CN109146212B - Large-scale isomorphic task allocation method in crowdsourcing system - Google Patents

Large-scale isomorphic task allocation method in crowdsourcing system Download PDF

Info

Publication number
CN109146212B
CN109146212B CN201710457647.0A CN201710457647A CN109146212B CN 109146212 B CN109146212 B CN 109146212B CN 201710457647 A CN201710457647 A CN 201710457647A CN 109146212 B CN109146212 B CN 109146212B
Authority
CN
China
Prior art keywords
worker
task
load
tasks
workers
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
Application number
CN201710457647.0A
Other languages
Chinese (zh)
Other versions
CN109146212A (en
Inventor
蒋嶷川
唐孟萍
张友红
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.)
Foshan University
Original Assignee
Foshan 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 Foshan University filed Critical Foshan University
Priority to CN201710457647.0A priority Critical patent/CN109146212B/en
Publication of CN109146212A publication Critical patent/CN109146212A/en
Application granted granted Critical
Publication of CN109146212B publication Critical patent/CN109146212B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Factory Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a large-scale isomorphic task allocation method in a crowdsourcing system, which comprises the following steps: (1) acquiring information of workers and tasks in the system; (2) inputting a relation model f between the task execution rate of a worker and the load; (3) setting a batch distribution threshold value theta and a distribution time interval T, and executing a task distribution process once at every interval T; (4) during the task allocation time interval T, the worker continues to perform the tasks on its task set; (5) and after all tasks in the system task set are distributed, performing load balancing adjustment on the worker set at intervals of T. Compared with the prior art, the invention has the following advantages: (1) dynamic characteristics of the worker task execution rate, which are influenced by load, are considered, so that the system task completion time is optimized; (2) by adopting the batch distribution method, the task execution rate of workers can be maintained at a better level.

