CN113434267A - Cloud computing workflow dynamic scheduling method, device, equipment and storage medium - Google Patents

Cloud computing workflow dynamic scheduling method, device, equipment and storage medium Download PDF

Info

Publication number
CN113434267A
CN113434267A CN202110569976.0A CN202110569976A CN113434267A CN 113434267 A CN113434267 A CN 113434267A CN 202110569976 A CN202110569976 A CN 202110569976A CN 113434267 A CN113434267 A CN 113434267A
Authority
CN
China
Prior art keywords
cloud computing
population
objective function
preset
archive
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.)
Granted
Application number
CN202110569976.0A
Other languages
Chinese (zh)
Other versions
CN113434267B (en
Inventor
梁正平
王志强
李辉才
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen University
Original Assignee
Shenzhen University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen University filed Critical Shenzhen University
Priority to CN202110569976.0A priority Critical patent/CN113434267B/en
Publication of CN113434267A publication Critical patent/CN113434267A/en
Application granted granted Critical
Publication of CN113434267B publication Critical patent/CN113434267B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • 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
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5072—Grid computing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/48—Indexing scheme relating to G06F9/48
    • G06F2209/485—Resource constraint

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention is applicable to the technical field of computers, and provides a method, a device, equipment and a storage medium for dynamically scheduling a cloud computing workflow, wherein the method comprises the following steps: the method comprises the steps of constructing a multi-objective function of a workflow scheduling optimization target of cloud computing according to a received multi-task processing request, obtaining available resource data of the cloud computing and constraint conditions of task processing preset in the cloud computing, obtaining an optimal solution set of the multi-objective function through a self-adaptive variable response dynamic multi-objective evolutionary algorithm according to the available resource data, the constraint conditions and the multi-objective function, and setting the optimal solution set as an execution sequence of multiple tasks, so that the workflow dynamic scheduling efficiency of the cloud computing is improved.

Description

Cloud computing workflow dynamic scheduling method, device, equipment and storage medium
Technical Field
The invention belongs to the technical field of computers, and particularly relates to a method, a device, equipment and a storage medium for dynamically scheduling a cloud computing workflow.
Background
Cloud computing is a large-scale heterogeneous and distributed computing infrastructure oriented to the scientific and commercial communities that provides high-quality and low-cost services with minimal hardware investment. Infrastructure as a service (IaaS), platform as a service (PaaS), and software as a service (SaaS) provide services through cloud computing, and users can access resources deployed on the cloud through IaaS, PaaS, and SaaS.
Workflow is a common technique for building large-scale computing and data-intensive applications from different research areas, application workflows are modeled using directed acyclic graphs, where the nodes of the graph are tasks interconnected by computing or data resources, and the workflow scheduling problem in cloud computing aims at mapping the tasks of a given application onto available resources. In the debugging of workflow schedules, the dynamic concept of workflow schedules includes two aspects, the first being the occurrence of new task requests over time, or transient resource failures, where resources may dynamically join or leave the cloud, which may be caused by various events, such as software failures (errors, overflows, etc.) or hardware failures (electrical power irregularities, hard disk failures, etc.). Another aspect is changing the number of targets during execution of the workflow, a real-world scenario faced by cloud computing, where the number of targets may change over time. For example, the completion time of a workflow may not be considered before submitting the workflow with a shorter deadline for execution. The dynamic property of workflow scheduling can affect the selection of the optimal workflow plan solution in different periods, and the existing workflow scheduling method has poor workflow scheduling efficiency due to the fact that the dynamic property in the workflow scheduling process cannot be effectively considered.
Disclosure of Invention
The invention aims to provide a method, a device, equipment and a storage medium for dynamically scheduling a cloud computing workflow, and aims to solve the problem of poor efficiency of the existing cloud computing workflow.
In one aspect, the present invention provides a method for dynamically scheduling a cloud computing workflow, where the method includes:
constructing a multi-objective function of a workflow scheduling optimization target of cloud computing according to the received multi-task processing request;
acquiring available resource data of the cloud computing and constraint conditions of task processing preset in the cloud computing;
and acquiring an optimal solution set of the multi-objective function through a dynamic multi-objective evolutionary algorithm of self-adaptive change response according to the available resource data, the constraint conditions and the multi-objective function, and setting the optimal solution set as a multi-task execution sequence.
In another aspect, the present invention provides a cloud computing workflow dynamic scheduling apparatus, including:
the function construction unit is used for constructing a multi-objective function of a workflow scheduling optimization target of cloud computing according to the received multi-task processing request;
the constraint obtaining unit is used for obtaining available resource data of the cloud computing and constraint conditions of task processing preset in the cloud computing; and
and the execution setting unit is used for acquiring an optimal solution set of the multi-objective function through a dynamic multi-objective evolutionary algorithm of self-adaptive change response according to the available resource data, the constraint condition and the multi-objective function, and setting the optimal solution set as a multi-task execution sequence.
In another aspect, the present invention also provides a computing device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method as described above when executing the computer program.
In another aspect, the present invention also provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps of the method as described above.
According to the method, the multi-objective function of the workflow scheduling optimization target of the cloud computing is constructed according to the received multi-task processing request, the available resource data of the cloud computing and the constraint condition of task processing preset in the cloud computing are obtained, the optimal solution set of the multi-objective function is obtained through the dynamic multi-objective evolutionary algorithm of the adaptive change response according to the available resource data, the constraint condition and the multi-objective function, the optimal solution set is set to the execution sequence of the multi-tasks, the multi-tasks are scheduled through the dynamic multi-objective evolutionary algorithm of the adaptive change response, and the dynamic scheduling efficiency of the workflow of the cloud computing is improved.
Drawings
Fig. 1 is a flowchart illustrating an implementation of a method for dynamically scheduling cloud computing workflows according to an embodiment of the present invention;
fig. 2 is a flowchart of an implementation of a cloud computing workflow dynamic scheduling method according to a second embodiment of the present invention;
fig. 3 is a schematic flow chart illustrating an implementation process of obtaining an optimal solution set of a multi-objective function in a cloud computing workflow dynamic scheduling method according to a second embodiment of the present invention;
fig. 4 is a schematic structural diagram of a cloud computing workflow dynamic scheduling apparatus according to a third embodiment of the present invention; and
fig. 5 is a schematic structural diagram of a computing device according to a sixth embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following detailed description of specific implementations of the present invention is provided in conjunction with specific embodiments:
the first embodiment is as follows:
fig. 1 shows an implementation flow of a cloud computing workflow dynamic scheduling method according to a first embodiment of the present invention, and for convenience of description, only parts related to the embodiment of the present invention are shown, which are detailed as follows:
in step S101, according to the received multitask processing request, a multi-objective function of a workflow scheduling optimization objective of cloud computing is constructed;
the embodiment of the invention is suitable for cloud computing, and is used for scheduling and processing multiple tasks on the cloud computing, so that the efficient processing of the multiple tasks of the cloud computing is realized. These tasks may be, by way of example, gravity wave analysis in physics, financial data analysis, weather forecasting in meteorology, and storm surge prediction. These all can rationally arrange task scheduling flow in cloud computing to make efficient use of cloud computing resources.
Specifically, the multitasking request received by the cloud computing can include a plurality of optimization objectives, and the optimization objectives can be to minimize computing time, cost, energy and unbalance degree and to maximize reliability and utilization rate.
In a preferred embodiment, the multi-objective function of the workflow scheduling optimization objective is
Figure BDA0003082289700000041
Wherein m represents the number of the workflow scheduling optimization targets, the value of m is 2, omega represents the available resource data set of cloud computing, f1(x) Representing the time at which the cloud computing processes all tasks, f2(x) The method has the advantages that the computing cost of completing all tasks by cloud computing is shown, x is a decision variable, and the decision variable is determined by the task parameter of each task, the available resource data of the cloud computing and the task execution sequence, so that the reliability and the utilization rate of the cloud computing are improved.
In step S102, available resource data of cloud computing and a constraint condition of task processing preset in the cloud computing are acquired;
in the embodiment of the present invention, the available resource data of the cloud computing includes the number of available CPUs of the cloud computing, an available storage space, an available time period of available resources, and the like, and the constraint conditions of task processing preset in the cloud computing are used to define or constrain each processing task when an optimization goal of minimizing time and cost is achieved. For example, the number of resources available, priority constraints between tasks, a task can only be scheduled to one cloud resource, a task cannot be interrupted once started, etc.
In step S103, an optimal solution set of the multi-objective functions is obtained through a dynamic multi-objective evolutionary algorithm of adaptive change response according to the available resource data, the constraint conditions, and the multi-objective functions, and the optimal solution set is set as an execution sequence of multiple tasks.
In the embodiment of the invention, the optimal solution set is a set of optimal solutions of the multi-objective function, wherein the solutions are feasible schemes for task scheduling in cloud computing, for example, the execution sequence of the multiple tasks can enable the cloud computing to use the least resources and the least time cost, and process all tasks with the highest efficiency as possible, so that a user can select one scheme better conforming to the actual situation to realize the scheduling processing of the multiple tasks according to the actual situation.
According to the embodiment of the invention, a multi-objective function of a workflow scheduling optimization target of cloud computing is constructed according to a received multi-task processing request, available resource data of the cloud computing and a constraint condition of task processing preset in the cloud computing are obtained, an optimal solution set of the multi-objective function is obtained through a dynamic multi-objective evolutionary algorithm of adaptive change response according to the available resource data, the constraint condition and the multi-objective function, the optimal solution set is set to an execution sequence of multi-tasks, so that the multi-tasks are scheduled through the dynamic multi-objective evolutionary algorithm of adaptive change response, and the dynamic scheduling efficiency of the workflow of the cloud computing is improved.
Example two:
fig. 2 shows an implementation process of obtaining an optimal solution set of a multi-objective function in a cloud computing workflow dynamic scheduling method provided by the second embodiment of the present invention, and for convenience of description, only parts related to the second embodiment of the present invention are shown, and details are as follows:
in step S201, an initial population of the multi-objective function is randomly generated according to the available resource data and the constraint conditions, and the length of each individual in the initial population is determined according to the number of tasks;
when the optimal solution set of the multi-objective function is obtained, an initial population of the multi-objective function is randomly generated according to available resource data and constraint conditions, the length of each individual in the initial population is determined according to the number of tasks, for example, when the number of the tasks is 2, the relative execution sequence of the tasks is stored in the first dimension of the population, and the mapping from the tasks to the resources is stored in the second dimension.
In step S202, non-dominant individuals in the initial population are screened according to the Pareto dominant relationship, and the non-dominant individuals are stored in a preset archive;
in the embodiment of the invention, the evolution direction of the population indicates the search direction and the search step length, and the preset archive is used for storing individuals with better convergence and diversity in the process of acquiring the optimal solution set of the multi-target function, so that the convergence and diversity of the solution in the optimal solution set are ensured.
In step S203, detecting whether the environment of the cloud computing is changed, if so, executing step S204, otherwise, executing step S205;
in step S204, when the cloud computing environment changes, updating the population and the preset archive by using a corresponding change response policy according to a difference between corresponding individual decision variables in the optimal population of the multi-objective function at the current time and the previous time;
in the embodiment of the invention, when the environment of cloud computing changes, that is, the cloud computing receives a new task processing request, the population and the preset archive are updated by using the corresponding change response strategy according to the difference between the corresponding individual decision variables in the optimal population of the multi-objective function at the current moment and the previous moment.
In a preferred embodiment, when updating the population and the preset archive by using the corresponding change strategy according to the difference between the corresponding individual decision variables in the optimal population of the multi-target function at the current time and the previous time, if the difference value between one individual decision variable in the optimal population of the multi-target function at the current time and the previous time is smaller than the dimension decision variable difference average value, performing variation operation on the individuals in the population, updating the preset archive, and if the difference value between one individual decision variable in the optimal population of the multi-target function at the current time and the previous time is larger than the dimension decision variable difference average value, using a formula
Figure BDA0003082289700000061
Reinitializing individuals in the population and updating a preset archive, wherein pi,jA decision variable in dimension j representing an ith individual in the population at a previous time,
Figure BDA0003082289700000062
represents a new startJ-th dimension decision variable, ac, of ith individual in the initialized populationjAnd
Figure BDA0003082289700000063
presetting the j-th dimension decision variable value, d, of the archiving center for archiving respectively at the current moment and the previous momentp_aci,jAnd
Figure BDA0003082289700000064
respectively p on the j-th decision variableiAnd
Figure BDA0003082289700000065
to respective corresponding archiving centres ac and aclGaussian is a gaussian interference function. Therefore, the variation strategy or the prediction strategy is selected in a self-adaptive mode based on the difference of the individual decision variables in each population, on one hand, the pertinence of change response can be improved, on the other hand, the balance of convergence and diversity can be promoted, and therefore the stability and the reliability of the subsequently obtained workflow dynamic scheduling scheme are improved.
In addition, acjIs calculated by the formula
Figure BDA0003082289700000066
Wherein | A | is the archive size, since the optimal population P is at time ttAnd the optimal population P at time t-1t-1The variation trend of the non-dominant individual set is favorable for predicting the distribution position of the population optimal solution set after the environment changes, and the most representative individual in the archive storing the non-dominant individuals is the archive center ac, so that when the environment changes, the archive center ac and ac are based on the time t and the time t-1lThe method is beneficial to the reinitialized population to approach to a Pareto optimal solution set (PS) in a new environment, so that the convergence of the population is improved, the prediction quality is improved, and the efficiency of cloud computing for processing all tasks is finally improved.
By way of example, fig. 3 shows that j-th dimension decision variables of two different individuals are adaptively changed and responded by a mutation strategy and a prediction strategy. Let xjAnd
Figure BDA0003082289700000071
respectively as the optimal population P at the time ttAnd the optimal population P at time t-1t-1Value distribution, p, of decision variables in the j-th dimensioni,jAnd pk,jAre respectively PtThe values of the decision variables of the ith and the kth individual dimension j,
Figure BDA0003082289700000072
and
Figure BDA0003082289700000073
are respectively Pt-1And (5) taking values of decision variables of the ith dimension and the kth dimension of the individual. As can be seen from the figure, pk,jAnd
Figure BDA0003082289700000074
is less than xjAnd
Figure BDA0003082289700000075
so that a variation strategy is adopted to generate a corresponding reinitialization value
Figure BDA0003082289700000076
And p isi,jAnd
Figure BDA0003082289700000077
is greater than xjAnd
Figure BDA0003082289700000078
so that a prediction strategy is adopted to generate a corresponding reinitialization value
Figure BDA0003082289700000079
In step S205, the population and the preset archive are evolved;
in the embodiment of the present invention, after the population and the preset archive are updated by using the corresponding change response policy, the population and the preset archive may be evolved, for example, using a static optimization framework to generate offspring, and then obtaining a next generation population through environment selection, specifically, using a preset static multi-objective evolution algorithm to evolve the population and the individuals in the preset archive. Of course, since the population and the preset archive are updated by using the corresponding change response policy, the population and the preset archive may not be evolved any more, and the step S206 is directly performed. When step S203 detects that the environment of the cloud computing is not changed, the population and the preset archive are directly evolved, for example, offspring are generated by using a static optimization framework, and then a next generation population is obtained through environment selection.
In step S206, it is determined whether the current iteration generation is smaller than a preset maximum iteration generation, if so, step S207 is executed, otherwise, step S208 is executed;
in step S207, if yes, updating the individuals in the population and the preset archive, adding 1 to the current iteration algebra, and skipping to the step of detecting whether the environment of cloud computing changes;
in step S208, otherwise, the individuals in the preset archive are determined to be the optimal solution set of the multi-objective function.
In the embodiment of the invention, when the optimal solution set of the multi-target function is obtained, the population and the preset archive are updated by adaptively using the variation and predicted variation response strategy, so that the optimal population with better balanced convergence and diversity is obtained, and the stability and the effectiveness of the multi-task scheduling processing scheme of cloud computing are ensured.
Example three:
fig. 4 shows a structure of a cloud computing workflow dynamic scheduling apparatus according to a third embodiment of the present invention, and for convenience of description, only a part related to the third embodiment of the present invention is shown, where the structure includes:
a function constructing unit 41, configured to construct a multi-objective function of a workflow scheduling optimization target of cloud computing according to the received multi-task processing request;
a constraint obtaining unit 42, configured to obtain available resource data of the cloud computing and a constraint condition of task processing preset in the cloud computing; and
and the execution setting unit 43 is configured to obtain an optimal solution set of the multi-objective functions through a dynamic multi-objective evolutionary algorithm of adaptive change response according to the available resource data, the constraint conditions and the multi-objective functions, and set the optimal solution set as an execution sequence of multiple tasks.
In a preferred embodiment, the execution setting unit 43 includes:
the population generating unit is used for randomly generating an initial population of the multi-objective function according to the available resource data and the constraint conditions, and the length of each individual in the initial population is determined according to the number of tasks;
the archive acquisition unit is used for screening the non-dominant individuals in the initial population according to the Pareto dominant relationship and storing the non-dominant individuals into a preset archive;
the population updating unit is used for detecting whether the environment of cloud computing changes or not, and if yes, updating the population and the preset archive by using a corresponding change response strategy according to the difference between corresponding individual decision variables in the optimal population of the multi-target function at the current moment and the previous moment; and
and the solution set determining unit is used for judging whether the current iteration algebra is smaller than a preset maximum iteration algebra or not, if so, updating the individuals in the population and a preset archive, adding 1 to the current iteration algebra, and jumping to the step of detecting whether the environment of cloud computing changes or not, otherwise, determining the individuals in the preset archive as the optimal solution set of the multi-objective function.
In this embodiment of the present invention, each unit of the cloud computing workflow dynamic scheduling apparatus may be implemented by a corresponding hardware or software unit, and each unit may be an independent software or hardware unit, or may be integrated into a software or hardware unit, which is not limited to this disclosure, and specific embodiments of each unit may refer to the description of the foregoing method embodiment, and are not described herein again.
Example four:
fig. 5 shows a structure of a computing device according to a fourth embodiment of the present invention, and for convenience of explanation, only a part related to the embodiment of the present invention is shown.
The computing device 5 of an embodiment of the invention comprises a processor 50, a memory 51 and a computer program 52 stored in the memory 51 and executable on the processor 50. The processor 50, when executing the computer program 52, implements the steps in the various method embodiments described above, such as the steps S101 to S103 shown in fig. 1. Alternatively, the processor 50, when executing the computer program 52, implements the functionality of the various units in the above-described apparatus embodiments, such as the functionality of the units 41 to 43 shown in fig. 4.
The computing device of embodiments of the present invention may be a server. The steps implemented when the processor 50 executes the computer program 52 in the computing device 5 to implement the above method can refer to the description of the foregoing method embodiments, and are not described herein again.
Example five:
in an embodiment of the present invention, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps in the above-described method embodiment, for example, steps S101 to S103 shown in fig. 1. Alternatively, the computer program realizes the functions of the units in the above-described apparatus embodiments, such as the functions of the units 41 to 43 shown in fig. 4, when executed by the processor.
According to the embodiment of the invention, a multi-objective function of a workflow scheduling optimization target of cloud computing is constructed according to a received multi-task processing request, available resource data of the cloud computing and a constraint condition of task processing preset in the cloud computing are obtained, an optimal solution set of the multi-objective function is obtained through a dynamic multi-objective evolutionary algorithm of adaptive change response according to the available resource data, the constraint condition and the multi-objective function, the optimal solution set is set to an execution sequence of multi-tasks, so that the multi-tasks are scheduled through the dynamic multi-objective evolutionary algorithm of adaptive change response, and the dynamic scheduling efficiency of the workflow of the cloud computing is improved.
The computer readable storage medium of the embodiments of the present invention may include any entity or device capable of carrying computer program code, a recording medium, such as a ROM/RAM, a magnetic disk, an optical disk, a flash memory, or the like.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims (10)

1. A method for dynamically scheduling a cloud computing workflow, the method comprising:
constructing a multi-objective function of a workflow scheduling optimization target of cloud computing according to the received multi-task processing request;
acquiring available resource data of the cloud computing and constraint conditions of task processing preset in the cloud computing;
and acquiring an optimal solution set of the multi-objective function through a dynamic multi-objective evolutionary algorithm of self-adaptive change response according to the available resource data, the constraint conditions and the multi-objective function, and setting the optimal solution set as a multi-task execution sequence.
2. The method of claim 1, wherein the multi-objective function is
Figure FDA0003082289690000011
Wherein m represents the number of the workflow scheduling optimization targets, the value of m is 2, Ω represents the available resource data set of the cloud computing, f1(x) Representing the time at which the cloud computing processes all tasks, f2(x) The method comprises the steps of representing the computing cost of completing all tasks by cloud computing, and x represents a decision variable, wherein the decision variable is determined by task parameters of each task, available resource data of the cloud computing and a task execution sequence.
3. The method of claim 1, wherein the step of obtaining the optimal solution set of the multi-objective function through a dynamic multi-objective evolutionary algorithm with adaptive change response comprises:
randomly generating an initial population of the multi-objective function according to the available resource data and the constraint conditions, wherein the length of each individual in the initial population is determined according to the number of tasks;
screening non-dominant individuals in the initial population according to a Pareto dominant relationship, and storing the non-dominant individuals in a preset archive;
detecting whether the environment of the cloud computing is changed, if so, updating the population and the preset archive files by using corresponding change response strategies according to the difference between corresponding individual decision variables in the optimal population of the multi-objective function at the current moment and the previous moment;
judging whether the current iteration algebra is smaller than a preset maximum iteration algebra, if so, updating the individuals in the population and the preset archive, adding 1 to the current iteration algebra, and skipping to the step of detecting whether the environment of the cloud computing is changed, otherwise, determining the individuals in the preset archive as the optimal solution set of the multi-target function.
4. The method of claim 3, wherein the step of updating the population and the preset archive using corresponding change response strategies based on differences between corresponding individual decision variables in the optimal population of the multi-objective function at a current time and a previous time comprises:
if the difference value between one body decision variable in the optimal population of the multi-target function at the current moment and the previous moment is smaller than the average value of the difference of the dimensional decision variables, carrying out variation operation on the individuals in the population and updating the preset archive;
if the difference value between one body in the optimal population of the multi-target function at the current moment and the previous moment is larger than the difference average value, a formula is used
Figure FDA0003082289690000021
Reinitializing individuals in the population and updating the preset archive, whichIn, pi,jA decision variable in dimension j representing an ith individual in the population at a previous time,
Figure FDA0003082289690000022
a j-dimension decision variable, ac, representing the i-th individual in the population after re-initializationjAnd
Figure FDA0003082289690000023
respectively representing the values of the j-th dimension decision variable of the preset archiving center at the current moment and the previous moment,
Figure FDA0003082289690000024
and
Figure FDA0003082289690000025
respectively p on the j-th decision variableiAnd
Figure FDA0003082289690000026
to respective corresponding archiving centres ac and aclGaussian is a gaussian interference function.
5. The method of claim 3, wherein after the step of updating the population and the preset archive using the corresponding change response policy, and before the step of determining whether the current iteration generation is less than a preset maximum iteration generation, the method further comprises:
and evolving the population and the preset archive.
6. The method of claim 3, wherein the optimal solution set of the multi-objective function is obtained by a dynamic multi-objective evolutionary algorithm with adaptive change response, further comprising:
and when detecting that the environment of the cloud computing is not changed, evolving the population and the preset archive.
7. An apparatus for dynamically scheduling cloud computing workflows, the apparatus comprising:
the function construction unit is used for constructing a multi-objective function of a workflow scheduling optimization target of cloud computing according to the received multi-task processing request;
the constraint obtaining unit is used for obtaining available resource data of the cloud computing and constraint conditions of task processing preset in the cloud computing; and
and the execution setting unit is used for acquiring an optimal solution set of the multi-objective function through a dynamic multi-objective evolutionary algorithm of self-adaptive change response according to the available resource data, the constraint condition and the multi-objective function, and setting the optimal solution set as a multi-task execution sequence.
8. The apparatus of claim 7, wherein the execution setting unit comprises:
a population generating unit, configured to randomly generate an initial population of the multi-objective function according to the available resource data and the constraint condition, where a length of each individual in the initial population is determined according to a task number;
the archive acquisition unit is used for screening the non-dominant individuals in the initial population according to the Pareto dominant relationship and storing the non-dominant individuals into a preset archive;
a population updating unit, configured to detect whether the cloud computing environment changes, and if so, update the population and the preset archive by using a corresponding change response policy according to a difference between corresponding individual decision variables in the optimal population of the multi-objective function at a current time and a previous time; and
and the solution set determining unit is used for judging whether the current iteration algebra is smaller than a preset maximum iteration algebra or not, if so, updating the individuals in the population and the preset archive, adding 1 to the current iteration algebra, skipping to the step of detecting whether the environment of the cloud computing is changed or not, and otherwise, determining the individuals in the preset archive as the optimal solution set of the multi-objective function.
9. A cloud computing device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the steps of the method according to any of claims 1 to 6 are implemented when the computer program is executed by the processor.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.
CN202110569976.0A 2021-05-25 2021-05-25 Cloud computing workflow dynamic scheduling method, device, equipment and storage medium Active CN113434267B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110569976.0A CN113434267B (en) 2021-05-25 2021-05-25 Cloud computing workflow dynamic scheduling method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110569976.0A CN113434267B (en) 2021-05-25 2021-05-25 Cloud computing workflow dynamic scheduling method, device, equipment and storage medium

Publications (2)

Publication Number Publication Date
CN113434267A true CN113434267A (en) 2021-09-24
CN113434267B CN113434267B (en) 2022-12-02

Family

ID=77802846

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110569976.0A Active CN113434267B (en) 2021-05-25 2021-05-25 Cloud computing workflow dynamic scheduling method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN113434267B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114912654A (en) * 2022-03-31 2022-08-16 北京百度网讯科技有限公司 Object scheduling method, device, equipment and storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102708011A (en) * 2012-05-11 2012-10-03 南京邮电大学 Multistage load estimating method facing task scheduling of cloud computing platform
CN108133260A (en) * 2018-01-17 2018-06-08 浙江理工大学 The workflow schedule method of multi-objective particle swarm optimization based on real-time status monitoring
CN108320059A (en) * 2018-02-22 2018-07-24 石家庄铁道大学 A kind of workflow schedule evolution optimization method and terminal device
CN108521446A (en) * 2018-03-20 2018-09-11 深圳大学 Dispatching method, device, equipment and the storage medium of cloud computing resources
CN109298930A (en) * 2017-07-24 2019-02-01 西安电子科技大学昆山创新研究院 A kind of cloud workflow schedule method and device based on multiple-objection optimization
US20190266534A1 (en) * 2016-07-20 2019-08-29 Worldline Multi-criteria adaptive scheduling method for a market-oriented hybrid cloud infrastructure
CN111027736A (en) * 2019-07-30 2020-04-17 中国人民解放军国防科技大学 Micro-service combined deployment and scheduling method under multi-objective optimization
US20210064432A1 (en) * 2018-02-05 2021-03-04 Telefonaktiebolaget Lm Ericsson (Publ) Resource needs prediction in virtualized systems: generic proactive and self-adaptive solution
US20210133534A1 (en) * 2019-10-30 2021-05-06 Inner Mongolia Agricultural University Cloud task scheduling method based on phagocytosis-based hybrid particle swarm optimization and genetic algorithm

