2, outputting all resource tasks with the dependent execution sequence to a task waiting queue;
s3, dividing a task running queue into a plurality of different task running channels according to different service types created by a plurality of resources; the plurality of different task operation channels comprise one or more of a newly-built task operation channel, a backup task operation channel, an updated task operation channel and a deleted task operation channel;
s4, calculating a user priority weight coefficient, a task waiting time weight coefficient and a task retry number weight coefficient;
s5, obtaining the comprehensive calculation expected value of each resource task according to the user priority weight coefficient, the task waiting time weight coefficient and the task retry time weight coefficient, and taking the comprehensive calculation expected value as the execution priority of the resource task;
s6, according to different service types of resource creation, a plurality of resource tasks are transmitted to different task operation channels;
s7, executing the resource tasks according to the descending order of the execution priority, and dynamically adjusting the execution priority of a plurality of resource tasks in different task operation channels in real time in the execution process of the resource tasks;
s8, removing the resource tasks which are successfully executed and the resource tasks which are failed to be executed from the task running channel;
s9, monitoring the current state of the resource task, adjusting the current state of the resource task, and if the current state of the resource task is adjusted to be a to-be-run state, outputting the resource task with the current state to be the to-be-run state to a task waiting queue;
s10, waiting until all the resource tasks in different task running channels are successfully executed, and ending the resource task scheduling;
the step of calculating the task waiting time weight coefficient in step S4 includes:
s411, obtaining queue time for executing the resource taskPQueuing time of said resource taskPIs the time period from the resource task entering the task waiting queue to the time before the resource task is executed;
s412, calculating the average queuing time of all resource tasksP avg ；
S413, queuing time according to the resource taskPAnd average queuing time of all resource tasksP avg To obtain a task latency weight coefficientα，
；
S414, the task waiting time weight coefficientαIncluding values between 0 and 1.
2. The intelligent task scheduling method based on the business efficiency optimization according to claim 1, characterized in that: the step of calculating the user priority weight coefficient in step S4 includes:
s41, obtaining the user level according to the monthly consumption amount or the annual consumption amount of the user;
s42, obtaining a user priority weighting coefficient according to the user levelR；
S43, the user priority weighting coefficientRIncluding values between 0 and 1.
3. The intelligent task scheduling method based on the business efficiency optimization according to claim 1, characterized in that: the step of calculating a task retry number weighting factor in step S4 includes:
s4111, obtaining retry times of executing the resource taskNNumber of retries of said resource taskNRetry execution times after the execution of the resource task fails;
s4112, calculating average retry times of all resource tasksN avg ；
S4113, retrying times according to the resource taskNAnd average number of retries for all resource tasksN avg To obtain the task retry number weight coefficientβ，
；
S4114, weighting coefficient of the number of task retriesβIncluding values between 0 and 1.
4. The intelligent task scheduling method based on service efficiency optimization according to any one of claims 1 to 3, wherein the step of obtaining the comprehensive calculated expected value of each resource task according to the user priority weight coefficient, the task waiting time weight coefficient and the task retry number weight coefficient in step S5 includes:
based on a comprehensive calculationExpected value formula:V=R×（α+β) Wherein, in the step (A),
，
then, the expected value is calculated comprehensively
Wherein, in the step (A),Rrepresents a user priority weight coefficient and,αa weight coefficient representing the task latency is represented,βa weight coefficient indicating the number of task retries,Pindicating the queuing time of the resource task,P avg representing the average queuing time of all resource tasks,Nindicating the number of retries of the resource task,N avg indicating the average number of retries for all resource tasks.
5. The intelligent task scheduling method based on business efficiency optimization according to claim 1, wherein the step of monitoring the current state of the resource task and adjusting the current state of the resource task in step S9 includes:
s91, monitoring the current state of the resource task through a human-computer interaction interface;
and S92, adjusting the current state of the resource task through the human-computer interaction interface.
6. The intelligent task scheduling method based on business efficiency optimization of claim 5, wherein the step of adjusting the current state of the resource task through the human-computer interface in step S92 comprises:
defining an expiration completion time for the resource task;
adjusting the current state of the resource task from a to-be-run state to a pause state;
adjusting the current state of the resource task from a pause state to a to-be-run state;
adjusting the current state of the resource task from a running failure state to a to-be-run state;
and adjusting the current state of the resource task from the running success state to the to-be-run state.
7. The intelligent task scheduling method based on the business efficiency optimization according to claim 1, characterized in that: the sequential execution of the resource tasks in step S7 is to execute the resource tasks in different task execution channels in parallel and sequentially.
8. An intelligent task scheduling system based on business efficiency optimization, which is characterized in that the intelligent task scheduling system based on business efficiency optimization comprises:
the resource task arranging equipment is used for arranging a plurality of resources according to the requirements of each user to obtain a plurality of resource tasks which are related to the requirements of the user and have a dependence execution sequence, and initializing the current state of each resource task as a to-be-run state;
the task delivery unit is used for outputting all resource tasks with the execution-dependent sequence to the task waiting queue;
the task delivery strategy control unit is used for dividing a task running queue into a plurality of different task running channels according to different service types established by a plurality of resources; the plurality of different task operation channels comprise one or more of a newly-built task operation channel, a backup task operation channel, an updated task operation channel and a deleted task operation channel;
the task delivery strategy control unit is used for calculating a user priority weight coefficient, a task waiting time weight coefficient and a task retry frequency weight coefficient;
the task delivery strategy control unit is used for obtaining the comprehensive calculation expected value of each resource task according to the user priority weight coefficient, the task waiting time weight coefficient and the task retry frequency weight coefficient, and taking the comprehensive calculation expected value as the execution priority of the resource tasks;
the task delivery strategy control unit is used for transmitting a plurality of resource tasks to different task operation channels according to different service types established by the resources;
the task execution equipment is used for executing the resource tasks according to the descending order of the execution priority;
the task execution strategy control unit is used for dynamically adjusting the execution priority of a plurality of resource tasks in different task operation channels in real time in the execution process of the resource tasks;
the task execution equipment is used for removing the resource tasks which are successfully executed and the resource tasks which are failed to be executed from the task operation channel;
the human-computer interaction interface is used for monitoring the current state of the resource task and adjusting the current state of the resource task;
the task execution device is used for outputting the resource task of which the current state is the to-be-operated state to a task waiting queue if the current state of the resource task is adjusted to the to-be-operated state;
and the task execution equipment is used for finishing the resource task scheduling after waiting until all the resource tasks in the different task operation channels are successfully executed.