Description

Large-scale isomorphic task allocation method in crowdsourcing system
Technical Field
The invention relates to a task allocation technology in a crowdsourcing system, in particular to a large-scale isomorphic task allocation method.
Background
With the advent of the network age, more and more network services have emerged, wherein the popularization of the crowdsourcing system provides a new mode for task allocation and execution. The crowdsourcing system mainly comprises a task publisher, a crowdsourcing platform, workers and tasks, wherein the crowdsourcing platform distributes the tasks published by the task publisher to the workers to execute according to a proper distribution standard. The task execution speed is high, the cost is low, the accuracy is high, and complex tasks which cannot be completed by some computers can be completed, and the characteristics make the crowdsourcing system increasingly paid attention and popularized. In recent years, research on crowdsourcing systems has been increasing, wherein task allocation methods are particularly the focus of research. When designing a task allocation method, the characteristics of the crowd sourcing platform and the characteristics of workers and tasks need to be considered.
In the past research on task allocation problems, the matching between task requirements and worker skills, the surrounding environment, and the influence of the resources owned by the workers on the task execution rate are mainly concerned. Related researches find that in a real environment (such as a crowdsourcing system) of social individuals performing tasks, the speed of a worker performing the tasks is influenced by the load born by the worker to change dynamically, an inverse U-shaped curve relationship of increasing and decreasing is formed between the two, the load of the worker is increased within a certain range to improve the speed of the worker performing the tasks, and the increase of the load of the worker is reduced after the certain range is exceeded. However, in the past research on the task allocation problem, the problem is often ignored, and the dynamic characteristics of workers, of which the speed is influenced by the load, are not considered, so that the existing task allocation method has some limitations in the environment of executing tasks by social individuals.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the limitation that dynamic characteristics are presented by ignoring the influence of the load on the task execution rate of workers in the traditional task allocation method is broken through, and the large-scale isomorphic task allocation method considering the influence of the load on the task execution rate of the workers in the crowdsourcing system is provided.
The solution of the invention for solving the technical problem is as follows: the large-scale isomorphic task allocation method in the crowdsourcing system comprises the following steps: (1) acquiring worker information and task information in the system; (2) inputting a relation model f between the task execution rate of a worker and the load; (3) setting a batch distribution threshold value theta and a distribution time interval T, executing a task distribution process once at each interval T, sequentially calculating the number of tasks to be distributed to each worker according to the distribution threshold value theta, and taking out the corresponding number of tasks from a task set to distribute to each worker; (4) during the task allocation time interval T, the worker continues to perform the tasks on its task set; (5) and after all tasks in the system task set are distributed, performing load balancing adjustment on the worker set at intervals of T.
As a further improvement of the above technical solution, the specific steps of acquiring the worker information and the task information in the system in step (1) are as follows:
(1.1) acquiring worker information in the system, and establishing a worker set A ═ a in the system1,a2,...,anAnd a Skill set of Skill(s)1,s2,...,sq}, each worker aiThere is a set of optimal task execution rates S (a)i)＝{si1,si2,...,siqAnd a task set R (a)i)＝{ri1,ri2,...,rikAnd the optimum load value IW corresponding to each workeri(ii) a Set of optimal task execution rates S (a)i) Element s in (1)iqIndicating worker aiExecutive skills sqOptimum rate of corresponding task, siq>When 0, indicates a worker aiPossessing skills sqOtherwise, it indicates that the worker is not skilled sq(ii) a Task set R (a) of workersi) Storage worker aiThe distributed task set; optimum load value IWiIndicating worker aiAn optimal amount of load to be assumed at an optimal rate level;
(1.2) acquiring task information in the system, and establishing a system task set R ═ R in the system1,r2,...,rmAll tasks in the isomorphic task environment are the same and all arrive initially, each task rjComprising a skill requirement twistjAnd a task workload requirement lenjEach task need only be assigned to one worker to perform.
As a further improvement of the above technical solution, the specific steps of inputting the relationship model f between the worker task execution rate and the load in step (2) are as follows:
(2.1) inputting a relation model f between the task execution rate of the worker and the load by the system: an inverse U-shaped relationship with increasing and decreasing first exists between the task execution rate of workers and the load of the workers in the system, and the relationship comprises the optimal load state and the limited load bearing range of the workers;
(2.2) the system preferentially selects a relation model between the worker task execution rate and the load, wherein the parameter alpha mainly controls the influence strength of the worker load on the worker task execution rate,the parameter beta represents the range of influence of the worker load on the worker task execution rate, IWiIndicating worker aiOptimum load value, loadiIndicating worker aiThe current value of the load,
Figure BDA0001324097020000031
indicating worker aiMinimum time to execute a task, tiIndicating worker aiThe actual time to perform a task, the relational model is specifically represented as follows:
Figure BDA0001324097020000041
(2.3) Each worker aiOf the optimum load amount IWiIt is initially known that all workers, each a, obey a same model of the relationship between task execution rate and loadiAll have an optimum load amount IWiAnd are initially known, so there is heterogeneity among workers for the same relationship model.
7. As a further improvement of the above technical solution, the step (3) of setting a threshold θ for allocating in batches and an allocation time interval T, executing a task allocation process once every interval T, sequentially calculating the number of tasks to be allocated to each worker according to the allocation threshold θ, and taking out a corresponding number of tasks from the task set to allocate to each worker includes the following specific steps:
(3.1) setting a batch distribution threshold value theta;
(3.2) setting a distribution time interval T;
(3.3) for worker aiAccording to the assigned threshold value theta, and the worker aiCurrent load amount ofiAnd an optimum load value IWiThe calculation should continue for worker aiNumber of tasks allocated allocatanumi＝(1+θ)*IWi-loadiAnd take min (allocatanum) out of the task set RiR) tasks to worker ai；
And (3.4) sequentially executing the task allocation processes in the step (3.3) for all workers in the worker set A according to the descending order of the optimal task execution rate.
As a further improvement of the above technical solution, in the step (4), during the task allocation time interval T, the specific steps of the worker continuously performing the tasks on the task set of the worker are as follows:
(4.1) the invention adopts a task batch distribution strategy, and workers continuously execute tasks on a task set of the workers in a distribution time interval T;
(4.2) if the task set of the workers is empty, the workers are in an idle waiting state in the distribution time interval;
(4.3) if the task set of the worker is not empty, the worker continuously executes the tasks on the task set according to the principle of first-come first-serve, and for the task rj＝(skillj,lenj) If the worker aiIs loaded at the current load amountjThen the task execution rate of the worker at this time is ratei(loadj)＝ai(skillj)*f(loadj) The time required to perform the task is tij＝lenj/ratei(loadj)。
As a further improvement of the above technical solution, after all tasks in the system task set in step (5) are allocated, the specific steps of performing load balancing adjustment on the worker set at intervals of time T are as follows:
(5.1) setting a timer, and performing load balancing adjustment on the worker set A once at intervals of a period of time T;
(5.2) load balancing adjustment: sequentially processing the worker set A according to the order from large to small of the optimal execution rate of the workers, and processing the worker a with an empty task setiSequentially arranging the workers a with lower speed from the worker set A according to the order of the best execution speed of the workers from small to largejTo the worker aiAccording to the worker aiCurrent load amount ofiAnd an optimum load value IWiAnd worker ajCurrent load amount ofjCalculating the task transfer quantity as taskTransferNum(j,i)＝min(IWi-loadi,loadj) So that the worker aiThe load amount on the catalyst reaches the optimum load amount.
The invention has the beneficial effects that: compared with the prior art, the invention has the following advantages: (1) the dynamic characteristics of the workers, which are influenced by the load, of the task execution rate are considered, the task allocation is reasonably carried out by effectively utilizing the relation between the task execution rate of the workers and the load, and the task execution rate of the workers is maintained at a better level so as to optimize the task completion time of the system; (2) the method adopts a batch distribution method, reasonably controls the task load born by a worker by executing a batch distribution strategy under a large-scale task environment, can effectively avoid the problem that the task execution rate of the worker is very low due to overload, adopts a mode of distributing according to a threshold value in each round, has much lower time complexity compared with other methods, and can simultaneously maintain the task execution rate of the worker at a better level.
Drawings
In order to more clearly illustrate the technical solution in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly described below. It is clear that the described figures are only some embodiments of the invention, not all embodiments, and that a person skilled in the art can also derive other designs and figures from them without inventive effort.
FIG. 1 is a flow chart of the present invention.
Detailed Description
The conception, the specific structure, and the technical effects produced by the present invention will be clearly and completely described below in conjunction with the embodiments and the accompanying drawings to fully understand the objects, the features, and the effects of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and those skilled in the art can obtain other embodiments without inventive effort based on the embodiments of the present invention, and all embodiments are within the protection scope of the present invention. In addition, all the coupling/connection relationships mentioned herein do not mean that the components are directly connected, but mean that a better coupling structure can be formed by adding or reducing coupling accessories according to specific implementation conditions. All technical characteristics in the invention can be interactively combined on the premise of not conflicting with each other.
Referring to fig. 1, a large-scale isomorphic task allocation method in a crowdsourcing system includes the following steps: (1) acquiring worker information and task information in the system; (2) inputting a relation model f between the task execution rate of a worker and the load; (3) setting a batch distribution threshold value theta and a distribution time interval T, executing a task distribution process once at each interval T, sequentially calculating the number of tasks to be distributed to each worker according to the distribution threshold value theta, and taking out the corresponding number of tasks from a task set to distribute to each worker; (4) during the task allocation time interval T, the worker continues to perform the tasks on its task set; (5) and after all tasks in the system task set are distributed, performing load balancing adjustment on the worker set at intervals of T.
Further as a preferred embodiment, the specific steps of acquiring the worker information and the task information in the system in the step (1) are as follows:
(1.1) acquiring worker information in the system, and establishing a worker set A ═ a in the system1,a2,...,anAnd a Skill set of Skill(s)1,s2,...,sq}, each worker aiThere is a set of optimal task execution rates S (a)i)＝{si1,si2,...,siqAnd a task set R (a)i)＝{ri1,ri2,...,rikAnd the optimum load value IW corresponding to each workeri(ii) a Set of optimal task execution rates S (a)i) Element s in (1)iqIndicating worker aiExecutive skills sqOptimum rate of corresponding task, siq>When 0, indicates a worker aiPossessing skills sqOtherwise, it indicates that the worker is not skilled sq(ii) a Task set R (a) of workersi) Storage worker aiThe distributed task set; optimum load value IWiIndicating worker aiAn optimal amount of load to be assumed at an optimal rate level;
(1.2) acquiring task information in the system, and establishing a system task set R ═ R in the system1,r2,...,rmAll tasks in the isomorphic task environment are the same and all arrive initially, each task rjComprising a skill requirement twistjAnd a task workload requirement lenjEach task need only be assigned to one worker to perform.
Further as a preferred embodiment, the specific steps in the step (2) are:
(2.1) the system inputs a relation model f between the worker task execution rate and the load. An inverted U-shaped relationship of increasing and decreasing between the rate at which a worker performs a task and the load of the worker exists in the system, and the relationship comprises the optimal load state and the limited load bearing range of the worker.
(2.2) the system preferentially selects a relation model between the worker task execution rate and the load, wherein the parameter alpha mainly controls the influence strength of the worker load on the worker task execution rate, the parameter beta represents the influence range of the worker load on the worker task execution rate, and IWiIndicating worker aiOptimum load value, loadiIndicating worker aiThe current value of the load,
Figure BDA0001324097020000081
indicating worker aiMinimum time to execute a task, tiIndicating worker aiThe actual time to perform a task, the relational model is specifically represented as follows:
Figure BDA0001324097020000082
(2.3) Each worker aiOf the optimum load amount IWiIt is initially known that all workers, each a, obey a same model of the relationship between task execution rate and loadiAll have an optimum load amount IWiAnd are initially known, so there is heterogeneity among workers for the same relationship model.
Further as a preferred embodiment, the specific steps in the step (3) are:
(3.1) setting a batch distribution threshold value theta. In a large-scale task environment, a relationship of increasing and decreasing between the task execution rate of a worker and the task load borne by the worker exists, and in order to maintain the worker rate at a better level, the task load borne by the worker needs to be reasonably controlled.
And (3.2) setting the distribution time interval T. In a large-scale task environment, the invention adopts a batch distribution method, needs to set a timer, and executes a task distribution process at intervals of a period of time T.
(3.3) for worker aiAccording to the assigned threshold value theta, and the worker aiCurrent load amount ofiAnd an optimum load value IWiThe calculation should continue for worker aiNumber of tasks allocated allocatanumi＝(1+θ)*IWi-loadiAnd take min (allocatanum) out of the task set RiR) tasks to worker ai。
And (3.4) sequentially executing the task allocation processes in the step (3.3) for all workers in the worker set A according to the descending order of the optimal task execution rate.
Further as a preferred embodiment, the specific steps in the step (4) are:
(4.1) the invention adopts a task batch distribution strategy, and in the distribution time interval T, workers continuously execute tasks on the task set, and the task distribution time is negligible.
(4.2) if the task set of the workers is empty, the workers are in an idle waiting state in the distribution time interval;
(4.3) if the worker task set is not empty, the worker follows the principle of first-come first-serveContinuously executing tasks on its task set, for task rj＝(skillj,lenj) If the worker aiIs loaded at the current load amountjThen the task execution rate of the worker at this time is ratei(loadj)＝ai(skillj)*f(loadj) The time required to perform the task is tij＝lenj/ratei(loadj)。
Further as a preferred embodiment, the specific steps in the step (5) are:
(5.1) when all tasks in the system task set R are distributed and completed, and partial worker task sets are not empty, due to the difference of the load capacity of workers and the sequence of task distribution, a phenomenon of load unbalance occurs on the worker set A, and therefore load balance adjustment needs to be carried out on the worker set A at the moment so as to optimize the task completion time.
(5.2) setting a timer, and carrying out load balancing adjustment on the worker set A once every a period of time T.
(5.3) sequentially processing the worker set A according to the order of the best execution rate of the workers from large to small, and processing the worker a with an empty task setiSequentially arranging the workers a with lower speed from the worker set A according to the order of the best execution speed of the workers from small to largejTo the worker aiAccording to the worker aiCurrent load amount ofiAnd an optimum load value IWiAnd worker ajCurrent load amount ofjThe number of task transfers is tasktransferrnum (j, i) ═ min (IW)i-loadi,loadj) So that the worker aiThe load amount on the catalyst reaches the optimum load amount.
A large-scale isomorphic task allocation method considering that the task execution rate of workers is influenced by load in a crowdsourcing system mainly comprises the following points: acquiring worker information and task information in the system; inputting a relation model f between the task execution rate of a worker and the load; setting a batch distribution threshold value theta and a distribution time interval T, executing a task distribution process once at each interval T, sequentially calculating the number of tasks to be distributed to each worker according to the distribution threshold value theta, and taking out the corresponding number of tasks from a task set to distribute to each worker; during the task allocation time interval T, the worker continues to perform the tasks on its task set; and after all tasks in the system task set are distributed, performing load balancing adjustment on the worker set at intervals of T. According to the method, a task batch distribution strategy is adopted according to the relationship of increasing and decreasing between the task execution rate of workers and the load, a large-scale task set is divided into a plurality of small-scale task subsets through batch distribution, and the task completion time is optimized.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that the present invention is not limited to the details of the embodiments shown and described, but is capable of numerous equivalents and substitutions without departing from the spirit of the invention as set forth in the claims appended hereto.

Claims (5)

1. The large-scale isomorphic task allocation method in the crowdsourcing system is characterized by comprising the following steps: the method comprises the following steps:
(1) acquiring worker information and task information in the system;
(2) inputting worker task execution rate and loadiA relation model f (load) betweeni)；
(3) Setting a batch distribution threshold value theta and a distribution time interval T, executing a task distribution process once at each interval T, sequentially calculating the number of tasks to be distributed to each worker according to the distribution threshold value theta, and taking out the corresponding number of tasks from a task set to distribute to each worker;
(4) during the task allocation time interval T, the worker continues to perform the tasks on its task set;
(5) after all tasks in the system task set are distributed, performing load balancing adjustment on the worker set at intervals of T;
the specific steps of inputting the relation model f between the worker task execution rate and the load in the step (2) are as follows:
(2.1) inputting a relation model f between the task execution rate of the worker and the load by the system: an inverse U-shaped relationship with increasing and decreasing first exists between the task execution rate of workers and the load of the workers in the system, and the relationship comprises the optimal load state and the limited load bearing range of the workers;
(2.2) the system selects a relation model between the worker task execution rate and the load, wherein the parameter alpha mainly controls the influence strength of the worker load on the worker task execution rate, the parameter beta represents the influence range of the worker load on the worker task execution rate, and IWiIndicating worker aiOptimum load value, loadiIndicating worker aiThe current value of the load,
Figure FDA0003415248130000011
indicating worker aiMinimum time to execute a task, tiIndicating worker aiThe actual time to perform a task, the relational model is specifically represented as follows:
Figure FDA0003415248130000021
(2.3) Each worker aiOf the optimum load amount IWiIt is initially known that all workers, each a, obey a same model of the relationship between task execution rate and loadiAll have an optimum load amount IWiAnd are initially known, so there is heterogeneity among workers for the same relationship model.
Wherein, the step (3) sets a batch distribution threshold value theta and a distribution time interval T, each time interval T executes a task distribution process, the number of tasks to be distributed to each worker is sequentially calculated according to the distribution threshold value theta, and the specific steps of taking out the corresponding number of tasks from the task set and distributing the tasks to each worker include:
(3.1) setting a batch distribution threshold value theta;
(3.2) setting a distribution time interval T;
(3.3) for worker aiAccording to the assigned threshold value theta, and the worker aiCurrent load amount ofiAnd an optimum load value IWiThe calculation should continue for worker aiNumber of tasks allocated allocatanumi＝(1+θ)*IWi-loadiAnd take min (allocatanum) out of the task set RiR) tasks to worker ai。
2. The method of large-scale isomorphic task allocation in a crowdsourcing system of claim 1, wherein: the specific steps of acquiring the worker information and the task information in the system in the step (1) are as follows:
(1.1) acquiring worker information in the system, and establishing a worker set A ═ a in the system1,a2,...,anAnd a Skill set of Skill(s)1,s2,...,sq}, each worker aiThere is a set of optimal task execution rates S (a)i)＝{si1,si2,...,siqAnd a task set R (a)i)＝{ri1,ri2,...,rikAnd the optimum load value IW corresponding to each workeri(ii) a Set of optimal task execution rates S (a)i) Element s in (1)iqIndicating worker aiExecutive skills sqOptimum rate of corresponding task, siq>When 0, indicates a worker aiPossessing skills sqOtherwise, it indicates that the worker is not skilled sq(ii) a Task set R (a) of workersi) Storage worker aiThe distributed task set; optimum load value IWiIndicating worker aiAn optimal amount of load to be assumed at an optimal rate level;
(1.2) acquiring task information in the system, and establishing a system task set R ═ R in the system1,r2,...,rmAll tasks in the isomorphic task environment are the same and all arrive initially, each task rjComprising a skill requirement twistjAnd a task workload requirement lenjEach task only needs to be divided intoAnd allocating a worker to perform the operation.
3. The method of large-scale isomorphic task allocation in a crowdsourcing system of claim 2, wherein: the specific steps of setting the batch distribution threshold θ and the distribution time interval T in the step (3), executing a task distribution process once every interval T, sequentially calculating the number of tasks to be distributed to each worker according to the distribution threshold θ, and taking out the corresponding number of tasks from the task set to distribute to each worker further include the following steps:
and (3.4) sequentially executing the task allocation process in the sub-step (3.3) in the step (3) for all workers in the worker set A according to the sequence of the optimal task execution rate from large to small.
4. The method of claim 3, wherein the method comprises: the specific steps of the worker continuously performing the tasks on the task set within the task allocation time interval T in the step (4) are as follows:
(4.1) the invention adopts a task batch distribution strategy, and workers continuously execute tasks on a task set of the workers in a distribution time interval T;
(4.2) if the task set of the workers is empty, the workers are in an idle waiting state in the distribution time interval;
(4.3) if the task set of the worker is not empty, the worker continuously executes the tasks on the task set according to the principle of first-come first-serve, and for the task rj＝(skillj,lenj) If the worker aiIs loaded at the current load amountiThen the task execution rate of the worker at this time is ratei(loadi)＝ai(skillj)*f(loadi) The time required to perform the task is tij＝lenj/ratei(loadi)。
5. The method of claim 4, wherein the method comprises: after all tasks in the system task set in the step (5) are distributed, the specific steps of performing load balancing adjustment on the worker set at intervals of T are as follows:
(5.1) setting a timer, and performing load balancing adjustment on the worker set A once at intervals of a period of time T;
(5.2) load balancing adjustment: sequentially processing the worker set A according to the order from large to small of the optimal execution rate of the workers, and processing the worker a with an empty task setiSequentially arranging the workers a with lower speed from the worker set A according to the order of the best execution speed of the workers from small to largejTo the worker aiAccording to the worker aiCurrent load amount ofiAnd an optimum load value IWiAnd worker ajCurrent load amount ofjThe number of task transfers is tasktransferrnum (j, i) ═ min (IW)i-loadi,loadj) So that the worker aiThe load amount on the catalyst reaches the optimum load amount.
CN201710457647.0A 2017-06-16 2017-06-16 Large-scale isomorphic task allocation method in crowdsourcing system Active CN109146212B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710457647.0A CN109146212B (en) 2017-06-16 2017-06-16 Large-scale isomorphic task allocation method in crowdsourcing system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710457647.0A CN109146212B (en) 2017-06-16 2017-06-16 Large-scale isomorphic task allocation method in crowdsourcing system

Publications (2)

Publication Number Publication Date
CN109146212A CN109146212A (en) 2019-01-04
CN109146212B true CN109146212B (en) 2022-03-25

Family

ID=64830601

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710457647.0A Active CN109146212B (en) 2017-06-16 2017-06-16 Large-scale isomorphic task allocation method in crowdsourcing system

Country Status (1)

Country Link
CN (1) CN109146212B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107220780B (en) * 2017-06-16 2020-07-28 佛山科学技术学院 Heterogeneous task execution sequence optimization method in crowdsourcing system
CN111415062A (en) * 2020-02-19 2020-07-14 平安科技（深圳）有限公司 Crowdsourcing task allocation method and device, electronic equipment and storage medium
WO2021163986A1 (en) * 2020-02-21 2021-08-26 Yangtze Memory Technologies Co., Ltd. Method and system for scheduling semiconductor fabrication

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002351678A (en) * 2001-05-28 2002-12-06 Atr Adaptive Communications Res Lab Scheduling method in continuous media stream processing
CN103647984A (en) * 2013-11-14 2014-03-19 天脉聚源(北京)传媒科技有限公司 Load distribution method and system for video processing servers
CN105141541A (en) * 2015-09-23 2015-12-09 浪潮(北京)电子信息产业有限公司 Task-based dynamic load balancing scheduling method and device
CN106557891A (en) * 2016-12-05 2017-04-05 苏州大学 Mass-rent method for allocating tasks based on user dependability
CN106776008A (en) * 2016-11-23 2017-05-31 福建六壬网安股份有限公司 A kind of method and system that load balancing is realized based on zookeeper

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150262111A1 (en) * 2014-03-12 2015-09-17 Nanyang Technological University Apparatus and method for efficient task allocation in crowdsourcing
CN104731663B (en) * 2015-03-31 2018-04-13 北京奇艺世纪科技有限公司 Task processing method and system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002351678A (en) * 2001-05-28 2002-12-06 Atr Adaptive Communications Res Lab Scheduling method in continuous media stream processing
CN103647984A (en) * 2013-11-14 2014-03-19 天脉聚源(北京)传媒科技有限公司 Load distribution method and system for video processing servers
CN105141541A (en) * 2015-09-23 2015-12-09 浪潮(北京)电子信息产业有限公司 Task-based dynamic load balancing scheduling method and device
CN106776008A (en) * 2016-11-23 2017-05-31 福建六壬网安股份有限公司 A kind of method and system that load balancing is realized based on zookeeper
CN106557891A (en) * 2016-12-05 2017-04-05 苏州大学 Mass-rent method for allocating tasks based on user dependability

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Load Balancing Task Scheduling;Tingting Wang 等;《2014 IEEE 12th International Conference on Dependable, Autonomic and Secure Computing》;20141006;第146-152页 *
网络结构化多Agent系统的任务分配;蒋嶷川;《模式识别与人工智能》;20120430;第25卷(第2期);第262-272页 *

Also Published As

Publication number Publication date
CN109146212A (en) 2019-01-04

Similar Documents

Publication Publication Date Title
US10474504B2 (en) Distributed node intra-group task scheduling method and system
CN102567080B (en) Virtual machine position selection system facing load balance in cloud computation environment
CN109146212B (en) Large-scale isomorphic task allocation method in crowdsourcing system
CN105260818B (en) Mix the on-line optimizing scheduling method with deadline constraint workflow group under cloud environment
CN107220780B (en) Heterogeneous task execution sequence optimization method in crowdsourcing system
CN103401939A (en) Load balancing method adopting mixing scheduling strategy
CN109582448A (en) A kind of edge calculations method for scheduling task towards criticality and timeliness
CN104239144A (en) Multilevel distributed task processing system
CN106991006B (en) Support the cloud workflow task clustering method relied on and the time balances
CN105975340A (en) Deployment algorithm of application distribution of virtual machine
CN108304256B (en) Task scheduling method and device with low overhead in edge computing
CN110308987B (en) Method for updating connection parameters of distributed training tasks on container cloud
CN106776025A (en) A kind of computer cluster job scheduling method and its device
CN107291550A (en) A kind of Spark platform resources dynamic allocation method and system for iterated application
CN106528288A (en) Resource management method, device and system
CN107239342A (en) A kind of storage cluster task management method and device
CN103257896B (en) A kind of Max-D job scheduling method under cloud environment
CN112540841A (en) Task scheduling method and device, processor and electronic equipment
CN104917839A (en) Load balancing method for use in cloud computing environment
CN108111337B (en) Method and equipment for arbitrating main nodes in distributed system
CN105740059A (en) Particle swarm scheduling method for divisible task
CN108984286A (en) A kind of resource regulating method and system of cloud computing platform
Li et al. Endpoint-flexible coflow scheduling across geo-distributed datacenters
CN106407007B (en) Cloud resource configuration optimization method for elastic analysis process
CN103744735A (en) Method and device for scheduling multi-core resource

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