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102708011A (en) * 2012-05-11 2012-10-03 南京邮电大学 Multistage load estimating method facing task scheduling of cloud computing platform
US20190266534A1 (en) * 2016-07-20 2019-08-29 Worldline Multi-criteria adaptive scheduling method for a market-oriented hybrid cloud infrastructure
CN109298930A (en) * 2017-07-24 2019-02-01 西安电子科技大学昆山创新研究院 A kind of cloud workflow schedule method and device based on multiple-objection optimization
CN108133260A (en) * 2018-01-17 2018-06-08 浙江理工大学 The workflow schedule method of multi-objective particle swarm optimization based on real-time status monitoring
US20210064432A1 (en) * 2018-02-05 2021-03-04 Telefonaktiebolaget Lm Ericsson (Publ) Resource needs prediction in virtualized systems: generic proactive and self-adaptive solution
CN108320059A (en) * 2018-02-22 2018-07-24 石家庄铁道大学 A kind of workflow schedule evolution optimization method and terminal device
CN108521446A (en) * 2018-03-20 2018-09-11 深圳大学 Dispatching method, device, equipment and the storage medium of cloud computing resources
CN111027736A (en) * 2019-07-30 2020-04-17 中国人民解放军国防科技大学 Micro-service combined deployment and scheduling method under multi-objective optimization
US20210133534A1 (en) * 2019-10-30 2021-05-06 Inner Mongolia Agricultural University Cloud task scheduling method based on phagocytosis-based hybrid particle swarm optimization and genetic algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
史振华: "基于改进的鲸鱼算法在云计算任务调度中的研究", 《科技通报》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114912654A (en) * 2022-03-31 2022-08-16 北京百度网讯科技有限公司 Object scheduling method, device, equipment and storage medium

Also Published As

Publication number Publication date
CN113434267B (en) 2022-12-02

Similar Documents

Publication Publication Date Title
Bhattacharjee et al. Barista: Efficient and scalable serverless serving system for deep learning prediction services
Hosseinzadeh et al. Multi-objective task and workflow scheduling approaches in cloud computing: a comprehensive review
Rahman et al. A dynamic critical path algorithm for scheduling scientific workflow applications on global grids
Liu et al. Deadline‐constrained coevolutionary genetic algorithm for scientific workflow scheduling in cloud computing
Shi et al. MDP and machine learning-based cost-optimization of dynamic resource allocation for network function virtualization
Zuo et al. A multi-objective hybrid cloud resource scheduling method based on deadline and cost constraints
Han et al. Tailored learning-based scheduling for kubernetes-oriented edge-cloud system
CN108133260B (en) Real-time state monitoring-based multi-target particle swarm optimization workflow scheduling method
CN109634742B (en) Time constraint scientific workflow optimization method based on ant colony algorithm
CN107656799B (en) Workflow scheduling method considering communication and computing cost under multi-cloud environment
Tang et al. Budget‐constraint stochastic task scheduling on heterogeneous cloud systems
Amoon A fault-tolerant scheduling system for computational grids
CN112463337B (en) Workflow task migration method used in mobile edge computing environment
Kaur et al. Deadline constrained scheduling of scientific workflows on cloud using hybrid genetic algorithm
CN113434267B (en) Cloud computing workflow dynamic scheduling method, device, equipment and storage medium
Han et al. EdgeTuner: Fast scheduling algorithm tuning for dynamic edge-cloud workloads and resources
Elsedimy et al. MOTS‐ACO: An improved ant colony optimiser for multi‐objective task scheduling optimisation problem in cloud data centres
Gupta et al. User defined weight based budget and deadline constrained workflow scheduling in cloud
Nasr et al. Cloudlet scheduling based load balancing on virtual machines in cloud computing environment
WO2023089350A1 (en) An architecture for a self-adaptive computation management in edge cloud
CN112130927B (en) Reliability-enhanced mobile edge computing task unloading method
CN115794323A (en) Task scheduling method, device, server and storage medium
CN106874215B (en) Serialized storage optimization method based on Spark operator
Rawat et al. Performance evaluation of an adopted model based on big-bang big-crunch and artificial neural network for cloud applications
Wang et al. A Novel Coevolutionary Approach to Reliability Guaranteed Multi-Workflow Scheduling upon Edge Computing Infrastructures

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