9. An electronic device comprising a processor and a memory, the memory storing program instructions, characterized in that: the processor executes program instructions to realize the intelligent task scheduling method based on the business efficiency optimization according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010623617.4A CN111506413B (en) | 2020-07-02 | 2020-07-02 | Intelligent task scheduling method and system based on business efficiency optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010623617.4A CN111506413B (en) | 2020-07-02 | 2020-07-02 | Intelligent task scheduling method and system based on business efficiency optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111506413A CN111506413A (en) | 2020-08-07 |
| CN111506413B true CN111506413B (en) | 2020-09-18 |
Family
ID=71870652
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010623617.4A Active CN111506413B (en) | 2020-07-02 | 2020-07-02 | Intelligent task scheduling method and system based on business efficiency optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111506413B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112667376A (en) * | 2020-12-23 | 2021-04-16 | 数字广东网络建设有限公司 | Task scheduling processing method and device, computer equipment and storage medium |
| CN113434291A (en) * | 2021-06-25 | 2021-09-24 | 湖北央中巨石信息技术有限公司 | Real-time scheduling optimization method based on channel |
| CN113835861A (en) * | 2021-09-24 | 2021-12-24 | 中汽创智科技有限公司 | Process scheduling method, device, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947552A (en) * | 2019-03-28 | 2019-06-28 | 南京邮电大学 | Margin control dynamic task scheduling method based on process and thread scheduling |
| CN110570011A (en) * | 2019-08-02 | 2019-12-13 | 中国船舶工业系统工程研究院 | Complex system resource optimization scheduling method and system under multi-constraint condition |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5469571A (en) * | 1991-07-15 | 1995-11-21 | Lynx Real-Time Systems, Inc. | Operating system architecture using multiple priority light weight kernel task based interrupt handling |
| CN107918806B (en) * | 2017-11-13 | 2021-01-26 | 浙江大学 | Intelligent optimal scheduling method |
| CN109933617B (en) * | 2019-03-08 | 2021-05-25 | 恒生电子股份有限公司 | Data processing method, data processing device, related equipment and related medium |
| CN110688229B (en) * | 2019-10-12 | 2022-08-02 | 阿波罗智能技术(北京)有限公司 | Task processing method and device |
-
2020
- 2020-07-02 CN CN202010623617.4A patent/CN111506413B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947552A (en) * | 2019-03-28 | 2019-06-28 | 南京邮电大学 | Margin control dynamic task scheduling method based on process and thread scheduling |
| CN110570011A (en) * | 2019-08-02 | 2019-12-13 | 中国船舶工业系统工程研究院 | Complex system resource optimization scheduling method and system under multi-constraint condition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111506413A (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111506413B (en) | Intelligent task scheduling method and system based on business efficiency optimization | |
| CN106790726B (en) | Priority queue dynamic feedback load balancing resource scheduling method based on Docker cloud platform | |
| CN109491790B (en) | Container-based industrial Internet of things edge computing resource allocation method and system | |
| EP3770774B1 (en) | Control method for household appliance, and household appliance | |
| Ghodsi et al. | Choosy: Max-min fair sharing for datacenter jobs with constraints | |
| US10037230B2 (en) | Managing data processing resources | |
| CN109117265A (en) | The method, apparatus, equipment and storage medium of schedule job in the cluster | |
| CN106874084B (en) | Distributed workflow scheduling method and device and computer equipment | |
| US20200174844A1 (en) | System and method for resource partitioning in distributed computing | |
| US20180052714A1 (en) | Optimized resource metering in a multi tenanted distributed file system | |
| CN111381950A (en) | Task scheduling method and system based on multiple copies for edge computing environment | |
| CN109783225B (en) | Tenant priority management method and system of multi-tenant big data platform | |
| Wang et al. | Job scheduling for large-scale machine learning clusters | |
| Li et al. | Enabling elastic stream processing in shared clusters | |
| CN111209091A (en) | Scheduling method of Spark task containing private data in mixed cloud environment | |
| CN111459684A (en) | Cloud computing resource fusion scheduling management method, system and medium for multiprocessor architecture | |
| Perret et al. | A deadline scheduler for jobs in distributed systems | |
| US10845997B2 (en) | Job manager for deploying a bundled application | |
| CN115033357A (en) | Micro-service workflow scheduling method and device based on dynamic resource selection strategy | |
| WO2018157768A1 (en) | Method and device for scheduling running device, and running device | |
| CN114327894A (en) | Resource allocation method, device, electronic equipment and storage medium | |
| CN116915869A (en) | Cloud edge cooperation-based time delay sensitive intelligent service quick response method | |
| CN117112199A (en) | Multi-tenant resource scheduling method, device and storage medium | |
| CN108429704B (en) | Node resource allocation method and device | |
| KR20130022707A (en) | Periodic and aperiodic task scheduling algorithm based on topological sort and residual time |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |