CN113157430A - Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment - Google Patents

Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment Download PDF

Info

Publication number
CN113157430A
CN113157430A CN202011475047.5A CN202011475047A CN113157430A CN 113157430 A CN113157430 A CN 113157430A CN 202011475047 A CN202011475047 A CN 202011475047A CN 113157430 A CN113157430 A CN 113157430A
Authority
CN
China
Prior art keywords
user
edge server
subtask
edge
service
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.)
Withdrawn
Application number
CN202011475047.5A
Other languages
Chinese (zh)
Inventor
向正哲
郑宇航
邓水光
王东京
陈垣毅
郑增威
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.)
Zhejiang University City College ZUCC
Original Assignee
Zhejiang University City College ZUCC
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 Zhejiang University City College ZUCC filed Critical Zhejiang University City College ZUCC
Priority to CN202011475047.5A priority Critical patent/CN113157430A/en
Publication of CN113157430A publication Critical patent/CN113157430A/en
Withdrawn legal-status Critical Current

Links

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
    • G06F9/50—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
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • 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
    • G06F9/505—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 considering the load
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/502—Proximity

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention provides a low-cost task allocation and service deployment method for a mobile group perception system in an edge computing environment, which comprises the following steps: s1) communication connection between the user and the edge server; s2) task submission and distribution; s3) establishing an objective function; s4) service deployment and task allocation. The invention has the advantages that: under the constraints of budget of application program developers, available resources of an edge server and user capacity, a mathematical model of Mixed Integer Quadratic Programming (MIQP) with the aim of balancing task quality and cost is provided, and the system operation cost can be reduced as far as possible on the basis of ensuring the task completion quality.

Description

Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment
Technical Field
The invention relates to the technical field of edge computing and mobile swarm intelligence perception, in particular to a low-cost task allocation and service deployment method for a mobile swarm perception system in an edge computing environment.
Background
With the development of mobile computing technology, we are embracing an era of mobile devices and services. According to the GSMA report, global mobile application users are approximately 51 hundred million, and will grow at 1.9% per year before 2025 years. As a result, mobile devices and mobile applications become more and more important, reshaping the communication between human and machine. The proliferation of mobile users and devices has created a huge market that attracts worldwide attention. In order to make themselves outstanding among competitors, mobile application enterprises may wish to better understand the preferences of these users, discovering their underlying patterns of behavior. Therefore, researchers at these businesses are always trying to collect records of user behavior as much as possible, and even directly interview their target users, and they are sure that these structured or unstructured and sequential/non-sequential context data will help them build a general user profile model to analyze and predict the user's future behavior.
However, due to subconscious privacy protection and concerns about external computing power consumption, few people are willing to provide their own data, and it is difficult for application developers to legally collect data of sufficiently high quality for their artificial intelligence models. To address this problem, more and more developers are turning to Mobile Crowd Sensing (MCS) techniques. In particular, MCS is a human-oriented technique that utilizes built-in sensors of the user's mobile device and the user's participation to collect data. It is not only concerned with the validity and accuracy of the data, but also with how to stimulate users to share their data. With MCS techniques, a reliable publish/subscribe interaction framework is established between users and developers so that high quality data can be collected, which users can accept and would like to collect if the developers would like to pay for their collaboration. However, the delay caused by long-distance transmission, the traffic congestion of massive data in the network, and the energy consumption caused by data preprocessing limit the application of MCS in a typical centralized architecture.
Fortunately, multiple access edge computing (MEC) techniques have been proposed to address the above-mentioned problems. MEC is a new paradigm that has recently emerged as an enhancement to mobile cloud computing to optimize the use of mobile resources and wireless networks to provide context-aware services. With the help of MECs, the computing and transport section between the mobile device and the cloud is migrated to the edge server. Thus, users can easily connect to edge servers in their vicinity over a wireless network. The short distance connection between the user and the edge server can greatly reduce latency, while the computing power of the edge server can fully meet the requirements of traditional tasks. More importantly, managing services (e.g., data pre-processing services) in the MEC environment will become easy with the help of an attractive container platform like kubernets. However, these advantages are not the cause of inattention in multi-source data acquisition planning-if the sensing tasks are not assigned to the appropriate user, the data acquisition task may even obtain low quality data at a higher cost. More importantly, because the edge servers are resource constrained, there are not enough resources to run the data pre-processing services if they are not deployed on the appropriate edge server. In the process of carrying out mobile crowd sensing, only the improvement on an excitation mechanism or only the improvement on an MEC task allocation method is considered in the prior art. Therefore, it is very important to design a task allocation scheme and a service deployment scheme to balance quality and cost.
Disclosure of Invention
The invention aims to provide a low-cost task allocation and service deployment method for a mobile group perception system in an edge-oriented computing environment, which balances quality and cost.
In order to achieve the purpose, the invention is realized by the following technical scheme:
a low-cost task allocation and service deployment method for a mobile group awareness system in an edge-oriented computing environment comprises the following steps:
s1) communication connection between user and edge server
Acquiring user position information to be served, calculating user vectors to be served by each edge server according to the communication distance between a user and the edge server and under the condition of ensuring the minimum communication distance by combining the known position information of each edge server, and splicing the user vectors according to columns to form a matrix L;
s2) submission and distribution of tasks
All edge servers form an MCS system, MCS marker responsible for submitting and distributing tasks is arranged in the system, different edge servers serve different users, each user can be selected to register as a service requester or a service executor, and when a service requester submits a task T to the MCS marker, the MCS marker divides the task into K subtasks and obtains a subtask list (T)1,T2,...,TK) Each subtask is deployed to each edge server, and the corresponding service S is equal to (S)1,s2,...,sK)；
S3) establishing an objective function
Establishing an objective function Cr：
Figure BDA0002834975590000031
Wherein, Pm,kHow many times a subtask k, D needs to be completed for user mj,kC whether subtask k is deployed on edge server jm,kThe number of incentives k required for the user m to complete the subtask once, and v is the unit cost for deploying the subtask to the edge server;
setting the constraint of the objective function:
constraint one is
Figure BDA0002834975590000032
Wherein the content of the first and second substances,
Figure BDA0002834975590000033
qm,kfor user m the quality of completion of subtask k, Wm,kWill of user m to complete subtask k, Lm,jWhether or not user m is at the edgeServer j's service scope; epsilonkThe minimum completion quality required for subtask k;
constraint two is
Figure BDA0002834975590000034
That is, the number of sub-tasks provided on each edge server cannot exceed the upper limit of the load of the edge server itself; wherein the content of the first and second substances,
Figure BDA0002834975590000035
an upper load limit for supported services for edge server j;
constraint III is
Figure BDA0002834975590000036
That is, the total number of subtasks that each user can execute simultaneously cannot exceed the load upper limit of the user; wherein the content of the first and second substances,
Figure BDA0002834975590000037
the load upper limit of the subtask which can be run by the user m;
s4) service deployment and task assignment
For the above objective function CrAnd performing minimum solution to obtain a matrix P and a matrix D, namely how many times each user m needs to complete the subtask k, and whether the subtask k is deployed on the edge server j, so as to perform service deployment on the edge server and task allocation of a service executor.
Further, the solution of the objective function in the step S3) is simplified into a mixed integer quadratic programming problem, and a matrix is defined
Figure BDA0002834975590000041
x＝[p d]TWherein P ═ P1,P2,...,PK]T， d＝[D1,D2,...,DK]T，
Figure BDA0002834975590000042
Defining an excitation matrix c ═ c1,c2,...,cK]TAnd
Figure BDA0002834975590000043
Figure BDA0002834975590000044
the above problem is translated into:
Figure BDA0002834975590000045
Figure BDA0002834975590000046
Figure BDA0002834975590000047
Figure BDA0002834975590000048
x∈N(M+N)·K。
further, the matrix L is Lm,jE {0,1}, each row corresponds to a user, each column corresponds to an edge server, and each user only connects with the edge server which is closest to the user in communication distance, so that the method has the advantages that
Figure BDA0002834975590000049
Compared with the prior art, the invention has the following advantages:
the invention relates to a low-cost task allocation and service deployment method for a mobile group perception system in an edge computing environment, which firstly researches the relation between task quality and cost in an MCS problem based on MEC and provides an analysis frame of the task quality and the cost on the basis of two classical cost-performance balance problems; the invention proposes a mathematical model of Mixed Integer Quadratic Programming (MIQP) with the aim of balancing task quality and cost under the constraints of budget of application program developers, available resources of edge servers and user capacity. The invention comprehensively considers the incentive mechanism and the mobile crowd sensing method of MEC task allocation, and can reduce the system operation cost as much as possible on the basis of ensuring the task completion quality.
Detailed Description
The following describes embodiments of the present invention in further detail.
A high-quality task allocation and service deployment method for a mobile group awareness system in an edge-oriented computing environment comprises the following steps:
s1) communication connection between user and edge server
Acquiring user position information to be served, calculating user vectors to be served by each edge server according to the communication distance between a user and the edge server and under the condition of ensuring the minimum communication distance by combining the known position information of each edge server, and splicing the user vectors according to columns to form a matrix L;
s2) submission and distribution of tasks
All edge servers form an MCS system, MCS marker responsible for submitting and distributing tasks is arranged in the system, different edge servers serve different users, each user can be selected to register as a service requester or a service executor, and when a service requester submits a task T to the MCS marker, the MCS marker divides the task into K subtasks and obtains a subtask list (T)1,T2,...,TK) Each subtask is deployed to each edge server, and the corresponding service S is equal to (S)1,s2,...,sK) (ii) a According to the decision, the edge server can distribute the subtask corresponding to a certain service to a certain user for execution for multiple times, corresponding to the decision variable Pm,kI.e. user m needs to perform P on subtask km,kSecondly;
acquiring the corresponding completion quality of each subtask, the willingness of a user to complete the subtask and the stimulation of the user to complete the subtask; each task has the lowest completion quality constraint, and each service executor and the edge server have the load upper limit constraint;
s3) establishing an objective function
Establishing an objective function Cr：
Figure BDA0002834975590000061
Wherein, Pm,kHow many times a subtask k, D needs to be completed for user mj,kC whether subtask k is deployed on edge server jm,kThe number of incentives k required for the user m to complete the subtask once, and v is the unit cost for deploying the subtask to the edge server;
the matrix L is Lm,jE {0,1}, each row corresponds to a user, each column corresponds to an edge server, and each user only connects with the edge server which is closest to the user in communication distance, so that the method has the advantages that
Figure BDA0002834975590000062
Setting the constraint of the objective function:
constraint one is
Figure BDA0002834975590000063
Wherein the content of the first and second substances,
Figure BDA0002834975590000064
qm,kfor user m the quality of completion of subtask k, Wm,kWill of user m to complete subtask k, Lm,jWhether the user m is in the service range of the edge server j or not; epsilonkThe minimum completion quality required for subtask k;
constraint two is
Figure BDA0002834975590000065
That is, the number of sub-tasks provided on each edge server cannot exceed the upper limit of the load of the edge server itself; wherein the content of the first and second substances,
Figure BDA0002834975590000066
an upper load limit for supported services for edge server j;
constraint III is
Figure BDA0002834975590000067
That is, the total number of subtasks that each user can execute simultaneously cannot exceed the load upper limit of the user; wherein the content of the first and second substances,
Figure BDA0002834975590000068
the load upper limit of the subtask which can be run by the user m;
s4) service deployment and task assignment
For the above objective function CrAnd performing minimum solution to obtain a matrix P and a matrix D, namely how many times each user m needs to complete the subtask k, and whether the subtask k is deployed on the edge server j, so as to perform service deployment on the edge server and task allocation of a service executor.
Further, the solution of the objective function in the step S3) is simplified into a mixed integer quadratic programming problem, and a matrix is defined
Figure BDA0002834975590000071
x＝[p d]TWherein P ═ P1,P2,...,PK]T， d＝[D1,D2,...,DK]T，
Figure BDA0002834975590000072
Then define matrix akIs composed of
Figure BDA0002834975590000073
Figure BDA0002834975590000074
Wherein
Figure DEST_PATH_GDA0003073233340000085
Is the Hadamard product of A and B;
defining an excitation matrix c ═ c1,c2,...,cK]TAnd
Figure BDA0002834975590000075
Figure BDA0002834975590000076
thus is provided with
Figure BDA0002834975590000077
Figure BDA0002834975590000078
The above problem is translated into:
Figure BDA0002834975590000079
Figure BDA00028349755900000710
Figure BDA00028349755900000711
Figure BDA0002834975590000081
x∈N(M+N)·K。
the foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the spirit of the present invention, and these modifications and improvements should also be considered as within the scope of the present invention.

Claims (3)

1. A low-cost task allocation and service deployment method for a mobile group awareness system in an edge computing environment is characterized by comprising the following steps:
s1) communication connection between user and edge server
Acquiring user position information to be served, calculating user vectors to be served by each edge server according to the communication distance between a user and the edge server and under the condition of ensuring the minimum communication distance by combining the known position information of each edge server, and splicing the user vectors according to columns to form a matrix L;
s2) submission and distribution of tasks
All edge servers form an MCS system, MCS marker responsible for submitting and distributing tasks is arranged in the system, different edge servers serve different users, each user can be selected to register as a service requester or a service executor, and when a service requester submits a task T to the MCS marker, the MCS marker divides the task into K subtasks and obtains a subtask list (T)1,T2,...,TK) Each subtask is deployed to each edge server, and the corresponding service S is equal to (S)1,s2,...,sK)；
S3) establishing an objective function
Establishing an objective function Cr：
Figure FDA0002834975580000011
Wherein, Pm,kHow many times a subtask k, D needs to be completed for user mj,kC whether subtask k is deployed on edge server jm,kThe number of incentives k required for the user m to complete the subtask once, and v is the unit cost for deploying the subtask to the edge server;
setting the constraint of the objective function:
constrainingOne is
Figure FDA0002834975580000012
Wherein the content of the first and second substances,
Figure FDA0002834975580000013
qm,kfor user m the quality of completion of subtask k, Wm,kWill of user m to complete subtask k, Lm,jWhether the user m is in the service range of the edge server j or not; epsilonkThe minimum completion quality required for subtask k;
constraint two is
Figure FDA0002834975580000021
That is, the number of sub-tasks provided on each edge server cannot exceed the upper limit of the load of the edge server itself; wherein the content of the first and second substances,
Figure FDA0002834975580000022
an upper load limit for supported services for edge server j;
constraint III is
Figure FDA0002834975580000023
That is, the total number of subtasks that each user can execute simultaneously cannot exceed the load upper limit of the user; wherein the content of the first and second substances,
Figure FDA0002834975580000024
the load upper limit of the subtask which can be run by the user m;
s4) service deployment and task assignment
For the above objective function CrAnd performing minimum solution to obtain a matrix P and a matrix D, namely how many times each user m needs to complete the subtask k, and whether the subtask k is deployed on the edge server j, so as to perform service deployment on the edge server and task allocation of a service executor.
2. The method for high-quality task allocation and service deployment of the mobile community awareness system in the edge-oriented computing environment according to claim 1, wherein:
simplifying the solution of the objective function in the step S3) into a mixed integer quadratic programming problem, and defining a matrix
Figure FDA0002834975580000025
x＝[p d]TWherein P ═ P1,P2,...,PK]T，
d＝[D1,D2,...,DK]T，
Figure FDA0002834975580000026
Defining an excitation matrix c ═ c1,c2,...,cK]TAnd
Figure FDA0002834975580000027
Figure FDA0002834975580000028
the above problem is translated into:
Figure FDA0002834975580000029
Figure FDA00028349755800000210
Figure FDA00028349755800000211
Figure FDA0002834975580000031
x∈N(M+N)·K。
3. the method for high-quality task allocation and service deployment of the mobile community awareness system in the edge-oriented computing environment according to claim 1 or 2, wherein:
the matrix L is Lm,jE {0,1}, each row corresponds to a user, each column corresponds to an edge server, and each user only connects with the edge server which is closest to the user in communication distance, so that the method has the advantages that
Figure FDA0002834975580000032
CN202011475047.5A 2020-12-14 2020-12-14 Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment Withdrawn CN113157430A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011475047.5A CN113157430A (en) 2020-12-14 2020-12-14 Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011475047.5A CN113157430A (en) 2020-12-14 2020-12-14 Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment

Publications (1)

Publication Number Publication Date
CN113157430A true CN113157430A (en) 2021-07-23

Family

ID=76882583

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011475047.5A Withdrawn CN113157430A (en) 2020-12-14 2020-12-14 Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment

Country Status (1)

Country Link
CN (1) CN113157430A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113419867A (en) * 2021-08-23 2021-09-21 浙大城市学院 Energy-saving service supply method in edge-oriented cloud collaborative computing environment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113419867A (en) * 2021-08-23 2021-09-21 浙大城市学院 Energy-saving service supply method in edge-oriented cloud collaborative computing environment
CN113419867B (en) * 2021-08-23 2022-01-18 浙大城市学院 Energy-saving service supply method in edge-oriented cloud collaborative computing environment

Similar Documents

Publication Publication Date Title
Xu et al. Blockchain-based cloudlet management for multimedia workflow in mobile cloud computing
CN104737132B (en) For the message queue in on-demand service environment based on the resource-sharing bidded
CN104303168B (en) Certification for the application of flexible resource demand
Hayyolalam et al. Single‐objective service composition methods in cloud manufacturing systems: Recent techniques, classification, and future trends
Stößer et al. Market-based pricing in grids: On strategic manipulation and computational cost
KR20200023706A (en) Distributed computing resources sharing system and computing apparatus thereof based on block chain system supporting smart contract
Ranjan et al. A case for cooperative and incentive-based federation of distributed clusters
Jäger et al. Crowdworking: working with or against the crowd?
CN112835708A (en) High-quality task allocation and service deployment method for mobile group perception system in edge computing environment
He et al. Programming framework and infrastructure for self-adaptation and optimized evolution method for microservice systems in cloud–edge environments
Mehrotra et al. Towards an autonomic performance management approach for a cloud broker environment using a decomposition–coordination based methodology
Nasrabadi et al. Resource allocation for performance improvement
Mohammadhosseini et al. An efficient energy-aware method for virtual machine placement in cloud data centers using the cultural algorithm
Addya et al. CoMCLOUD: Virtual machine coalition for multi-tier applications over multi-cloud environments
Tong et al. Stackelberg game-based task offloading and pricing with computing capacity constraint in mobile edge computing
Han et al. Tiff: Tokenized incentive for federated learning
CN113157430A (en) Low-cost task allocation and service deployment method for mobile group perception system in edge computing environment
CN111738757A (en) Multi-channel service platform
Chen et al. Incorporating geographical location for team formation in social coding sites
Cirne et al. Scheduling in bag-of-task grids: The PAUÁ case
CN106453557A (en) Two-time scale dynamic bidding and resource management algorithm for user in IaaS service
Ziafat et al. A method for the optimum selection of datacenters in geographically distributed clouds
Zhang et al. Price competition with service level guarantee in web services
Senthil Kumar et al. A novel resource management framework in a cloud computing environment using hybrid cat swarm BAT (HCSBAT) algorithm
Zhang et al. You calculate and I provision: A DRL-assisted service framework to realize distributed and tenant-driven virtual network slicing

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
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20210723